QNX Licensing ... is anyone listening?
Steve Drake
Dont bother them for that, they want mass quantity deals 1000's....
Thats why the price is literaly a moving target. Your guess was off a
full boat developer system will run about $3000.00 single license,
then when you get your stuff running have them quote you on quantity
100 or so, single runtime purchases are roughly $750 and thats just
for the O.S. you want Socket? Thats another $300 or so....
Then there's the ever elusive modular pricing.....
If price is a major concern and your not doing mass quantity then QNX
is NOT for you. You'll see posts here making it sound like QNX can
walk on water but it cant, call them up and ask them for a price on
the IAT toolkit , go ahead, they will give the run around... If they
do give a price please post here.
Better yet swing into one of their new offices and get a demo and talk
to someone there. You cant. Those offices do not exist, every phone
number on the web site is the Canadian office.
On Sun, 08 Nov 1998 00:05:37 +1100, Ron Dunn <r...@relational.net>
wrote:
>I know it is an old topic, but I just can't resist one more try ...
>
>Have you looked at the web-page for the QNX IAT demonstration lately?
>I was there tonight, and it reports over 500,000 downloads of the
>toolkit. Mine's one of them. And if the rest of the downloads are
>like mine, they represent people who are seeking an alternative
>operating system, who are impressed by the demonstration, and who are
>subsequently burned by the bizarre QNX distribution and licensing
>system.
>
>One of the most FAQ on this newsgroup must be "What does QNX cost...",
>and it is one of very few questions which cannot be answered clearly
>and precisely by the otherwise helpful and informative people here.
>
>I have difficulty understanding why QNX is advertised in relatively
>mass-market media, attracting people like me, when the client base
>apparently sought by QNX is so different. It is just too hard to find
>someone from whom one can buy the system. There is no consistent and
>public pricing model on which R&D or simple exploratory purchasing
>decisions can be made. There is no encouragement to "buy and try".
>As a result, my opinion is that QNX is losing thousands of potential
>customers, many potential applications, and maybe the opportunity to
>become a much bigger force in the OS market.
>
>I really wish that QNX would would change its focus and ENCOURAGE
>developers to try the system.
>
>Come on, guys, put together a standard bundling of products similar to
>those used for the IAT. Price it at a single, bundled figure which is
>appealing to small developers, and cost competitive for potential
>users. A guess? Somewhere between USD 300 and USD 500 per copy. In
>fact, why not make it even easier, and put together a time-limited
>demonstration CD bundling all of this stuff and push it out to anyone
>who takes the time to write and ask!
>
>What are you going to lose? By your own figures, there are hundreds
>of thousands of people have downloaded your demonstration, and my
>guess is that only a very small fraction of those have bothered to
>wade through the morass of your current licensing and distribution
>practices. Even if just one percent of those people bought a licence
>at USD 300, that's USD 1,500,000. Isn't that enough to reconsider?
>
>Every application ported to or written for QNX represents extra OS
>sales from the developer of that application. Most people don't go to
>the trouble of porting software for just one use, they expect to sell
>it to other customers, who in turn will represent purchasers of QNX to
>run that application. From my own example, we have ten computers, and
>would probably licence five copies. Our applications sell in the
>banking market, and we have around five installations per month, of
>around 3-4 computers each. Our customers don't care about the OS we
>run, so maybe we would install 100-150 licenses per year. We're
>small, and that's not a big figure. But take 100 companies like ours,
>or even 1,000 companies like ours, and all of a sudden those numbers
>look much bigger.
>
>Furthermore, one of those developers might just be sitting on the
>world's next killer application. No, it's not me (I wish!), but the
>networking / messaging model of QNX opens up so many ideas for new
>software that the possibility can't be ignored.
>
>Then imagine the possibilities which emerge if some of those users
>begin porting office software to QNX. There's a LOT of companies out
>there with hardware which is no longer capable of running Microsoft's
>increasingly ponderous software, who don't want to dabble in the Linux
>market, but who have serious need for a competent, network-capable
>multi-tasking environment to replace their older applications. There
>is also an installed base of happy QNX customers, who would surely be
>happy to see extended use of the OS within their organisations,
>representing a new software market for developers.
>
>Well, maybe QNX wants to remain an embedded OS, or an industrial OS.
>I think it's a waste. The IAT demonstrates the application / user
>potential of this OS. I would dearly love to try working with it, and
>I'm sure I'm not alone.
>
>Please, QNX, take a little risk, and take the chance to realise the
>potential of your OS.
>
>Ron Dunn
>CEO
>Relational Networks
>
Ron Dunn
Armin Steinhoff_de>
<BA8CE78712E72421.0179B0F2...@library-proxy.airnews.net>,
sdr...@kia.net says...
>
>QNX does'nt care about guys like you, one and two O.S. purchases....
>Dont bother them for that, they want mass quantity deals 1000's....
>Thats why the price is literaly a moving target. Your guess was off a
>full boat developer system will run about $3000.00 single license,
OK, go with M$ ... in order to write any lines of code you have to be a member
of the M$ developer network MSDN (SDK, DDK). That means you have to pay year
after
year $2.400 for the MSDN Universal Subscription. After 5 years it takes
$12.000
and there is _no end_ for your payment :-(.
With QNX you have just to pay the initial cost (~$1000) for the compiler :-)
>then when you get your stuff running have them quote you on quantity
>100 or so, single runtime purchases are roughly $750 and thats just
>for the O.S. you want Socket? Thats another $300 or so....
>Then there's the ever elusive modular pricing.....
You want to write a socket based application ? ... subscribe to M$DN !
>If price is a major concern and your not doing mass quantity then QNX
>is NOT for you.
OK, buy a buggy Win98 for $100 and a buggy Office 97 for ~DM 1000 and you will
feel the M$ stuff is nothing for you.
Go with LINUX and Applixware or StarOffice ....
You need a powerful server cluster with more than 2 servers in the 'cluster'?
... go with LINUX and you are at secure site. The market share of UNIX server
(%46) is growing nearly two times faster than the market share (16%) of NT
servers.
>You'll see posts here making it sound like QNX can
>walk on water but it cant, call them up and ask them for a price on
>the IAT toolkit , go ahead, they will give the run around... If they
>do give a price please post here.
Visit the home page of the company 'become' or the homepage of the
QNX FAQ ... there is a price.
>Better yet swing into one of their new offices and get a demo and talk
>to someone there. You cant. Those offices do not exist, every phone
>number on the web site is the Canadian office.
>
Well, they are looking still for sales poeple ... talk with QSSL :-)
Armin Steinhoff
Mario Charest
>Better yet swing into one of their new offices and get a demo and talk
>to someone there. You cant. Those offices do not exist, every phone
>number on the web site is the Canadian office.
>
Indeed, it's my opinion as well, their marketing strategy in the US is/was a
flop.
Valuable time wasted here...
Peter North
developers:
NOT ONE THING ON THIS LIST HAS HAPPENED:
(http://www.qnx.com/company/indirect/user.html)
We are opening QNX regional offices across the United States for one
reason: growth. Growth of QSSL as a company,
growth of our dedication to
customer service, and, of course, the rapid
growth of QNX technology in
the marketplace. But more importantly, this
change means growth for you
as a QNX developer. You now have a dedicated
regional sales team that:
more closely shares your business hours
has more time and resources to address
your specific application
needs
can visit you more often to help with
business or technical issues
can work cooperatively with you on sales
opportunities and local trade
shows
is experienced in providing sales and
application-engineering support
for high-volume products
can put you in touch with other local QNX
developers who may wish to
incorporate your technology into their
products - or whose technology
could benefit your products
can work more closely with other suppliers
in your region to provide
you with smoother, more integrated service
has knowledge and experience in supporting
a range of QNX products
has more opportunity to learn about your
product first-hand and can
consequently keep a lookout for sales
opportunities for you
On Sun, 8 Nov 1998 20:16:58 -0500, "Mario Charest" <m...@videotron.ca>
wrote:
>>
>>Better yet swing into one of their new offices and get a demo and talk
>>to someone there. You cant. Those offices do not exist, every phone
>>number on the web site is the Canadian office.
>>
>
John Birch
wrote:
>>>Better yet swing into one of their new offices and get a demo and talk
>>>to someone there. You cant. Those offices do not exist, every phone
>>>number on the web site is the Canadian office.
>>>
>>
>>Indeed, it's my opinion as well, their marketing strategy in the US is/was a
>>flop.
>>Valuable time wasted here...
>>
>>
>>
>
Might not be happening over there, but it's happening over here!
regards John B.
Igor Kovalenko
There were so many such postings, but I can't remember any signed by a
CEO :)
Well, I used to be in the same boat and you are right, this is deadly
topic. It was discussed over and over again, but very few things
changing. And those changing, not always make people happier.
Now I'm in completely different boat (well, Motorola looks more like a
ship) and can enjoy all those gold service plans, etc. But still no calm
;)
The sad truth is, QSSL business model is simply not suited for the
market you (and me) proposing. Dealing wit lots of small volume customer
implies huge investments into marketing and support, what is not
particularly easy to handle for a privately owned company with <150
employees. And another fact is, QSSL owners don't appear to wish to make
an initial public offering. As I sayd many times, the main problem of
QSSL is "L" letter in its name ;)
Actually, it is easy to understand them. Their current model is nice:
sell 1,000 copies for $3,000, get $3,000,000, keep 10 support guys and
30 developers, concentrating on the OS itself.
Another model is: sell 10,000 copies for $300, get the same $3,000,000,
BUT ... have 100 support guys and 300 developers to develop all the
applications required to stay on much more competitive market.
Well, if they'll go to millions of sales, then the picture may become
different. But then one micro-company from Redmond may politely ask them
to shut down their business and may be QSSL then will switch into an
Open Source model, like Netscape ;))
That sayd, I really would like if they will listen to you ;)
- Igor
Guy Macon
>QNX does'nt care about guys like you, one and two O.S. purchases....
>Dont bother them for that, they want mass quantity deals 1000's....
>Thats why the price is literaly a moving target. Your guess was off a
>full boat developer system will run about $3000.00 single license,
>then when you get your stuff running have them quote you on quantity
>100 or so, single runtime purchases are roughly $750 and thats just
>for the O.S. you want Socket? Thats another $300 or so....
>Then there's the ever elusive modular pricing.....
>
>If price is a major concern and your not doing mass quantity then QNX
>is NOT for you.
There is one other market for QNX, and I am an example of it. I develop
entire production lines for CDs/CD-ROMs/DVDs, and each one costs a bunch
of money (for the real high speed ones, we are talking millions of dollars).
A large part of my project cost is software - and a large part of my
software costs are fighting the bugs and general flakyness of microsoft NT
(a fine OS for the desktop, but poorly suited to realtime control).
I am in the process of switching to QNX, and I would pay $100,000 for
the benefits that I am seeing already. I will probably never buy my
50th QNX system. So, there are your two markets: high development system
cost / high quality / high volume / low unit cost, and high development
system cost / high quality / low volume / high unit cost. And no stories
like the one where Microsoft pulled all the developers off of NT 4.0
Service Pack 4 to work on NT 5.0 and try to keep it on schedule. QNX
takes a much more serious antibug position; the number one goal is
reliability and stability, not adding features or meeting new version
ship dates.
>You'll see posts here making it sound like QNX can
>walk on water but it cant,
Yes it can. ;)
Peter North
>In article <BA8CE78712E72421.0179B0F2...@library-proxy.airnews.net>, sdr...@kia.net wrote:
>
>>QNX does'nt care about guys like you, one and two O.S. purchases....
>>Dont bother them for that, they want mass quantity deals 1000's....
>>Thats why the price is literaly a moving target. Your guess was off a
>>full boat developer system will run about $3000.00 single license,
>>then when you get your stuff running have them quote you on quantity
>>100 or so, single runtime purchases are roughly $750 and thats just
>>for the O.S. you want Socket? Thats another $300 or so....
>>Then there's the ever elusive modular pricing.....
>>
>>If price is a major concern and your not doing mass quantity then QNX
>>is NOT for you.
>
>There is one other market for QNX, and I am an example of it. I develop
>entire production lines for CDs/CD-ROMs/DVDs, and each one costs a bunch
>of money (for the real high speed ones, we are talking millions of dollars).
Too bad QNX cant burn cd's.
>
>A large part of my project cost is software - and a large part of my
>software costs are fighting the bugs and general flakyness of microsoft NT
>(a fine OS for the desktop, but poorly suited to realtime control).
>
>I am in the process of switching to QNX, and I would pay $100,000 for
>the benefits that I am seeing already. I will probably never buy my
>50th QNX system. So, there are your two markets: high development system
>cost / high quality / high volume / low unit cost, and high development
>system cost / high quality / low volume / high unit cost. And no stories
>like the one where Microsoft pulled all the developers off of NT 4.0
>Service Pack 4 to work on NT 5.0 and try to keep it on schedule. QNX
>takes a much more serious antibug position; the number one goal is
>reliability and stability, not adding features or meeting new version
>ship dates.
>
>>You'll see posts here making it sound like QNX can
>>walk on water but it cant,
>
>Yes it can. ;)
Armin Steinhoff
Mario,
someone is abusing your name ...
Armin
In article <Kgr12.300$4b2...@weber.videotron.net>, "Mario says...
>
>>
>>Better yet swing into one of their new offices and get a demo and talk
>>to someone there. You cant. Those offices do not exist, every phone
>>number on the web site is the Canadian office.
>>
>
Lance Roberts
<ABBA3FB34DF8AD88.26EF6580...@library-proxy.airnews.ne
t>...
> On 09 Nov 1998 17:47:38 PST, guym...@deltanet.com (Guy Macon) wrote:
>
> >There is one other market for QNX, and I am an example of it. I develop
> >entire production lines for CDs/CD-ROMs/DVDs, and each one costs a bunch
> >of money (for the real high speed ones, we are talking millions of
dollars).
> Too bad QNX cant burn cd's.
>
smart and friendly writer.
Young Minds have a CD writing product that has a QNX port.
I'd also assume he is actually manufacturing CD's. Qnx is good for
production systems, we have many such systems.
regards
Lance Roberts
Process Computer Engineer
BHP NZ Steel Ltd
Mario Charest
No no it's really me ;-) It's fun to shake the tree once in a while,
always interesting to see was falls off!
You know me Armin ;-) I love QNX and have based my career around it.
But it doesn't mean I have to be blind to what's happening.
IMHO QSSL marketing and sales strategies sucks. But it's their business,
they can run it as they see fit, I respect that. And to some respect I
beleive
it's part of the problem cause by their growth.
Armin Steinhoff wrote in message <7297hb$b...@drn.newsguy.com>...
>
>Mario,
>
>someone is abusing your name ...
>
>Armin
>
>
>In article <Kgr12.300$4b2...@weber.videotron.net>, "Mario says...
>>
>>>
>>>Better yet swing into one of their new offices and get a demo and talk
>>>to someone there. You cant. Those offices do not exist, every phone
>>>number on the web site is the Canadian office.
>>>
>>
Gregor Brandt
You don not have to be an MSDN subscriber to develop for Microsoft.
All compilers for windows come with as much documentation as Watcom does
with QNX. The MS web based 'Knowledge Base' is also quite good. Go buy
a $600 compiler and program your heart out.
Gregor
> ----------
> From: Armin Steinhoff_de>[SMTP:Ar...@newsguy.com]
> Posted At: Sunday, November 08, 1998 9:13 AM
> Posted To: comp.os.qnx
> Conversation: QNX Licensing ... is anyone listening?
> Subject: Re: QNX Licensing ... is anyone listening?
>
> In article
> <BA8CE78712E72421.0179B0F2...@library-proxy.airn
> ews.net>,
> sdr...@kia.net says...
> >
> >QNX does'nt care about guys like you, one and two O.S. purchases....
> >Dont bother them for that, they want mass quantity deals 1000's....
> >Thats why the price is literaly a moving target. Your guess was off a
> >full boat developer system will run about $3000.00 single license,
>
> OK, go with M$ ... in order to write any lines of code you have to be
> a member
> of the M$ developer network MSDN (SDK, DDK). That means you have to
> pay year
> after
> year $2.400 for the MSDN Universal Subscription. After 5 years it
> takes
> $12.000
> and there is _no end_ for your payment :-(.
>
> With QNX you have just to pay the initial cost (~$1000) for the
> compiler :-)
>
> >then when you get your stuff running have them quote you on quantity
> >100 or so, single runtime purchases are roughly $750 and thats just
> >for the O.S. you want Socket? Thats another $300 or so....
> >Then there's the ever elusive modular pricing.....
>
> You want to write a socket based application ? ... subscribe to M$DN !
>
>
> >If price is a major concern and your not doing mass quantity then QNX
> >is NOT for you.
>
> OK, buy a buggy Win98 for $100 and a buggy Office 97 for ~DM 1000 and
> you will
> feel the M$ stuff is nothing for you.
>
> Go with LINUX and Applixware or StarOffice ....
>
> You need a powerful server cluster with more than 2 servers in the
> 'cluster'?
> ... go with LINUX and you are at secure site. The market share of UNIX
> server
> (%46) is growing nearly two times faster than the market share (16%)
> of NT
> servers.
>
> >You'll see posts here making it sound like QNX can
> >walk on water but it cant, call them up and ask them for a price on
> >the IAT toolkit , go ahead, they will give the run around... If they
> >do give a price please post here.
>
> Visit the home page of the company 'become' or the homepage of the
> QNX FAQ ... there is a price.
>
> >Better yet swing into one of their new offices and get a demo and
> talk
> >to someone there. You cant. Those offices do not exist, every phone
> >number on the web site is the Canadian office.
> >
>
Geoff Roberts
>In article <ABBA3FB34DF8AD88.26EF6580...@library-proxy.airnews.net>, pno...@kia.net wrote:
>>
>>On 09 Nov 1998 17:47:38 PST, guym...@deltanet.com (Guy Macon) wrote:
>>>
>>>There is one other market for QNX, and I am an example of it. I develop
>>>entire production lines for CDs/CD-ROMs/DVDs, and each one costs a bunch
>>>of money (for the real high speed ones, we are talking millions of dollars).
>>
>>Too bad QNX cant burn cd's.
>>
>
>If you don't understand the difference between a CD replication line
>and a CD-R burner, you may be more comfortable in another newsgroup.
>comp.os.qnx is for QNX users, and QNX is a realtime operating system,
>not a general purpose desktop operating system. If you do not do any
>realtime work, I think that you will be happier with Linux. If you
>do realtime work, then you understand why so many of us seem so happy
>with QNX.
>
That was awfully polite. I would have simply told him to piss off...
douglas shawhan
> QNX does'nt care about guys like you, one and two O.S. purchases....
> Dont bother them for that, they want mass quantity deals 1000's....
Who doesn't?
>>
>>I really wish that QNX would would change its focus and ENCOURAGE
>>developers to try the system.
>>practices. Even if just one percent of those people bought a licence
>>at USD 300, that's USD 1,500,000. Isn't that enough to reconsider?
*sigh* guess not...
This very afternoon I called a QNX salesperson asking about dealerships (I
own and operate a bookstore that dabbles in software). I was politely told
to go to hell.
Too bad. I think I could sell it all the livelong day for $500 US. I would
love to get the UofE kids of the "Visual Basic" and NT brain drain train.
Come on folks, just a leeetle bit cheaper?
--
___________________________________________________________
while you are passing out the spam, don't forget these guys.
root@localhost
$LOGNAME@localhost
ab...@tiac.net
$LOGIN@localhost
$USER@$HOST
$USER@localhost
____________________________________________________
Inge Vabekk
>This very afternoon I called a QNX salesperson asking about dealerships (I
>own and operate a bookstore that dabbles in software). I was politely told
>to go to hell.
>
>Too bad. I think I could sell it all the livelong day for $500 US. I would
>love to get the UofE kids of the "Visual Basic" and NT brain drain train.
>Come on folks, just a leeetle bit cheaper?
For most people who need a real-time system, QNX is *THE* moneysaver, so
it's cheap enough (if not *TOO* cheap) already.
Why should QSSL be interested in selling *CHEAP* to people with no interest
for data processing (or knowledge about it)?
It doesn't make sense at all. Less money -- more trouble, more support....
for purposes remotely different from QSSLs objectives?
Come on. You don't really mean this.
Start selling cars in your store instead. Very little knowledge is required
to drive one, and the nearest garage takes the support responsibility.
--
in...@arxi.nospam (remove spam....)
Armin Steinhoff
>
[ clip ... ]
>This very afternoon I called a QNX salesperson asking about dealerships (I
>own and operate a bookstore that dabbles in software). I was politely told
>to go to hell.
>
>Too bad. I think I could sell it all the livelong day for $500 US. I would
>love to get the UofE kids of the "Visual Basic" and NT brain drain train.
>Come on folks, just a leeetle bit cheaper?
Oh yes, it's possible. The initial costs could be lower if QSSL _sells_ all
major updates or releases for a _single user_ entry level QNX which boots
directly into PHONTON! Wouldn't it be nice when all - I guess more than a half
million - users would pay for the updates ??
And of cause ... each major release must be a marketing/press happening !!
And could it not possible to offer different license schemes for QNX??
One with the high initial price and 'zero update costs' ... and a second one
with a low initial price and 'none zero update costs' for a special single user
QNX/PHOTON with some restrictions e.g. no FLEET ??
What schould a low cost entry level QNX/PHOTON include ??
Any ideas ??
Armin
Igor Kovalenko
>
> That was awfully polite. I would have simply told him to piss off...
Hey Geoff, don't shot him for freedom of speech ;)
All this vendetta somehow reminds me times when some guys here aimed the
gun to me ;))
I don't mean the guy is doing very right things, but sometimes some
things may drive any of us crazy enough to write something like that...
you must know this I believe ;)
Anyways, whichever evil things he will do, this so much attention to him
actually just provides extra exposure to his messages, what is not
exactly what you probably want ;~\
The best way to shut down a really evil guy is just ignore him totally.
Or you guys can fill and send him that "standard generalized bonehead
reply form" :)
Cheers,
- Igor
Felix S. Gallo
Inge Vabekk writes:
>Why should QSSL be interested in selling *CHEAP* to people with no interest
>for data processing (or knowledge about it)?
You're misclassifying the people who want to meddle with an inexpensive
RTOS. They want to do data processing. And what's more, the answer
to your question is: they are the people who help shape product decisions
at development corporations. For instance, me.
Felix
Felix S. Gallo
>This very afternoon I called a QNX salesperson asking about dealerships (I
>own and operate a bookstore that dabbles in software). I was politely told
>to go to hell.
>
>Too bad. I think I could sell it all the livelong day for $500 US. I would
>love to get the UofE kids of the "Visual Basic" and NT brain drain train.
>Come on folks, just a leeetle bit cheaper?
Douglas has hit upon the key here, in my opinion.
UNIX, Windows 95, Windows CE, Windows NT, Linux, and FreeBSD --
all competitors in one way or another in the x86 space -- make their
name for themselves, and a serious amount of money, by seeding
the market with extremely inexpensive starter kits and providing
much-more-expensive tools to those who want to go to the next level.
Vide free 2-user Solaris, free 2-user SCO, academic pricing, Win95
preinstalled on every computer, Windows CE development a free
upgrade to existing Microsoft development tools, Windows NT coming
free with every server, Linux being free, and FreeBSD being free --
and then look at the price of 'real' Solaris, 'real' SCO, corporate
desktop licensing, dev tools for Solaris and SCO, Microsoft Office,
Windows CE hardware, Windows NT Server, Caldera Linux, and
BSDI's BSD boxes. It's a very successful tactic that drives
mindshare.
Now, other embedded companies like Wind River do it like QNX
currently does -- no free lunch, huge cost for the first seat, very
rapidly diminishing costs for the seats thereafter. But those
embedded companies' products have nowhere near the promise
and technological coolness of QNX (and I can say this as someone
who works in Tornado half the day), and so we naturally want QNX
to step up to the next level (a la Linux, Windows, Solaris) and shoot
for the big serious targets.
While working at four different companies, I have every time come
to the conclusion that QNX was absolutely the best OS to use for
the product I was designing. I've even gone so far as to buy
the expensive development licenses (hi, Pierre! :)) and build
fully functional replica demo models of our products (some of
which were projected to sell in the tens-to-hundreds-of-thousands
range). I've been so hungry for a decent embedded operating
system, and so impressed by the core technology of QNX, that
I've evangelized QNX to the moon.
Company #1: "Since we can't get a good idea of how much it will
cost us, and SCO is (small integer) per copy, we'll use SCO."
Company #2: "What amazing technology. But I've never heard
of it. Let's stick with something widely known and available that
we can get developers for, like IRIX."
Company #3: "Sounds great, but we have got to go with Microsoft.
Don't ask any questions. We just do." (okay, there's nothing
QNX could have done about this one.)
Company #4: "QNX? Oh yeah, we used that in a previous project.
Great, great OS. Priced, of course, way out of line for even high
volume consumer devices like we make. Kinda sad, really."
(another message out of company #4: "QNX would be even
better if it ran on a serious modern chip like the StrongARM."
Just mentioning that as an aside. :))
Like Douglas, I'm confident I could get a yacht in the Mediterranean
by being provided QNX at slightly over cost to sell to the Linux
Masses. The direct market is huge. But what's more, the
_indirect_ market -- what these kids do when they get into real
companies -- is what QNX is at least in theory aiming for.
Felix
Mitchell Schoenbrun
> This very afternoon I called a QNX salesperson asking about dealerships (I
> own and operate a bookstore that dabbles in software). I was politely told
> to go to hell.
Well now, I presume the "hell" part is your interpretation. As you
might know already, QSSL changed their sales model to where they
sell QNX directly only. I don't know of any customers who've
been happy about this, but that's their decision. In any case,
they've never sold QNX in a book store.
Guy Macon
>
>douglas shawhan wrote in message.....
>
>>own and operate a bookstore that dabbles in software). I was politely told
>>to go to hell.
>>
>>love to get the UofE kids of the "Visual Basic" and NT brain drain train.
>>Come on folks, just a leeetle bit cheaper?
>
>
>it's cheap enough (if not *TOO* cheap) already.
>
>for data processing (or knowledge about it)?
>for purposes remotely different from QSSLs objectives?
For just about anyone who has problems with the price, Either an ordinary
RTOS or Linux would be a better choice. Think of QNX as a fighter jet;
if cost is a concern, buy a general aviation plane. If you are going to
be shot at by MIGs, you don't care how much your plane costs...
Guy Macon
>
Sorry. I forgot that I wasn't in soc.religion.quaker anymore. <smile>
Ken Schumm
> >You know me Armin ;-) I love QNX and have based my career around it.
> >But it doesn't mean I have to be blind to what's happening.
> >IMHO QSSL marketing and sales strategies sucks. But it's their business,
> >they can run it as they see fit, I respect that. And to some respect I
> >beleive
> >it's part of the problem cause by their growth.
>
> every one of them that I ever met at the conferences. But let's face it, they
> aren't marketing geniuses and will never have the resources to really make QNX
> a household word in the computer industry as long as they refuse to go public,
> be bought out, or merge with a sugar daddy. In 5 years QNX will go the way of
> Beta video (I bought one of the last Beta Hi-Fi's because of it's obvious
> superiority). We will all be using NT, CE, Linux, or out of work like
> carbeurator engineers, because we refused to adapt.
Lets see, I've been using QNX since ~1984, and since about 1988 I've been
hearing that QNX will become the next dodo bird. Hasn't happened yet.
IMHO, there are too many savvy engineers that are unsatisfied with the
technical inadequacy of MS products for technically superior products to
disappear. Not to say it can't happen to QNX or any other company, but
there will always be a market niche for superior, or percieved superior,
products.
--
kwsc...@qsolv.com QSolv, Inc.
Purveyor of custom QNX software solutions
Geoff Roberts
<kova...@comm.mot.com> wrote:
>Geoff Roberts wrote:
>>
>> That was awfully polite. I would have simply told him to piss off...
>
>Hey Geoff, don't shot him for freedom of speech ;)
>All this vendetta somehow reminds me times when some guys here aimed the
>gun to me ;))
>
>I don't mean the guy is doing very right things, but sometimes some
>things may drive any of us crazy enough to write something like that...
>you must know this I believe ;)
>
>Anyways, whichever evil things he will do, this so much attention to him
>actually just provides extra exposure to his messages, what is not
>exactly what you probably want ;~\
>
>The best way to shut down a really evil guy is just ignore him totally.
>Or you guys can fill and send him that "standard generalized bonehead
>reply form" :)
>
>Cheers,
>- Igor
You're probably right Igor. It's just that I'm getting older and
sometimes my liver goes sour on me...
Cheers,
Geoff.
PS: you homesick yet??? ;-)
CR91LJL
Inge Vabekk
>
>Inge Vabekk writes:
>>Why should QSSL be interested in selling *CHEAP* to people with no
interest
>>for data processing (or knowledge about it)?
>
>RTOS. They want to do data processing. And what's more, the answer>to
your question is: they are the people who help shape product decisions>at
development corporations. For instance, me.
And me. I didn't pay *cheap*. But in the first project it saved me a similar
amount of money. And the next. And the next. So it turned out to be quite
"cheap" after all.
To me, cheap is near-synonymous to low quality. And to QSSL, cheap would
probably mean that they would have to give up their design and marketing
objectives, meaning that guys like us would have to find yet another OS,
because QNX wouldn't be the right product for the job any more.
Besides, the fact is that most people who ask for QNX are NOT interested in
its RT capabilities. They don't even know what on earth they should need RT
for (or what it is). They'd probably be more happy with Linux, because what
they need most is an OS for general (office?)data processing.
--
in...@arxi.nospam (remove spam....)
Inge Vabekk
>
>Hey, I love QNX, admire the technical geniuses in Kanata and I actually
liked>every one of them that I ever met at the conferences. But let's face
it, they>aren't marketing geniuses and will never have the resources to
really make QNX>a household word in the computer industry as long as they
refuse to go public,>be bought out, or merge with a sugar
Right. Have you ever considered the fact that QSSL don't WANT to be
marketing geniuses?
>daddy. In 5 years QNX will go the way of>Beta video (I bought one of >the
last Beta Hi-Fi's because of it's obvious>superiority). We will all be
>using NT, CE, Linux, or out of work like>carbeurator engineers, >because we
refused to adapt.
I don't think so, but of course, only time will show.
I'd probably be among the first ones to jump 20 feet high if I could run all
my Microshit programs (or look-alikes) on QNX, but I have to accept that
that's not the reason for its existence. When it comes to programming, I
wouldn't spend an hour writing a program for M$ Windows, or a dime for a new
compiler (I have a 10 years old one that's OK for very simple,
quick-and-dirty DOS utility programs).
We have a demo CDs from Metrowerks, containing Windows code, but no QNX
code. I wouldn't even bother to install it on an M$ system. I have more
useful things to do.
But that's MY point of view. I'll let others have their own. And tomorrow
we'll have Metrowerks visiting. Just to give us the first look.....(+:
--
in...@arxi.nospam (remove spam....)
Glenn Sherman
Inge Vabekk <in...@arxi.no> wrote in article <72e2os$5br$1...@elle.eunet.no>...
Thought I would add my .02
I think we can all agree that when it comes to RTOS - Qnx is at the
top of its class, and it is probably priced right where it should be for
that.
Why then does Qnx keep advertising its 1.44 meg floppy with everything
you need for the internet...blah, blah, blah, if they don't want people
using it ( or trying to ) for general purpose stuff. If I am dealing with
RT,
why do I care that I can get on the internet with less than 1.44 meg of
space. More importantly, why does everyone get so surprised that
after people get a FREE download of this wonderful thing, that they
figure that they might be able to get the OS for less than $1,000.
If Qnx doesn't want to bother with the small user ( and I don't blame them,
10,000 dummy users = lots of $ in support ), STOP advertising the
1.44 floppy to everyone.
> in...@arxi.nospam (remove spam....)
>
--
Glenn Sherman gsherman@remove_this.jlc.net
Granite State Software
Steve Munnings
Glenn Sherman wrote in message
<01be0e4e$e36c0de0$439f...@Pgsherm.jlc.net>...
>
>Thought I would add my .02
>
>I think we can all agree that when it comes to RTOS - Qnx is at the
>top of its class, and it is probably priced right where it should be for
>that.
>
>Why then does Qnx keep advertising its 1.44 meg floppy with everything
>you need for the internet...blah, blah, blah, if they don't want people
>using it ( or trying to ) for general purpose stuff. If I am dealing with
>RT,
>why do I care that I can get on the internet with less than 1.44 meg of
>space. More importantly, why does everyone get so surprised that
>after people get a FREE download of this wonderful thing, that they
>figure that they might be able to get the OS for less than $1,000.
>
>If Qnx doesn't want to bother with the small user ( and I don't blame them,
>10,000 dummy users = lots of $ in support ), STOP advertising the
>1.44 floppy to everyone.
I think that the market QSSL is after here is the people building set-top
internet appliances, and other embedded systems that use the Internet. I
believe that they have already convinced a significant number of these
poeple of the merits of QNX - see their web-site...
If "everybody" learns about the technical marvels of the QNX, and its
incredibly small footprint, then the "target" population has a better chance
of hearing about it, and also convincing their management, etc., of its
merits.
While QNX is a leading RTOS, they are also targetting the embedded market
where RT may not be so important, but small footprint definitely is. And,
they sure do have small footprint for what they are doing. Also, finally,
if I owned such a "slick" piece of technology, I sure would want people to
know about it and appreciate it, even if they did not have a need to buy it!
:)
My $0.02 (Canadian, no less!)
Igor Kovalenko
I have some information that QSSL actually made some attempts to put
their shoe into general purpose OS area. I mean, they had some
negotiations with some software vendors of GP-class, like SQL servers
and even complete office packages.
Don't ask me details, but I trust the source 100%.
Are you guys surprized? ;)
- Igor
P.S.
I'm speaking only for myself, not for QSSL or my employeer. Nothing of
above has been declared by QSSL officially, and it all might be just my
imagination as well.
Glenn Sherman wrote:
>
> Thought I would add my .02
>
> I think we can all agree that when it comes to RTOS - Qnx is at the
> top of its class, and it is probably priced right where it should be for
> that.
>
> Why then does Qnx keep advertising its 1.44 meg floppy with everything
> you need for the internet...blah, blah, blah, if they don't want people
> using it ( or trying to ) for general purpose stuff. If I am dealing with
> RT,
> why do I care that I can get on the internet with less than 1.44 meg of
> space. More importantly, why does everyone get so surprised that
> after people get a FREE download of this wonderful thing, that they
> figure that they might be able to get the OS for less than $1,000.
>
> If Qnx doesn't want to bother with the small user ( and I don't blame them,
> 10,000 dummy users = lots of $ in support ), STOP advertising the
> 1.44 floppy to everyone.
>
Armin Steinhoff
says...
>
>>You know me Armin ;-) I love QNX and have based my career around it.
>>But it doesn't mean I have to be blind to what's happening.
>>IMHO QSSL marketing and sales strategies sucks. But it's their business,
>>they can run it as they see fit, I respect that. And to some respect I
>>beleive
>>it's part of the problem cause by their growth.
>
>every one of them that I ever met at the conferences. But let's face it, they
>aren't marketing geniuses and will never have the resources to really make QNX
>a household word in the computer industry as long as they refuse to go public,
>Beta video (I bought one of the last Beta Hi-Fi's because of it's obvious
>superiority).
QNX isn't based on dieing 'industrial standard' ...
> We will all be using NT, CE, Linux, or out of work like
>carbeurator engineers, because we refused to adapt.
... fatalism based on M$ marketing.
The opposite will happen. Any idea why Matrox is using QNX?? Any idea why the
market share of UNIX based server is growing faster than the market share of NT
server?? Oh yes, I heard the marketing hypes ... NT will replace UNIX :-)
But the reality is different .....
Armin Steinhoff
http://www.DACHS.net
>
Armin Steinhoff
>
>Hmm, I'm afraid to say this, but too tempted ;)
>I have some information that QSSL actually made some attempts to put
>their shoe into general purpose OS area. I mean, they had some
>negotiations with some software vendors of GP-class, like SQL servers
>and even complete office packages.
>
>Don't ask me details, but I trust the source 100%.
>Are you guys surprized? ;)
Hi Igor,
I'm not really surprized, because it should be possible to port Applixware or
StarOffice to X or Xfree86 without big efforts.
But ... is behind this 'single step' something like a strategie ?
Or is it just a 1.44M flop-py w/o marketing background ??
Greetings
Armin
Guy Macon
>Besides, the fact is that most people who ask for QNX are NOT interested in
>its RT capabilities. They don't even know what on earth they should need RT
>for (or what it is). They'd probably be more happy with Linux, because what
>they need most is an OS for general (office?)data processing.
The way I see it, there are only three reasons to prefer QNX over Linux.
[1] You need realtime capabilities. (If you don't know what this means,
then you don't need realtime capabilities!)
[2] You need very high reliability, as in "if this crashes, someone dies"
or "if this crashes, this company goes bankrupt".
[3] You want to impress the many supermodels who are secret QNX groupies
and who get really hot whenever they see your small tight Kernel.
Mitchell Schoenbrun
While these are clearly important features of QNX over Linux,
especially #3, this does not come close to describing all the
"whys" that people might have for using QNX. Here are just a few
more from my own experience.
A) QNX comes from a company. I know I could start a flame war
over support for Linux, and I'm not even saying this is true
for me, but some customers need to know exactly where a product
comes from. This give them the opportunity to count on, or
lean on them when the going gets tough.
B) QNX's tightly coupled networking.
C) Ease of creating and integrating drivers.
D) Small footprint, and high efficiency.
A final note. Just because you don't know what
real-time is, you don't necessarily don't need it.
On more than one occaision I've seen pea-brain
management try to oust QNX as the OS, and fail
because the replacement did not have the
needed real-time capabilities.
--
Mitchell Schoenbrun --------- masc...@pobox.com
Inge Vabekk
>you need for the internet...blah, blah, blah, if they don't want people
>using it ( or trying to ) for general purpose stuff. If I am dealing with
>RT,
>why do I care that I can get on the internet with less than 1.44 meg of
>space. More importantly, why does everyone get so surprised that
>after people get a FREE download of this wonderful thing, that they
>figure that they might be able to get the OS for less than $1,000.
>
>If Qnx doesn't want to bother with the small user ( and I don't blame them,
>10,000 dummy users = lots of $ in support ), STOP advertising the
>1.44 floppy to everyone.
You've got a point there. Well worth a thought.
--
in...@arxi.nospam (remove spam....)
Tom Sheppard
product that runs QNX, then you must purchase the appropriate license."
Until then, the OS is sold at cost to encourage developers and
experimenters.
People who wouldn't respect this licensing arrangement probably will steal
the s/w anyway. It will be more difficult with limited distribution, but
not impossible.
Documentation could be provided as PDF files so no manuals need be
shipped. The product itself could be downloaded from the net keeping
distribution costs low.
There would be no support available through QNX without additional cost.
"May I have your license number sir? I'm sorry, that is an experimenter's
license number and does not entitle you to support. May I transfer you to
our sales department or would you prefer to dial a 900 number for per-call
technical support?" You could automate this by having people enter their
license number into an automated voice response system.
I'm not a QNX developer, but I've heard good things about it. I think it's
very important to ensure that high quality s/w survives in this market
because the alternative is "desktop" quality RTOSes -- and that's real
scary.
...Tom
Guy Macon
>
>Glenn Sherman wrote in message <01be0e4e$e36c0de0
>>
>>you need for the internet...blah, blah, blah, if they don't want people
>>using it ( or trying to ) for general purpose stuff. If I am dealing with
>>RT,
>>why do I care that I can get on the internet with less than 1.44 meg of
>>space. More importantly, why does everyone get so surprised that
>>after people get a FREE download of this wonderful thing, that they
>>figure that they might be able to get the OS for less than $1,000.
>>
>>If Qnx doesn't want to bother with the small user ( and I don't blame them,
>>10,000 dummy users = lots of $ in support ), STOP advertising the
>>1.44 floppy to everyone.
>
>
>You've got a point there. Well worth a thought.
On the other hand, it may (or may not!) be a wise marketting decision
to target a larger audience to reach more of your target audience.
Consider the Sun webserver ads on netwok television; 99.9% of the
viewers don't buy webservers, but those who do are worth the expense.
If this is the thinking at QSSL, they should try to craft the ads and
the demo disks to make it clear up front that QNX is not low cost.
I am reminded of all of the bank ads claiming that they love to say yes
to those seeking a loan. Most of those who respond get a big no, but a
few of the kind of customer they want get swept in with the crowd...
Steve McPolin
Guy Macon <guym...@deltanet.com> wrote:
>In article <72gmch$oa$1...@elle.eunet.no>, in...@arxi.no wrote:
>>
>>Glenn Sherman wrote in message <01be0e4e$e36c0de0
>>>
>>>Why then does Qnx keep advertising its 1.44 meg floppy with everything
>>>you need for the internet...blah, blah, blah, if they don't want people
>>>using it ( or trying to ) for general purpose stuff. If I am dealing with
>>>RT,
>>>why do I care that I can get on the internet with less than 1.44 meg of
>>>space. More importantly, why does everyone get so surprised that
>>>after people get a FREE download of this wonderful thing, that they
>>>figure that they might be able to get the OS for less than $1,000.
>>>
>>>If Qnx doesn't want to bother with the small user ( and I don't blame them,
>>>10,000 dummy users = lots of $ in support ), STOP advertising the
>>>1.44 floppy to everyone.
Every morning I read a newspaper called the "Globe and Mail".
It always contains 1/2 page or better ad from Mercedes or BMW.
I can't afford either. I don't write letters to the editor
complaining about how I can't afford what they are advertising.
I look, think "what a pretty ad". ( in my more cynical moments
think "thank god someone is stupid enough to buy that crap; it
keeps the price of Acura's lower"). Neither Mercedes Benz nor
BMW has ever offered me a free test drive. Maybe if they did,
I would throw fiscal responsibility to the wind and sign a piece
of paper. Maybe I would decide some things are worth it.
#include <std-disclaimer.h>
--
Steve McPolin (st...@qnx.com); QNX Software Systems, Ltd.
Igor Kovalenko
>
> Every morning I read a newspaper called the "Globe and Mail".
> It always contains 1/2 page or better ad from Mercedes or BMW.
> I can't afford either. I don't write letters to the editor
> complaining about how I can't afford what they are advertising.
> I look, think "what a pretty ad". ( in my more cynical moments
> think "thank god someone is stupid enough to buy that crap; it
> keeps the price of Acura's lower").
Hey, they aren't that 'crap'. I liked to drive BMW, despite it was an
older one ;)
> Neither Mercedes Benz nor
> BMW has ever offered me a free test drive. Maybe if they did,
> I would throw fiscal responsibility to the wind and sign a piece
> of paper. Maybe I would decide some things are worth it.
>
You use not a very good analogy. Mercedes Benz can't offer you a lower
price, because there are high manufacturing costs, aside from high
design costs. They can't put it into 1.44 drive and get a copy for
cheap. But they can offer you a lease, did you consider it for your
analogy?
So, the real question is the way QSSL may earn money from QNX. So far
they prefer to enforce developers to pay a lot from the beginning. All
this flame comes from the fact, that most people don't believe this is
the only way, or even the best way.
Another way would be to encourage developers to try QNX (free test
drive), then earn money from runtime licenses being sold by lots of new
developers (like car dealers earn money from lease).
Actually (AFAIK) QSSL doesn't really make money from development
licenses. Moreover, I heard that QSSL may be loosing money if we'll look
at development systems business alone. So, what prevents QSSL from
trying another approach?
Of course, nobody should consider my speculations about QSSL business
'authoritative'. I speak only for myself.
- Igor
Ron Dunn
[.. in respose to suggestions that QNX should not mass-advertise ..]
> Every morning I read a newspaper called the "Globe and Mail".
> It always contains 1/2 page or better ad from Mercedes or BMW.
> I can't afford either. I don't write letters to the editor
> complaining about how I can't afford what they are advertising.
> I look, think "what a pretty ad". ( in my more cynical moments
> think "thank god someone is stupid enough to buy that crap; it
> keeps the price of Acura's lower"). Neither Mercedes Benz nor
> BMW has ever offered me a free test drive. Maybe if they did,
> I would throw fiscal responsibility to the wind and sign a piece
> of paper. Maybe I would decide some things are worth it.
When a newsgroup discussion slides into analogy, it's just one short step away
from the "You're a NAZI!" stage. The problem with analogies is that, for every
writer's similarity, there is a corresponding reader's difference.
I started this thread out of frustration with the perceived mismatch between QNX
advertising and licensing. I saw the advertising, tried the product, and LOVED
it. Then I made the mistake of trying to BUY it. Deja News will show many
similar stories (including mine) of the problems trying to buy QNX as a developer
or end user.
We got some sensible feedback in this thread. The argument that QNX is a small
company, having fun doing what they like in a market they enjoy, is one I
respect. My own company is like that -- we build banking software, we reject
requests to move into other markets, and we have fun at our job. I also
appreciated the argument that increasing the distribution of QNX will probably
mean increased end-user support costs.
Suggestions to use Linux aren't appropriate for me. I've tried it, but am
disappointed with the bloat which it is experiencing at present. Further, there's
too much confusion with desktops, turf-wars with competing projects, and an
anti-commercial bias against major organisations porting to and supporting the
platform. I expect people to pay for our software, and therefore I don't mind
paying for the software we use.
Whilst I think that the pricing of QNX is too high, and that they would make more
money by reducing the cost and selling more, my biggest problem is that it is too
complex to licence! It seems that if you're not in the territory of one of the
"good" distributors, you might as well forget the product.
The response from one participant, that QNX offer two levels of distribution (with
support / none) is a possible solution. An alternative solution, which we
employed with our own software, is to "retail" the product.
We used to have a complex pricing model like QNX, where you paid a base licence,
then paid per-function (or per transaction) increments. It was complex to sell,
and complex to support the sales and implementation process because there wasn't
enough standardisation.
Recently we re-wrote our sales model, and made three bundlings - Lite, Complete,
Source, and substantially reduced the cost of all three bundlings. We now have so
much demand from our sales partners that we've doubled our staff in three months
just to keep pace with installation support. The interesting feature, though, is
that the effort in pre- and post sales support has actually FALLEN, and our
profits have INCREASED. Because there are fewer combinations of things to sell,
because we clearly document the contents and dependencies of each bundling, there
are fewer proposal-support enquiries. We're selling more. Further, we took the
approach of providing a copy of the product to every sales person in our
distribution channel to ensure that they were able to learn, demonstrate and
properly promote our system.
QNX could take the same approach. Put together a workable bundling that will let
people implement the equivalent of the IAT demonstration. Call it QNX-Lite. Give
it basic QNX, Photon, and maybe the Voyager stuff. Offer a "complete" bundling
with all file systems, FLEET, etc, for those users wanting advanced features. For
BOTH offerings, have a "developer" option that provides compilers, database,
tools, etc.
With appropriate pricing, and mass market distribution (mail order? programmer
shop?), this would certainly satisfy MY desire to try QNX. I can imagine that a
"lite" bundling would be the solution to the Book Seller's problem raised
previously in this thread. The engineer / RT types could continue to buy/use the
"complete" offering, maybe benefitting from more simple packaging and
distribution.
Finally, I want to make the point that I'm writing this because I *LIKE* QNX. I'm
not trying to denigrate the product or its users. I'm not flaming the company. I
simply (?) want an available solution to the desktop and delivery OS problems we
face in our company, and I'm frustrated that it is *JUST* out of reach.
Ron Dunn
CEO
Relational Networks
douglas shawhan
> douglas shawhan writes:
>>This very afternoon I called a QNX salesperson asking about dealerships (I
>>own and operate a bookstore that dabbles in software). I was politely told
>>to go to hell.
Okay, to "go to heck". :)
>>
>>Too bad. I think I could sell it all the livelong day for $500 US. I would
>>love to get the UofE kids of the "Visual Basic" and NT brain drain train.
>>Come on folks, just a leeetle bit cheaper?
> Douglas has hit upon the key here, in my opinion.
> UNIX, Windows 95, Windows CE, Windows NT, Linux, and FreeBSD --
> all competitors in one way or another in the x86 space -- make their
> name for themselves, and a serious amount of money, by seeding
> the market with extremely inexpensive starter kits and providing
> much-more-expensive tools to those who want to go to the next level.
> Vide free 2-user Solaris, free 2-user SCO, academic pricing, Win95
> preinstalled on every computer, Windows CE development a free
> upgrade to existing Microsoft development tools, Windows NT coming
> free with every server, Linux being free, and FreeBSD being free --
> and then look at the price of 'real' Solaris, 'real' SCO, corporate
> desktop licensing, dev tools for Solaris and SCO, Microsoft Office,
> Windows CE hardware, Windows NT Server, Caldera Linux, and
> BSDI's BSD boxes. It's a very successful tactic that drives
> mindshare.
What he said. I wonder how many more copies of Mathematica Wolfram has sold
(I know, apples and plungers...) since their aggressive pricing for
students? I set up about three to five linux boxes for college and high
school kids per week. The interest in stable OS's among the youngns'has
grown phenominally in my town- in no small part due to my constant ranting
on the asskickedness of linux and it's ilk. True, most college CS and
Engineering students need a RTOS like an extra limb (with the exception of
the robotics kids...), but considering what most of them spend on constant
upgrades to keep up with NT's hardware jones I think an argument can be made
that one would save money in the long run.
> system, and so impressed by the core technology of QNX, that
> I've evangelized QNX to the moon.
I have dd'd about 30 floppies with the internet demo. (an excellent use for
my old win 3.11 diskies) Whenever CS kids come to my store to sell or buy
textbooks I hand them one. I feel like a moonie....
> Like Douglas, I'm confident I could get a yacht in the Mediterranean
> by being provided QNX at slightly over cost to sell to the Linux
> Masses. The direct market is huge. But what's more, the
> _indirect_ market -- what these kids do when they get into real
> companies -- is what QNX is at least in theory aiming for.
Worth a shot.
(maybe include gcc and port some libs while you are at it- while we are
piping, that is!)
d
Mike Davies
<kova...@comm.mot.com> writes
>So, the real question is the way QSSL may earn money from QNX. So far
>they prefer to enforce developers to pay a lot from the beginning.
Have you ever had the misfortune to speak to Wind River about
development seat prices, Igor ? :-)
>- Igor
--
Mike Davies
Armin Steinhoff
>
>In article <72huof$k...@chronicle.concentric.net>,
>Guy Macon <guym...@deltanet.com> wrote:
>>In article <72gmch$oa$1...@elle.eunet.no>, in...@arxi.no wrote:
>>>
>>>Glenn Sherman wrote in message <01be0e4e$e36c0de0
>>>>
>>>>Why then does Qnx keep advertising its 1.44 meg floppy with everything
>>>>you need for the internet...blah, blah, blah, if they don't want people
>>>>using it ( or trying to ) for general purpose stuff. If I am dealing with
>>>>RT,
>>>>why do I care that I can get on the internet with less than 1.44 meg of
>>>>space. More importantly, why does everyone get so surprised that
>>>>after people get a FREE download of this wonderful thing, that they
>>>>figure that they might be able to get the OS for less than $1,000.
>>>>
>>>>If Qnx doesn't want to bother with the small user ( and I don't blame them,
>>>>10,000 dummy users = lots of $ in support ), STOP advertising the
>>>>1.44 floppy to everyone.
>
>It always contains 1/2 page or better ad from Mercedes or BMW.
>I can't afford either.
It would be nice if QSSL would learn from the marketing strategy of
Mercedes ... there is also a 'Mercedes light' on our street :-)
Armin Steinhoff
http://www.DACHS.net
Paul May
sorry for butting in on my first subscribe but i think
its relavent to this thread.
today (sat 14th Nov 98) its been confirmed that AMIGA inc/GateWay
will use the QNX kernal for the new AMIGA OS5/Dev and /Prod Machines.
http://ugn.amiga.org/events/logs/havemose.shtml (and more logs there)
"
QNX will provide Kernel, file system, device drivers ect,
AMIGA will provide UI, GFX, 3D, Multimedia, video ect"
the plan is basicly to give the new (3rd party built) machines the
same leading edge Tec as the original AMIGA 1000's had in the 80's.
its good for QNX as EVERYONES heard of the AMIGA and now you can bet
the masses in the Micro$loth markets will take notice (i'd imagine).
comments and your long standing thoughts on the QNX OS strenths/weaknesses
in a mass market for the many people that are as yet a bit vaige about its
past/future possabilitys Please.
PS.
any more NG's relating to QNX other than this one (on my ISP) ?.
>Actually (AFAIK) QSSL doesn't really make money from development
>licenses. Moreover, I heard that QSSL may be loosing money if we'll look
>at development systems business alone. So, what prevents QSSL from
>trying another approach?
>Of course, nobody should consider my speculations about QSSL business
>'authoritative'. I speak only for myself.
--
Paul May ,UK
Team
*AMIGA* WorlWide Member
<tsb> User Of The small OS With A BIG Future
--
<tsb> Looking For Disabled persons Ideas
<tsb> For Better Computer Interaction & Inclusion
--
John Grant
Armin Steinhoff <Armin@Steinhoff_de> wrote:
>In article <364B07C0...@comm.mot.com>, Igor says...
>>
>>Hmm, I'm afraid to say this, but too tempted ;)
>>I have some information that QSSL actually made some attempts to put
>>their shoe into general purpose OS area. I mean, they had some
>>negotiations with some software vendors of GP-class, like SQL servers
>>and even complete office packages.
>>
>>Don't ask me details, but I trust the source 100%.
>>Are you guys surprized? ;)
>
>Hi Igor,
>
>I'm not really surprized, because it should be possible to port Applixware or
>StarOffice to X or Xfree86 without big efforts.
>
>But ... is behind this 'single step' something like a strategie ?
>Or is it just a 1.44M flop-py w/o marketing background ??
You might want to ask Amiga Inc. (Gateway). Word coming out of Computer98
in Germany this weekend is that QNX is partnering with them on the
next-gen Amiga OS.
As I type this there's nothing on either the QNX or AI web sites to
support that yet, but that's the buzz on a couple mailing lists.
[snip]
Jim Lambert
>Hi everyone.
>
[snipped]
>
>PS.
>
>any more NG's relating to QNX other than this one (on my ISP) ?.
There is a QNX news server with all the QUICS newsgroups. It is
news.qnx.com.
Jim
Inge Vabekk
This was really good news, because the QUICS system is not very user
friendly. I've had the opportunity to use it for several years, but haven't
done so until recently because the only way I could reach it was via modem.
Now we can use the Internet, but the "regular" access method depends on
cooperation between several people.
The 'tin' system doesn't handle 'quics' files with an owner ID not identical
to the current user's, so it sometimes sends messages to the bit bucket
(with no error message).
I tried to ping it (the news server), but otherwise there was no response.
Is it closed on Sundays? (+:
Or maybe we need to get an access before we can use it?
--
in...@arxi.nospam (remove spam....)
Jim Lambert
Inge Vabekk wrote in message <72mcjf$n61$1...@elle.eunet.no>...
You don't need any special access to access through your news reader. I
would try it again as it seems this news server goes down or is unavailable
a few times every week. It could also have been backup time or something.
Give it a try again.
As to using tin and QUICS you will still have to use them to post messages.
The internet newsgroup access is read-only.
Jim
Inge Vabekk
unavailable
>a few times every week. It could also have been backup time or something.
>Give it a try again.
>
>As to using tin and QUICS you will still have to use them to post messages.
>The internet newsgroup access is read-only.
Oh. Too bad.
--
in...@arxi.nospam (remove spam....)
qnx...@ibm.net
from canada to california, i assure you that the offices are real, and
perhaps not surprisingly busy. qnx sells on an oem basis and is priced
accordingly, much like our competitors. we are also pleased to provide, more
or less at cost to our third party partners we have also initated an
educational program. you may also have noted that qnx has been announced as
the foundation os technology for their next generation AMiga OS. I doubt
that several thousands of dollars per development system would work in this
model. you may also want to look at wind rivers model for pricing of
developement seats. Should you ever have any questions about QNX feel free
to contact rhu...@qnx.com I will be more than glad to help. (or call 408
821 4601.
Steve Drake wrote in message ...
>QNX does'nt care about guys like you, one and two O.S. purchases....
>Dont bother them for that, they want mass quantity deals 1000's....
>Thats why the price is literaly a moving target. Your guess was off a
>full boat developer system will run about $3000.00 single license,
>then when you get your stuff running have them quote you on quantity
>100 or so, single runtime purchases are roughly $750 and thats just
>for the O.S. you want Socket? Thats another $300 or so....
>Then there's the ever elusive modular pricing.....
>
>If price is a major concern and your not doing mass quantity then QNX
>is NOT for you. You'll see posts here making it sound like QNX can
>walk on water but it cant, call them up and ask them for a price on
>the IAT toolkit , go ahead, they will give the run around... If they
>do give a price please post here.
>
>Better yet swing into one of their new offices and get a demo and talk
>to someone there. You cant. Those offices do not exist, every phone
>number on the web site is the Canadian office.
>
>
>
>
>On Sun, 08 Nov 1998 00:05:37 +1100, Ron Dunn <r...@relational.net>
>wrote:
>
>>I know it is an old topic, but I just can't resist one more try ...
>>
>>Have you looked at the web-page for the QNX IAT demonstration lately?
>>I was there tonight, and it reports over 500,000 downloads of the
>>toolkit. Mine's one of them. And if the rest of the downloads are
>>like mine, they represent people who are seeking an alternative
>>operating system, who are impressed by the demonstration, and who are
>>subsequently burned by the bizarre QNX distribution and licensing
>>system.
>>
>>One of the most FAQ on this newsgroup must be "What does QNX cost...",
>>and it is one of very few questions which cannot be answered clearly
>>and precisely by the otherwise helpful and informative people here.
>>
>>I have difficulty understanding why QNX is advertised in relatively
>>mass-market media, attracting people like me, when the client base
>>apparently sought by QNX is so different. It is just too hard to find
>>someone from whom one can buy the system. There is no consistent and
>>public pricing model on which R&D or simple exploratory purchasing
>>decisions can be made. There is no encouragement to "buy and try".
>>As a result, my opinion is that QNX is losing thousands of potential
>>customers, many potential applications, and maybe the opportunity to
>>become a much bigger force in the OS market.
>>
>>I really wish that QNX would would change its focus and ENCOURAGE
>>developers to try the system.
>>
>>Come on, guys, put together a standard bundling of products similar to
>>those used for the IAT. Price it at a single, bundled figure which is
>>appealing to small developers, and cost competitive for potential
>>users. A guess? Somewhere between USD 300 and USD 500 per copy. In
>>fact, why not make it even easier, and put together a time-limited
>>demonstration CD bundling all of this stuff and push it out to anyone
>>who takes the time to write and ask!
>>
>>What are you going to lose? By your own figures, there are hundreds
>>of thousands of people have downloaded your demonstration, and my
>>guess is that only a very small fraction of those have bothered to
>>wade through the morass of your current licensing and distribution
>>practices. Even if just one percent of those people bought a licence
>>at USD 300, that's USD 1,500,000. Isn't that enough to reconsider?
>>
>>Every application ported to or written for QNX represents extra OS
>>sales from the developer of that application. Most people don't go to
>>the trouble of porting software for just one use, they expect to sell
>>it to other customers, who in turn will represent purchasers of QNX to
>>run that application. From my own example, we have ten computers, and
>>would probably licence five copies. Our applications sell in the
>>banking market, and we have around five installations per month, of
>>around 3-4 computers each. Our customers don't care about the OS we
>>run, so maybe we would install 100-150 licenses per year. We're
>>small, and that's not a big figure. But take 100 companies like ours,
>>or even 1,000 companies like ours, and all of a sudden those numbers
>>look much bigger.
>>
>>Furthermore, one of those developers might just be sitting on the
>>world's next killer application. No, it's not me (I wish!), but the
>>networking / messaging model of QNX opens up so many ideas for new
>>software that the possibility can't be ignored.
>>
>>Then imagine the possibilities which emerge if some of those users
>>begin porting office software to QNX. There's a LOT of companies out
>>there with hardware which is no longer capable of running Microsoft's
>>increasingly ponderous software, who don't want to dabble in the Linux
>>market, but who have serious need for a competent, network-capable
>>multi-tasking environment to replace their older applications. There
>>is also an installed base of happy QNX customers, who would surely be
>>happy to see extended use of the OS within their organisations,
>>representing a new software market for developers.
>>
>>Well, maybe QNX wants to remain an embedded OS, or an industrial OS.
>>I think it's a waste. The IAT demonstrates the application / user
>>potential of this OS. I would dearly love to try working with it, and
>>I'm sure I'm not alone.
>>
>>Please, QNX, take a little risk, and take the chance to realise the
>>potential of your OS.
W.Stokes
10 YEARS AGO!
Where do I get the new "Kickstart w/Qnx" diskette?
Maybe QNX will be the Next OS for the Atari ST platform.
Can I also get the 6502 version of QNX for my Atari 800 and Apple IIe
?
Is this QNX's marketing startegy? Capture a dead market....
Why in the hell would anyone buy into an Amiga platform?
I can see why your busy in California, besides looking at oem deals
maybe you should be checking the classifieds, with this new direction
of QNX I have a feeling you wont be "dissplaced" for very long.
W.Stokes
April fools joke!
The Amiga platform is dead.
Its main market in the end was video and that has long dried up.
To tie their name to an out of date GAME machine is really a smart
move.
Where can I download the Amiga/QNX version of Marble Madness?
I hope they port their DOOM version over too!!!
On 14 Nov 98 21:27:51 +0000, "Paul May" <paul...@mcmail.com> wrote:
>Hi everyone.
>
>sorry for butting in on my first subscribe but i think
>its relavent to this thread.
>
>today (sat 14th Nov 98) its been confirmed that AMIGA inc/GateWay
>will use the QNX kernal for the new AMIGA OS5/Dev and /Prod Machines.
>
>http://ugn.amiga.org/events/logs/havemose.shtml (and more logs there)
>
>"
>
>QNX will provide Kernel, file system, device drivers ect,
>AMIGA will provide UI, GFX, 3D, Multimedia, video ect"
>
>the plan is basicly to give the new (3rd party built) machines the
>same leading edge Tec as the original AMIGA 1000's had in the 80's.
>
>its good for QNX as EVERYONES heard of the AMIGA and now you can bet
>the masses in the Micro$loth markets will take notice (i'd imagine).
>
>comments and your long standing thoughts on the QNX OS strenths/weaknesses
>in a mass market for the many people that are as yet a bit vaige about its
>past/future possabilitys Please.
>
>PS.
>
>any more NG's relating to QNX other than this one (on my ISP) ?.
>
Shaun C. Murray
wrote:
>If this wasnt the month of November I would have to say this is an
>April fools joke!
>
>The Amiga platform is dead.
Down but not out for the count. Which is why Gateway/Amiga Inc are
designing a new Amiga II with a new OS. QNX provides an Amiga-like
kernel. Gateway provides a lot of potential backing plus Amiga Inc.
have been talking about licencing for any kind of device from palmtops
to workstations to phones to washing machines.
>
>Its main market in the end was video and that has long dried up.
It's main market was always video - the Newtek Toaster in particular.
A lot of people doing raytracing and video work now use PC's and Mac's
running Toaster derived software. QNX's scalability would make a
kick-ass rendering OS.
The Amiga's video market dried up because CBM went belly up and no
parent meant no CPU upgrades beyond the 68060.
>
>To tie their name to an out of date GAME machine is really a smart
>move.
Yeah right. Like the PC isn't todays game machine. Some would say it's
today's out of date game machine with Dreamcast and Playstation 2
coming.
>
>Where can I download the Amiga/QNX version of Marble Madness?
You could never download the Amiga version of marble madness legally.
>
>I hope they port their DOOM version over too!!!
Whose? QNX's or Amiga's?
Bear in mind that at the announcement the QNX Pres. showed DOOM
running on QNX on two PC's.
Smart comments aside, Amiga working with QNX with a better GUI,
OpenGL, Games and media kits along with the active and talented Amiga
programming community should work well for both companies and both
communities. I look forward to a bright future.
It's interesting to note that Amiga originally (according to reports)
considered BeOS before QNX. With Amiga providing the media parts of
the OS, QNX has a chance of becoming a better mediaOS than BeOS where
it's realtime prowess will show. Work with us people.
--
Shaun
s...@enterprise.net PGP Key available
Michael Hunter
: >
: >Why then does Qnx keep advertising its 1.44 meg floppy with everything
: >you need for the internet...blah, blah, blah, if they don't want people
: >using it ( or trying to ) for general purpose stuff. If I am dealing with
: >RT,
: >why do I care that I can get on the internet with less than 1.44 meg of
: >space. More importantly, why does everyone get so surprised that
: >after people get a FREE download of this wonderful thing, that they
: >figure that they might be able to get the OS for less than $1,000.
: >
: >If Qnx doesn't want to bother with the small user ( and I don't blame them,
: >10,000 dummy users = lots of $ in support ), STOP advertising the
: >1.44 floppy to everyone.
: You've got a point there. Well worth a thought.
The demo floppy isn't about RT, its about embeddability. As with
most advertisements if it doesn't fit don't buy it. How many <name
product> advertisements do you see in the common media that you
ignore?
mph
--
* Michael Hunter (mphu...@qnx.com, http://www.qnx.com/~mphunter)
Mitchell Schoenbrun
> Is this QNX's marketing startegy? Capture a dead market....
Why would anyone assume so many years later that the target
computer has anything to do with the old Amigas?
There might not be any vestige of the old company, someone
might have just bought the trademark for name recognition.
That's what happens at bankrupcy sales.
Even if this isn't the case, who knows what niche this company
has been surviving in. I think I remember seeing a set of
networked computers with the Amiga name on it a few years ago
in one of the first multi-user 3D arcade games. The one where
you shoot other players on a square 3D board, with 4 raised
platforms.
Igor Kovalenko
>
> The demo floppy isn't about RT, its about embeddability. As with
> most advertisements if it doesn't fit don't buy it. How many <name
> product> advertisements do you see in the common media that you
> ignore?
>
With all my respect Mike, I don't believe you are right here.
It doesn't really matter how many of ads you ignore. It matters how many
of them are _potentially_ affordable to you, assuming you have some
interest in the subject.
I'm not a marketing guy, but what I feel is, every ads should be
targeted to some reasonably selected audience. And, according to common
sense, it should meet common expectations of that audience. If you show
a candy ads in the Cartoon Network, children (well, their parents) don't
expect to spend thousands of dollars to buy that candy.
What QSSL did with the demo, is something opposite. Common expectations
are already established by other software vendors. And then the audience
see a pretty new guy who shows them a sweet candy. It looks so sweet
that they throw their business and run to the touchphones like crazy,
with their credit card handy, but ... only to find out that 99% of them
simply can't afford it.
Well, I guess the next trick would be to try The Christmas Sale Approach
(tm). Rise the price to $10,000, then send tons of Santas to
supermarkets with big banners like "QNX for sale. 90% discount for
limited time only". When people will start to attack Santas, they should
just politely answer: "Ignore us please. We don't know price and we
don't even have a copy. Ask you nearest distributor".
Wow, that turns out to look like a nice Christmas tale ;)
Cheers,
- Igor
Glenn Sherman
Michael Hunter <mphu...@qnx.com> wrote in article <72pdg8$l...@qnx.com>...
> Inge Vabekk (in...@arxi.no) wrote:
> : Glenn Sherman wrote in message <01be0e4e$e36c0de0
> : >Inge Vabekk <in...@arxi.no> wrote in article
> : >
> : >Why then does Qnx keep advertising its 1.44 meg floppy with everything
> : >you need for the internet...blah, blah, blah, if they don't want
people
> : >using it ( or trying to ) for general purpose stuff. If I am dealing
with
> : >RT,
> : >why do I care that I can get on the internet with less than 1.44 meg
of
> : >space. More importantly, why does everyone get so surprised that
> : >after people get a FREE download of this wonderful thing, that they
> : >figure that they might be able to get the OS for less than $1,000.
> : >
> : >If Qnx doesn't want to bother with the small user ( and I don't blame
them,
> : >10,000 dummy users = lots of $ in support ), STOP advertising the
> : >1.44 floppy to everyone.
>
> : You've got a point there. Well worth a thought.
>
> The demo floppy isn't about RT, its about embeddability. As with
> most advertisements if it doesn't fit don't buy it. How many <name
(if it doesn't fit, don't BUY it). BUT you are giving it away free to
every
Tom, Dick, and Harry who stumbles across it. I REALLY don't mind that
you do that, but you don't give people any idea of what to expect when
they want to BUY it. Then they all start in on this newsgroup flamming Qnx
for not telling them.
I like Qnx, but I would like to see less posts to this newsgroup about
a) how much does it cost ?
b) where can I get it ?
c) now that I got this free demo, why does the OS cost so d**n much ?
( an aside here...
Now you are going to tell me that it is so expensive because of the
RT,
but you said before that the demo isn't about RT.
)
Why not give people a little more info on the web page before they download
it ?
> product> advertisements do you see in the common media that you
> ignore?
>
> mph
>
> --
> * Michael Hunter (mphu...@qnx.com, http://www.qnx.com/~mphunter)
Shaun C. Murray
wrote:
>W.Stokes <wst...@hotmail.com> wrote:
>
>> Is this QNX's marketing startegy? Capture a dead market....
>
>Why would anyone assume so many years later that the target
>computer has anything to do with the old Amigas?
Exactly. The new Amiga doesn't have that much to do with the old as a
hardware basis. Where it (hopefully) will be similar is in feel,
elegance and efficiency.
>There might not be any vestige of the old company, someone
>might have just bought the trademark for name recognition.
>That's what happens at bankrupcy sales.
Actually, Gateway originally bought Amiga for the patent portfolio but
realised there was a community of people attached with it that were
strangely attached to the old thing. They now have a number of staff
on board from the old Commodore and as was pointed out at the
QNX<>Amiga announcement, both QNX's founders had Amiga 1000's and 60%
of the staff are Amiga owners past or present.
>
>Even if this isn't the case, who knows what niche this company
>has been surviving in.
The alternative one that doesn't like the mainstream offerings, the
old romantics that like the old girl, the video market that still
can't find anything that does NTSC/PAL as well for as little money....
> I think I remember seeing a set of
>networked computers with the Amiga name on it a few years ago
>in one of the first multi-user 3D arcade games. The one where
>you shoot other players on a square 3D board, with 4 raised
>platforms.
Virtuality. They use a network of Amigas with custom graphics boards.
Or at least did. I suspect they've moved on now.
ur...@my-dejanews.com
> I think we can all agree that when it comes to RTOS - Qnx is at the
> top of its class, and it is probably priced right where it should be for
> that.
>
> Why then does Qnx keep advertising its 1.44 meg floppy with everything
> you need for the internet...blah, blah, blah, if they don't want people
> using it ( or trying to ) for general purpose stuff. If I am dealing with
> RT,
> why do I care that I can get on the internet with less than 1.44 meg of
> space. More importantly, why does everyone get so surprised that
> after people get a FREE download of this wonderful thing, that they
> figure that they might be able to get the OS for less than $1,000.
>
> If Qnx doesn't want to bother with the small user ( and I don't blame them,
> 10,000 dummy users = lots of $ in support ), STOP advertising the
> 1.44 floppy to everyone.
Amen, brother. You read my mind (how do you do that? :-)
I think when it comes to RT, you know what price is right for you, so no
argument there. The problem is QNX's marketing focus: all over their web
site, even when they're talking RT, they keep emphasizing the compact kernel,
embeddability, etc. etc. ad nauseum. Again and again: internet kiosks,
embedded devices, lean OS. Pardon my presumption, but if I'm going to sell a
set-top box for $400, there's NO WAY I'm paying that much again (or
considerably more) for the OS running it. How can I pass that cost on to the
user?
Please no more arguments about me not being their intended market! I seem to
be EXACTLY the intended market; they put at the very least as much emphasis
on embeddability as on RT. I download the 1.44" demo, run it and say: great,
that's what I want in my device, only in ROM. How much? That's where the deal
falls through.
I think they have to wake up and formulate their markets. I can see at least
two: RT and embedded, and you can't possibly treat them both the same. We're
talking completely different economies. If they sold QNX with all its
features EXCEPT RT capabilities for a fraction of the price, that would work.
Paying $30 to $50 royalties for the OS on a $400 device is doable. As someone
on this thread said, most embedded customers don't need RT. There's no need
for the developer of an internet kiosk to be on the same pricing plan as the
developer of a kidney dialysis machine when it comes to the OS. After all,
THEY TO are after different markets willing to pay vastly different prices.
I think everybody in here defending the QNX marketing has it wrong. Then
again, I think a lot of them aren't spending their own money, so they don't
mind. It's easy to justify flying first class when your expense account pays
for it.
I wasn't being hypothetical about the $400 device, either: this morning I
demonstrated the QNX 1.44" demo to my boss for possible use in an RF PDT
(portable data terminal). It's perfect: we fill out a lot of forms that get
logged into an SQL backend, but some of the PDTs also do some processing of
their own, where Java comes in handy. Sometimes the RF link might be down, in
which case the forms would be hosted by the embedded web server until a
connection is reestablished. And since we are trying to minimize the cost of
the PDT, QNX's frugality is a godsend. Obviously we don't care about RT. Pity
we found out the royalties per device, which sent us running for the hills.
Now I'm investigating other solutions: maybe embedded Linux? I don't know how
far along they are with that. DOS won't work because web browser and web
server both need to be running on the PDT at the same time. Oh well...
Uwe Wolfgang Radu
American Engineering
despamm...@americanengr.com
-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
Igor Kovalenko
>
[snip]
>
> http://ugn.amiga.org/events/logs/havemose.shtml (and more logs there)
>
> QNX will provide Kernel, file system, device drivers ect,
> AMIGA will provide UI, GFX, 3D, Multimedia, video ect"
>
Ok, after reading that I feel curious. It looks like Amiga OS5 will
actually be plain Neutrino+Photon, with multimedia extensions. Not that
I dislike it, actually I'm happy about it. Never had Amiga but heard
only good things about it. So, guys behind it must be good enough to
implement what many of us dreamed about for so long time.
What is curious, is how QSSL may treat this new OS. It actually may
become a direct competitor, unless some functionality will be
intentionally restricted or Amiga will run only on a totally different
platform. What about runtime prices? They can't be as high as for QNX
for Amiga market, but otherwise QSSL will have to drop runtime prices
too... or it will compete against itself like with QNX Windows & Photon?
> PS.
>
> any more NG's relating to QNX other than this one (on my ISP) ?.
>
Unless you have some specific techincal questions, you better stay here.
It is the only public newgroup generally accessible through ISPs
worldwide.
- Igor
Felix S. Gallo
douglas shawhan writes:
>I have dd'd about 30 floppies with the internet demo. (an excellent use for
>my old win 3.11 diskies) Whenever CS kids come to my store to sell or buy
>textbooks I hand them one. I feel like a moonie...
I've done that too. Usually with the obvious result: "wow! This is
pretty damn cool! How much does it cost to develop for?" "Uh...
several thous...wait, come back!"
Another point that is easy to miss is that things like Linux are not
only almost as embeddable as QNX (I once put Linux in 2M ROM
and 4M RAM), but Moore's Law is making the bar move up. Pretty
soon, it'll be silly not to put 32M of RAM in a cheap embedded
device, because they won't even make 4M RAM modules any more.
The 'size' advantage of QNX is shrinking rapidly.
And, the 'RTOS' feature of QNX is similarly disappearing. RT-Linux
has a subsystem which is just as realtime as QNX's, and even if
you didn't want to use that, the hardware is getting so fast that
just relying on preemptive multitasking is getting more viable for
more and more tasks.
QNX's attempts to move 'up' in the world (graphical desktops,
browsers, phone dialers) are acknowledgements of the fact
that real time is now 'complete' -- QNX can't get any more
'real time', customers don't need the time to be any realer,
and the only way QNX can add new neat stuff is by going into
the usual desktop world. Here Linux competes directly
against QNX and totally dominates the playing field. It's
as reliable as QNX, as small as a tricked-out QNX, and
has much better tools, interfaces, and third party software.
Theoretically, I'm the exact customer QNX should be aiming
for -- I design tiny high volume internet appliances with embedded
operating systems for a living as part of a series of major hardware
and software design companies. And even though their technical
advantage is decreasing, they've successfully won me over on
tech merit four times in a row. That's not an easy task in this
industry, much less the dog-eat-dog embedded operating system
world.
But right now I have windows open on vxworks embedded targets
and have another embedded system under my desk, which used
to run the full QNX dev suite, running Linux. It was QNX's territory --
they owned the mind of the guy making the tech decisions -- they
had a good customer contact -- they had the product in the door --
I had bought the full dev system at list price quantity 1 -- my needs
were an exact fit against QNX's OS...
...but a product has to cost less than N dollars to make it. RAM,
flat screen and CPU use up most of the budget. An OS which
then costs, say, $15 per copy translates to about $40 in end user
price. Compared with an OS that costs $4 per copy ($10 end
user price), that's a premium of $30 price per unit. Even if
QNX is a hell of a lot more fun to program for than vxworks,
you can bet that a $30 relative price disadvantage vs. the
inevitable low-price cloning competitor does not make up for
the innately beautiful architecture that the end user will never see.
In my wild fantasies from which I routinely wake up shouting
hallelujahs, QSSL calls me and says "Felix, we will sell you QNX
with graphics, networking, browser, and JVM for less than $8
per copy." 6 months later, the world becomes a far better place
to live. But until then, alas, back to Tornado.
F.
Lance Roberts
> embedded devices, lean OS. Pardon my presumption, but if I'm going to
sell a
> set-top box for $400, there's NO WAY I'm paying that much again (or
> considerably more) for the OS running it. How can I pass that cost on to
the
> user?
> demonstrated the QNX 1.44" demo to my boss for possible use in an RF PDT
> (portable data terminal). It's perfect: we fill out a lot of forms that
get
don't care about RT. Pity
> we found out the royalties per device, which sent us running for the
hills.
I'd always estimated that a large market device would have a small royalty
fee, especially since you would probably use a subset of Qnx OS &
utilities. what size market were you estimating, and how much were the
royalties per device that Qssl wanted?
regards
Lance Roberts
Lance Roberts
<72q05a$a...@enews1.newsguy.com>...
>
> douglas shawhan writes:
> >I have dd'd about 30 floppies with the internet demo. (an excellent use
for
> >my old win 3.11 diskies) Whenever CS kids come to my store to sell or
buy
> >textbooks I hand them one. I feel like a moonie...
>
> I've done that too. Usually with the obvious result: "wow! This is
> pretty damn cool! How much does it cost to develop for?" "Uh...
> several thous...wait, come back!"
That's really not the final answer, it's just that some of the developers
and consultants keep their cards close to their hands. Most of my
development work is done from a W95 PC on my desk that is running a
phindows session to a qnx server. It's handy in my case that I am the sole
user of this server for speed of compilation, testing, etc, but it is easy
to see that the server doesn't need to be just down the hall in this case,
I just need a password onto it.
I can envisage that anyone wanting to develop and test a user/pc product -
such as a word processor or similar, would only need a copy of phindows (or
just telnet if you don't want graphics) and just log into a remote qnx
server that is equipped with compiler, etc. the limitations here are
processor speed, memory, but I haven't seen much in the way that qnx
licenses it's OS that would prohibit this.
This would enable many developers to share a server for a much reduced
cost.
It really dependeds on what you are developing, whether it is feasable to
develop in this manner. To save on compiler costs, you could send any C
code to a server for compilation, then run the exe on your own qnx pc.
There are many possibilities for sharing because qnx is so flexible.
To save on Tcp/ip you could just DosFsys programs to W95, then use their
TCP/IP to ftp to a server, telnet to request compile, etc.
But you will quickly be able to justify buying your own development system
if you want to do this stuff often, and faster, more efficiently, without
extra shell scripts, time sharing with others, etc.
But for a beginner it could be a good starting point. (you can't phindow's
to quics.qnx.com or compile yet though, maybe someone could see a market
here though.)
> Another point that is easy to miss is that things like Linux are not
> only almost as embeddable as QNX (I once put Linux in 2M ROM
> and 4M RAM), but Moore's Law is making the bar move up. Pretty
> soon, it'll be silly not to put 32M of RAM in a cheap embedded
> device, because they won't even make 4M RAM modules any more.
> The 'size' advantage of QNX is shrinking rapidly.
No, it means you will be able to develop bigger/better systems with qnx
than others for the same memory.
> And, the 'RTOS' feature of QNX is similarly disappearing. RT-Linux
> has a subsystem which is just as realtime as QNX's, and even if
> you didn't want to use that, the hardware is getting so fast that
> just relying on preemptive multitasking is getting more viable for
> more and more tasks.
viable yes, but any product running qnx will still have an edge. Look at
mp3's, real movies, etc. They require a certain processing speed, when I
bought my old 486 I didn't know about them, then when I did I found with
W95 I couldn't quite run them at a usable speed, with a more efficient os,
some old hardware can creep over the definate line of not acceptable to
acceptable. It's easier to get this sort of speed out of qnx os than
others, and you can have more things running without it skipping/stalling
if you set it up right.
> QNX's attempts to move 'up' in the world (graphical desktops,
> browsers, phone dialers) are acknowledgements of the fact
> that real time is now 'complete' -- QNX can't get any more
> 'real time', customers don't need the time to be any realer,
> and the only way QNX can add new neat stuff is by going into
> the usual desktop world. Here Linux competes directly
> against QNX and totally dominates the playing field. It's
> as reliable as QNX, as small as a tricked-out QNX, and
> has much better tools, interfaces, and third party software.
it just shows there is room for growth with qnx, just total capability of a
hardware device is limited once the owner decides not to upgrade it (eg.
playstations are all same hardware, but new improvements to graphics
engines, etc, are bringing out more impressive games all the time. same
hardware, more efficient software is really how most systems are upgraded
around here, except the M$ guys who put advertised more capable software,
then need better hardware to get same useability)
If you are forward thinking enough when you build your embedded devices,
the first product can be improved just with software later down the track.
> Theoretically, I'm the exact customer QNX should be aiming
> for -- I design tiny high volume internet appliances with embedded
> operating systems for a living as part of a series of major hardware
> and software design companies. And even though their technical
> advantage is decreasing, they've successfully won me over on
> tech merit four times in a row. That's not an easy task in this
> industry, much less the dog-eat-dog embedded operating system
> world.
If you allow it to have it's software upgradeable somehow, then you will be
surprised how many extra features people will want later on down the track.
m You will probably be stuck with a large consumer base with same old
hardware, but if you can improve the existing device with just software,
then it will still be worth something. Many consumers will appreciate a
low cost upgrade, because when new features come out that they want, they
won't have to throw away the old hardware (being a physical object, they
will often form an attachment to it)
> ...but a product has to cost less than N dollars to make it. RAM,
> flat screen and CPU use up most of the budget. An OS which
> then costs, say, $15 per copy translates to about $40 in end user
> price. Compared with an OS that costs $4 per copy ($10 end
> user price), that's a premium of $30 price per unit. Even if
> QNX is a hell of a lot more fun to program for than vxworks,
> you can bet that a $30 relative price disadvantage vs. the
> inevitable low-price cloning competitor does not make up for
> the innately beautiful architecture that the end user will never see.
yes this is a good point, the times where qnx really makes a difference is
when you are making something that is really making the most use of your
hardware, so you are doing something with the same hardware cost that you
physically couldn't achieve with another os.
Maybe you will have to add more features to it until you reach this limit.
(people like features after all)
> In my wild fantasies from which I routinely wake up shouting
> hallelujahs, QSSL calls me and says "Felix, we will sell you QNX
> with graphics, networking, browser, and JVM for less than $8
> per copy." 6 months later, the world becomes a far better place
> to live. But until then, alas, back to Tornado.
I expect it will get cheaper as they sell more of it. Forward planning for
this is risky though. I expect they want a slightly larger user base before
they take that risk.
I guess from their point of view, if they sell it to you for $8, they sell
it to similar customers for that price.
these other customers may place a higher value on it as it gives them the
edge they want, so they might already be prepared to pay more.
if you can't justify the full features that qnx gives with the price they
want, then you will have to settle for something else I guess. But someone
else will see/use the real potentials, and maybe they'll be better for it
in the long run.
For people that say that they can't see the potentials without playing with
it, I think that is where the demo comes in.
regards
Lance Roberts
Paul May
>> Is this QNX's marketing startegy? Capture a dead market....
simple answer ,ASK THEM.
we ,in the AMIGA community see your kind of 'knee jerk' reaction
all the time ,the simple fact is that the Original AMIGA 1000/256k
machine of the 80's was a ground breaking machine AT THAT TIME.
the most common IBM compatable at that time was the 8086 & 286AT
machines, the difference is that the 1000 CAN be expanded to run
todays current S/W ,can you say the same of the 286 !!!.
pushing the distant past aside for the moment, as of today
the AMIGA market consists (mainly) of 68040/68060 motorola CPU's
with the option of adding a PPC coprosessor board.
TRUE is the fact that this CURRENT AMIGA market isn't in the same leage
as the WinTel market of TODAY,
*However* the market is as we speak is moving to 'Digital Convergence' ,
as in an upto date the C=CDTV of years gone by.
as for the next AMIGA2 and its relation to QNX, thats the reason
i'm reading this list, to get informed feedback from the long time
QNX users/3rd party developers, NOT to be informed by such as YOU
about how my platform of choice is dead.
the funny thing is that if you were to DO A SIMPLE SEARCH
of the web YOU WILL find that we are indeed alive and ready
to enlighten such as you about the future, who knows, YOU
MIGHT EVEN BUY ONE in 2000+ when you see what its cabable of ?.
>Why would anyone assume so many years later that the target
>computer has anything to do with the old Amigas?
thank you Mitchell, its nice to see that that NG at least
has 1 person thats open minded and willing to listen.
it might be that you wont be interested in the new machines
(and thats fine), i guess it all depends on weather your
main reason for (presumably) useing QNX is specialist or pleasure ?.
>There might not be any vestige of the old company, someone
>might have just bought the trademark for name recognition.
>That's what happens at bankrupcy sales.
as it turns out (officially), thats exactly what happened,
GATEWAY(2000) payed for the pattents and were really suprised
to receave masses of emails.letters and so on from the
(sometimes rabid <G>) AMIGA core that they re assesed their
position and (long-short) ended up setting up AMIGAinc
and taking the old AMIGA international under its wing.
the thing that gets me, is NON/EX AMIGA people dont realise/forget
that Commodore were infact mainly a x86PC company and only
ended up with the AMIGA because ATARI
(they have a new machine out now to Mr Stokes)
wouldnt pay the extra cash to the lorain (AMIGA) people.
so infact we have gone full circle and ended up here,
owned by another BIG PC company and looking to a relatively
unknown 3rd party (no offence to QNX) for help recreating
a BASE kernel for the next BIG thing.
as it happens (presumably Mr stokes doesnt know) AMIGA inc
asked all the people at QNX if they had heard of the AMIGA
the response was that the 2 head guys first computers were
AMIGAs and infact 60%+ of the 200 ? people there also owned one
and were VERY exited to be placed in a position to help
bring it back to the massess.
i guess that explains why EXEC (current AMIGA kernel)
and QNX are said to be simular in concept/feel
although the EXEC has suffered due to the lack of
concurrent owners to develop it until now.
anyone here familiar with both these to comment on
that being the case or NOT ?.
>Even if this isn't the case, who knows what niche this company
>has been surviving in. I think I remember seeing a set of
>networked computers with the Amiga name on it a few years ago
>in one of the first multi-user 3D arcade games. The one where
>you shoot other players on a square 3D board, with 4 raised
>platforms.
i'm not really a games player so i cant help you on that one.
they are still used in NASA (as i'm told so is QNX ? what systems)
the simple fact is that we DID survive, contrary to Stokes
comments (do a search), we are now going to use QNX as the BASE
crossplatform kernel, because of limited time constraints.
the market will be devided in to 'Classic' covering everything
from the past including the PPC options now available,
and the next 'AMIGA2' 3rd party built 'Digital Convergence'
STB's, DeskTop Machines and distributed Networking Manchines.
its unclear (covered by NDA's) right now weather AMIGAinc will
just produce a reference machine for the 3rd partys to follow
as a base, or infact they will make & market the first machines.
AMIGA inc's main goal right now, is to build a NEW OS (HAL and so on)
on top of the QNX kernel.
i dont know about the QNX Community, but in the AMIGA Community
we are used to talking directly with the S/W and H/W developers
directy, and indeed we have many of them in Team AMIGA.
AMIGA inc have & are listened to the grass roots users/developers
and are incorporating MANY things at our request, such as the
disability working group projects, anti Piracy projects and
many more.
there were details posted a while ago about the expected minumum
spec for the new machines but i cant find the info right now
so i'll wait before reposting that (if your interested ?).
on a side note, perhaps people here might take a look at
REBOL from the original AMIGA OS developer Carl S
at http://www.rebol.com i'm not sure if he's already
done a QNX OS port yet but i'm sure it will now be in the works
he's gone an AMIGA, win9x, Unix and other ports so far.
rebol will complament the strengths of the QNX networking
and probably be included(providing they can agree on terms)
in the next OS along with AREXX.
--
Paul May
Team
*AMIGA*
<tsb> User Of The small OS With A BIG Future
--
<tsb> Looking For Disabled persons Ideas
<tsb> For Better Computer Interaction & Inclusion
--
Paul May
>April fools joke!
theres only 1 fool here and its not me , look in the mirror sometime.
>The Amiga platform is dead.
sure, if YOU say so
>Its main market in the end was video and that has long dried up.
really, well if you say so IT MUST BE TRUE, not.
>To tie their name to an out of date GAME machine is really a smart
>move.
Hmmm a games machine that set the standard LONG BEFORE your
current gaming win9x box EVER DID, rather a good move i'd say.
i'm presuming your a spoty 14 year old that loves his games
and his win9x OS given your very limited short sighted views.
>Where can I download the Amiga/QNX version of Marble Madness?
ARRR so your a feaving PIRATE THEN ?, you think its CoOl to get
free WaREz rather than pay a fair price for a fair product ?.
i guess the NEW AMIGAs with their ANTI Piracy devices want be the
thing for you then.
good news for the AMIGA/QNX S/W and H/W developers
past and future though.
>I hope they port their DOOM version over too!!!
they already did, that was part of the Press demo, as you would have
realised if you had bothered to read the logs before shouting off your
mouth.
whats the matter ?, perhaps you dont like the idea that a NEW AMIGA
could retake your win9x gameing box crown at its release.
just like the Original AMIGA, the new one will be FAR more than
a mear game box, it just happenes that the average guy in the street
will want to use it as such ,so why not include gaming in the thing,
after all its going to be used for HDTV, Video, Sound, PC emulation
among other things, so why not use 2% of its potentual for FUN things.
you do remember haveing FUN useing the computer, Ohh win9x i forgot
perhaps you DONT.
if you think the new AMIGA will not suit your needs in a new
machine , then FINE i have no problem with that, but please
its plain that you know NOTHING about the current machines
never mind the potential/spec of the New one's so stop
trying to be clever and making childish remarks about
many peoples prefered choice of OS/machine.
who knows ,perhaps given time you might be even consider buying
one, if it base price/spec far out performs the PC at *that* time ?.
>On 14 Nov 98 21:27:51 +0000, "Paul May" <paul...@mcmail.com> wrote:
>>Hi everyone.
>>
>>sorry for butting in on my first subscribe but i think
>>its relavent to this thread.
>>
>>today (sat 14th Nov 98) its been confirmed that AMIGA inc/GateWay
>>will use the QNX kernal for the new AMIGA OS5/Dev and /Prod Machines.
>>
>>http://ugn.amiga.org/events/logs/havemose.shtml (and more logs there)
--
Paul May ,UK
Team
*AMIGA* Worldwide MEMBER
Guy Macon
>>Its main market in the end was video and that has long dried up.
>
>It's main market was always video - the Newtek Toaster in particular.
>A lot of people doing raytracing and video work now use PC's and Mac's
>running Toaster derived software. QNX's scalability would make a
>kick-ass rendering OS.
>
>The Amiga's video market dried up because CBM went belly up and no
>parent meant no CPU upgrades beyond the 68060.
Every TV studio has a couple of old Amigas stiil in use. It was and is
the only PC to output standard NTSC video as it's main output. They make
great titlers. Also, check out the various websites for the TV show
BABYLON 5. Guess what computer they use to do all of their graphics?
Yup. Stacks of Amigi.
Now consider Gateway. They are very profitable now, but in a businesss
where even the biggest can die quickly. It makes sense for them to have
a fallback position selling something that they have exclusive rights to.
With Apple dying from self inflicted wounds, there is a place for a 10%
market share alternative to Microsquish.
Paul May
>10 YEARS AGO!
shame you never spent any money on them to expand they above
the basic out of the box and slap a game in.
i'v got a IBM 8080/640k, can i run win9x on it <G>,
see we can all make silly statments,it doesnt impress anyone though
(get the picture ?).
>Where do I get the new "Kickstart w/Qnx" diskette?
in the new OS5/developer boxs around dec/jan OR
in the OS5/Production models in AMIGA2 around 18 months.
>Maybe QNX will be the Next OS for the Atari ST platform.
ROTFL , perhaps so, YOU DO REALISE that there
IS A NEW ST Machine out now as well.
>Can I also get the 6502 version of QNX for my Atari 800 and Apple IIe
>?
well these machines ARE infact well emulated on many platforms
includeing the amiga, so i guess when we move to QNX the anser
is YES.
appart from that, perhaps QNX has been ported to some of these CPU's,
i dont know, but QNX is AFAIK currently an embeded OS
i'm sure the guys will enlighten you.
>Is this QNX's marketing startegy? Capture a dead market....
Hmmm so you consider 'DIGITAL CONVERGENCE' a dead market ?,
perhaps you should do some reading mate ,you dont seem very
well up on current trends in the world markets.
>Why in the hell would anyone buy into an Amiga platform?
ask that same question AFTER you have seen the NEW MACHINES.
>I can see why your busy in California, besides looking at oem deals
>maybe you should be checking the classifieds, with this new direction
>of QNX I have a feeling you wont be "dissplaced" for very long.
Arrr, so you 'have a feeling' ,well that explains it then 8)
perhaps Gateway AMIGAinc and the QNX guys should be worryed then.
--
Paul May
Team
*AMIGA*
W.Stokes
>i'm presuming your a spoty 14 year old that loves his games
>and his win9x OS given your very limited short sighted views.
Not quite. I happen to still have that Amiga 1000, do you?
>>Where can I download the Amiga/QNX version of Marble Madness?
>
>ARRR so your a feaving PIRATE THEN ?, you think its CoOl to get
>free WaREz rather than pay a fair price for a fair product ?.
The Amiga platform was one of THE BIGGEST piracy groups I ever saw!
Larger than the Atari platform!
I'm sure you havent installed the same QNX license on a different
machine that wasnt hooked the same network, because your an honest
pilar of the community.... what bullshit.
>i guess the NEW AMIGAs with their ANTI Piracy devices want be the
>thing for you then.
What a DONGLE?
>good news for the AMIGA/QNX S/W and H/W developers
>past and future though.
>
>>I hope they port their DOOM version over too!!!
>they already did, that was part of the Press demo, as you would have
>realised if you had bothered to read the logs before shouting off your
>mouth.
So there is a Neutrino port, wow please post the url to it...
>whats the matter ?, perhaps you dont like the idea that a NEW AMIGA
>could retake your win9x gameing box crown at its release.
You really are a moron Paul. 3D0 ring a bell, this was from the makers
of the Amiga... Qnx going to port to that too?
Jim Lambert
This guy/gal (W. Stokes) is not a regular poster of this newsgroup and does
not represent my opinion or most likely the opinion of the other regulars on
this newsgroup. This newsgroup is frequented by mostly professionals and I
for one would sure like to keep it that way. I never used an Amiga but I am
always interested when QNX gets exposure to new people/ideas and hope that
this new announcement has nothing but good news for both QNX and Amiga.
Welcome to the group and just try to ignore the trollers.
Jim
Paul May wrote in message <2722.624T21...@mcmail.com>...
>>If this wasnt the month of November I would have to say this is an
>>April fools joke!
>
>theres only 1 fool here and its not me , look in the mirror sometime.
>
>>The Amiga platform is dead.
>
>sure, if YOU say so
>
>>Its main market in the end was video and that has long dried up.
>
>really, well if you say so IT MUST BE TRUE, not.
>
>>To tie their name to an out of date GAME machine is really a smart
>>move.
>
>Hmmm a games machine that set the standard LONG BEFORE your
>current gaming win9x box EVER DID, rather a good move i'd say.
>
>i'm presuming your a spoty 14 year old that loves his games
>and his win9x OS given your very limited short sighted views.
>
>>Where can I download the Amiga/QNX version of Marble Madness?
>
>ARRR so your a feaving PIRATE THEN ?, you think its CoOl to get
>free WaREz rather than pay a fair price for a fair product ?.
>
>i guess the NEW AMIGAs with their ANTI Piracy devices want be the
>thing for you then.
>
>good news for the AMIGA/QNX S/W and H/W developers
>past and future though.
>
>>I hope they port their DOOM version over too!!!
>
>they already did, that was part of the Press demo, as you would have
>realised if you had bothered to read the logs before shouting off your
>mouth.
>
>whats the matter ?, perhaps you dont like the idea that a NEW AMIGA
>could retake your win9x gameing box crown at its release.
>
>--
>
John Birch
<snip>
Thanks for the url. Like other posters here, I am interested in the
announcement. I guess however we will now see a 'fire' of inflamatory
speculation about this news.
Unlike R. West, W Stokes et al, I am not interested in mindless
debates about my one's bigger / better than yours! Seeing who can piss
highest up the wall is a game best left for those who like urine on
their trousers (pants).
I was always interested in the Amiga, although I never owned one. It
is interesting how there might, just might be another contender in the
desktop market - even if it does end up being a niche. I just hope
that no attempt is made to provide Windoze emulation. A far better
approach IMHO is to provide compatibility at file level rather than to
host foreign apps. Alternatively Java applet support.
regards John B.
John Birch
wrote:
My two pence worth;
The QNX IAT 1.44Mb demo disk is an impressive demonstration of the
capabilities of the QNX OS. Evidently it shows that QNX is suitable
for the development of Internet Appliances, BUT I think its purpose is
multi-fold, in demonstrating how compact etc etc it is as an IAT it is
showing that QNX is generally small and efficient. Thus people who
have no interest in developing an Internet Appliance can still be
impressed by QNX. Most developers can appreciate the complexity of
implementing an Internet web browser, thus to hand out diskettes with
an OS and a browser on them sends a strong message about QNX's overall
capabilities, not just it's suitability for Internet use.
There is a lot of bandwidth in this newsgroup thread about how QNX
appears to be targetting the 'wrong' set of people, I.e. why develop
something that looks like I can use it, then to find that I can't
because it's too expensive. At the end of the day, ALL advertising
works that way. The hit rate for advertising is way down in the noise
statistically. Even if 1 person in ten thousand is a serious contender
for QNX then the IAT disk has done its job. After all, most software
demos now come on CD, so by comparison the QNX IAT sends a very strong
message.
The client base that QSSL seek is one that is prepared to fit THEIR
marketing / sales projections. The moment that they see a decline in
market, will be the moment that they change the strategy. I certainly
agree that it would be nice if QNX was much cheaper, etc. etc, but I
can't see how the business model would work. After all, would you want
the same type of support / sales model as Windows / Borland (sorry
Imprise) et al.
I don't believe that QSSL is that interested in small developers,
unless they have the kind of application that fits their pricing
strategy. I am fortunate, I work for a company where the software / PC
hardware represents about 1% of the product cost. Consequently we are
able to use QNX and enjoy the benefits of QNX because the cost of
doing so is irrelevant in the scheme of things. There are other
posters in the thread who appear to be in a similar situation. I
sympathise with those who cannot join the party for budgetary reasons.
BTW, for those readers on the US / Canadian side of the pond, you
already enjoy QNX at considerably cheaper than us over here. This goes
for all software / hardware in general (basically the US dollar price
is the UK sterling price!), so consider yourselves lucky ;-).
The problem with a 'Lite' version is that although there might be
extra revenue from sales, the support costs to QSSL would probably be
considerably higher than those in the current sales strategy. What
might be possible is an unsupported cut-down version e.g. an academic
version. (Any support would come from this newsgroup for example).
I still believe that if somebody out there is developing a set top
box, and they have an order for several hundred thousand of them (and
that is probably a minimum quantity) then QSSL will be talking to them
about pricing in the ballpark that has been discussed in this thread.
(~$20 ?). Why, because from QSSL's point of view, they are supporting
maybe ten to twenty developers. The duplication / installation of the
OS is done by the customer, so the cost to QSSL is limited. The
revenue is a figure that QSSL set in order to get the market share
against Wind River, WinCE et al.
If you are a developer working in a company of less than 100 people,
and you don't have a software contract from a giant corporate in your
hand to develop their set top / PDA software - can you seriously
expect QSSL talk about volume discounts? Show me the volume - I'll
show you the discount must surely be the response!
Final parting thought. I expect that every company seriously
developing set top / internet appliances / PDA's has experienced QNX
via the IAT demo. That was the point of it!
No direct mails please!
regards John B.
Craig Sutton
W.Stokes <wst...@hotmail.com> wrote in article
<EAD575537CB054DD.13A8461F...@library-proxy.airnews.ne
t>...
> An Amiga , I used to have an Amiga 1000 and an Amiga 500!
> 10 YEARS AGO!
> Where do I get the new "Kickstart w/Qnx" diskette?
>
Your sounding like a total fool, I bet you don't even know what the current
amiga models are capeable of.
Andrew Thomas
Well, if your unit volumes are high enough, you CAN get the price to
come down a long way. One of the reasons that QNX does not give out
prices is because they are flexible based on current and future
volume. If you are only going to build 1,000 units, what is QNX's
incentive to drop the price? If you are going to build 1,000,000
units, my guess is that you can cut a deal for under $20. Only your
sales rep knows for sure, but their volume pricing is pretty
agressive.
I do not speak for QNX, or with any actual knowledge of their pricing
structure. I've just had customers who have purchased QNX in volume.
Andrew
-----
Andrew Thomas, President, Cogent Real-Time Systems Inc.
2430 Meadowpine Boulevard, Suite 105, Mississauga, Ontario, Canada L5N 6S2
Email: and...@cogent.ca WWW: http://www.cogent.ca
Guy Macon
>Paul,
>
>This guy/gal (W. Stokes) is not a regular poster of this newsgroup and does
>not represent my opinion or most likely the opinion of the other regulars on
>this newsgroup. This newsgroup is frequented by mostly professionals and I
>for one would sure like to keep it that way. I never used an Amiga but I am
>always interested when QNX gets exposure to new people/ideas and hope that
>this new announcement has nothing but good news for both QNX and Amiga.
>
>Welcome to the group and just try to ignore the trollers.
We should expect a bunch of Amiga fans dropping in, with well over 90%
of them being reasonable (but ignorant of what QNX is). Gentle
explainations are in order for this group. for the non reasonable ones,
just move on without posting a reply and they will get tired of shouting
into an empty hall.
Leif Bloomquist
> UNIX, Windows 95, Windows CE, Windows NT, Linux, and FreeBSD --
> all competitors in one way or another in the x86 space -- make their
> name for themselves, and a serious amount of money, by seeding
> the market with extremely inexpensive starter kits and providing
Apparently, someone from QNX was on the University of Waterloo campus a
couple of
weeks ago, and handed out 1,000 (yes, 1 with three zeroes) free CDs with
QNX and Watcom. (There's more info on the QNX and UW websites).
What's interesting here is, that means they *do* have a "starter kit"
prepared on a CD, ready for someone to install and test it out in a bit
more depth than the single-floppy demo. Good for them, I hope to see
them doing more of that.
-Leif B.
--
Leif Bloomquist B.A.Sc. | leifb (at) robominer (dot) com
Research and Development Engineer | Phone: (905) 713-3700 x713
Automated Mining Systems Inc. | Fax: (905) 713-3708
Aurora, Ontario, Canada | http://www.robominer.com
sa...@yahoo.com
"Paul May" <paul...@mcmail.com> wrote:
> >W.Stokes <wst...@hotmail.com> wrote:
>
> >> Is this QNX's marketing startegy? Capture a dead market....
>
> simple answer ,ASK THEM.
>
We are FAR from dead, just under everyone else's RADAR... ;)
Incidentally, they still use Amigas for space shuttle and other launches,
too... (re: story that QNX is used on the shuttle).
<snip>
> >Even if this isn't the case, who knows what niche this company
> >has been surviving in. I think I remember seeing a set of
> >networked computers with the Amiga name on it a few years ago
> >in one of the first multi-user 3D arcade games. The one where
> >you shoot other players on a square 3D board, with 4 raised
> >platforms.
>
> i'm not really a games player so i cant help you on that one.
>
<snip> --
He's referring to Virtuality. The first (and only successful AFAIK) Virtual
Reality arcade game (so far). It was a bunch of networked A3000s (and later,
A4000s, though the company had pretty much died by then)...
> Paul May
> Team
> *AMIGA*
> <tsb> User Of The small OS With A BIG Future
> --
> <tsb> Looking For Disabled persons Ideas
> <tsb> For Better Computer Interaction & Inclusion
> --
>
>
Looking forward to the QNX/AmigaOS partnership. Welcome aboard all
QNXers...‰
Mitchell Schoenbrun
> What's interesting here is, that means they *do* have a "starter kit"
> prepared on a CD, ready for someone to install and test it out in a bit
> more depth than the single-floppy demo. Good for them, I hope to see
> them doing more of that.
Well I don't know what they handed out, but if it is what they talked
about at the show, it is no surprise. They said they would have a
CD with all their software. The licenses are separate however.
They might be including some limited time licenses.
Felix S. Gallo
>I can envisage that anyone wanting to develop and test a user/pc product -
>such as a word processor or similar, would only need a copy of phindows (or
>just telnet if you don't want graphics) and just log into a remote qnx
>server that is equipped with compiler, etc. the limitations here are
>processor speed, memory, but I haven't seen much in the way that qnx
>licenses it's OS that would prohibit this.
A nice idea, but when you develop code for an OS, you also want
to be able to run the code on that OS and meddle with it in person.
This is especially true for developing under RTOSes, where you
may have arcane physical devices paired with device drivers.
>This would enable many developers to share a server for a much reduced
>cost.
Timesharing has its place, but even in those cases, you still need to
buy at least one server copy of QNX with dev tools, networking, and
so forth -- which effectively eliminates most Linux users from contention.
>> Another point that is easy to miss is that things like Linux are not
>> only almost as embeddable as QNX (I once put Linux in 2M ROM
>> and 4M RAM), but Moore's Law is making the bar move up. Pretty
>> soon, it'll be silly not to put 32M of RAM in a cheap embedded
>> device, because they won't even make 4M RAM modules any more.
>> The 'size' advantage of QNX is shrinking rapidly.
>
>No, it means you will be able to develop bigger/better systems with qnx
>than others for the same memory.
Not necessarily. An OS that takes 1.5M and an OS that takes 256K are
remarkably similar-looking if you have 32M of RAM, all else being equal.
If they both have to run an 8M browser, the differences get even smaller.
I'm not saying, note, that QNX is toast -- just that the OS's primary
advantage is naturally decreasing, making its price/performance
value get dangerously closer to its competitors.
>If you allow it to have it's software upgradeable somehow, then you will be
>surprised how many extra features people will want later on down the track.
True if you're designing computers -- much less so if you're designing,
say, Internet appliances, where the underlying OS only needs enough
features to support the infrastructure.
>yes this is a good point, the times where qnx really makes a difference is
>when you are making something that is really making the most use of your
>hardware, so you are doing something with the same hardware cost that you
>physically couldn't achieve with another os.
I sorta disagree. The big win for QNX right now is in those projects that
maximize the use of its novel architecture. There are many operating
systems that will make as much out of hardware as QNX does, at a
lower price point. If those operating systems ever get nicer to program
for, have a usable target shell ( :) ), and/or come natively ROMmable,
even the QNX architectural advantage will vanish against the price/
performance issue.
>if you can't justify the full features that qnx gives with the price they
>want, then you will have to settle for something else I guess. But someone
>else will see/use the real potentials, and maybe they'll be better for it
>in the long run.
>
>For people that say that they can't see the potentials without playing with
>it, I think that is where the demo comes in.
QNX's potential is enormous. It's the accompanying sticker shock that's
causing me heartburn. :)
>regards
>Lance Roberts
Felix
Guy Macon
>I don't believe that QSSL is that interested in small developers,
>unless they have the kind of application that fits their pricing
>strategy. I am fortunate, I work for a company where the software / PC
>hardware represents about 1% of the product cost. Consequently we are
>able to use QNX and enjoy the benefits of QNX because the cost of
>doing so is irrelevant in the scheme of things. There are other
>posters in the thread who appear to be in a similar situation. I
>sympathise with those who cannot join the party for budgetary reasons.
I also work for a company where the software & PC hardware represents
about 1% of the product cost. Alas, the custom programming needed
represents about 60% of the product cost. I believe that this figure
is larger than it should be, and I think I know why; Windows NT.
That's why I am converting to QNX.
John Helliwell
W.Stokes wrote in message
<469652E0471C7DD3.F1CE4E54...@library-proxy.airnews.ne
t>...
>If this wasnt the month of November I would have to say this is an
>April fools joke!
<sigh>
>
>The Amiga platform is dead.
>
It's not dead, it's sleepin'. It's pinin' for the fjords.... <end monty
pyton parody>
To come out with a statement like that is extremely shortsighted. My Amiga
isn't dead, anyway, it runs an extremely effiecient pre-emptively
multitasking OS, and uses a Motorola PowerPC processor.
>To tie their name to an out of date GAME machine is really a smart
>move.
I never use my Amiga for games. I use it for web browsing, email. news etc.
It's much nicer than this win95 box I'm on now, anyway.
John Birch
But surely you should be converting to embedded NT! ;-)
regards John B.
John Birch
wrote:
>snip<
>
>I sorta disagree. The big win for QNX right now is in those projects that
>maximize the use of its novel architecture. There are many operating
>systems that will make as much out of hardware as QNX does, at a
>lower price point. If those operating systems ever get nicer to program
>for, have a usable target shell ( :) ), and/or come natively ROMmable,
>even the QNX architectural advantage will vanish against the price/
>performance issue.
>
I sorta disagree here! The big win for QNX is those developers that
are looking for a well defined, structured and tested OS and who can
afford to pay a premium price for very good support (at least compared
with most other OS's mentioned in this and previous threads).
As regards huge volume applications, QNX is still applicable as I am
sure QSSL will talk big discounts for SERIOUS enquiries.
AS far as a cheaper QNX is concerned, I suspect the new Amiga OS
(based on QNX it is rumored) may go a long way to answering your
concerns. Although I guess you'll probably have to buy an Amiga to run
it on :-(
regards John B.
Andrew Thomas
They were complete, non-expiring licenses. As far as I know, they
included Photon development as well. It was worth A LOT of money.
deja...@my-dejanews.com
> But that's MY point of view. I'll let others have their own. And tomorrow
> we'll have Metrowerks visiting. Just to give us the first look.....(+:
>
> --
> in...@arxi.nospam (remove spam....)
>
>
So, how does one get one's hand on the Code Warrior for QNX demo CD? And
were you impressed? What ballpark is the price in?
thanks
==
Chiradeep
sa...@yahoo.com
The "games machine" statement is getting REALLY stale... I could understand
six or seven years ago when there might have been that perception. But, then,
PC people were not only blowing off the Amiga as a game machine, they were
also saying things like, "Why do I need more than 16 colors?" or, "Why do I
need multitasking? I can only do one thing at a time." No, a majority of
Amiga users that are still around (read: the REAL Amiga users) don't use it
for games as much as for professional uses or net surfing. I personally use
mine for graphics and sound. Games every now and then. The PC is much more of
a game machine now than the Amiga. Besides, that's why I got a Playstation...
;)‰
Mitchell Schoenbrun
> They were complete, non-expiring licenses. As far as I know, they
> included Photon development as well. It was worth A LOT of money.
This could be so, but it would be unlike anything QSSL has ever done.
--
Mitchell Schoenbrun --------- masc...@pobox.com
Tom Sheppard
<f...@newsguy.com> wrote:
...
>>> Another point that is easy to miss is that things like Linux are not
>>> only almost as embeddable as QNX (I once put Linux in 2M ROM
>>> and 4M RAM), but Moore's Law is making the bar move up. Pretty
>>> soon, it'll be silly not to put 32M of RAM in a cheap embedded
>>> device, because they won't even make 4M RAM modules any more.
>>> The 'size' advantage of QNX is shrinking rapidly.
>>
>>No, it means you will be able to develop bigger/better systems with qnx
>>than others for the same memory.
>
>Not necessarily. An OS that takes 1.5M and an OS that takes 256K are
>remarkably similar-looking if you have 32M of RAM, all else being equal.
...
Let's take a simple case where each instruction is a fixed 4 bytes long. A
1.5 MB OS contains approximately 400 k instructions to implement some
units of functionality. A 256 KB OS only has 65 k instructions. If "all
else being equal" and that means functionality too, then the large OS
executes more instructions to accomplish the same work. That means
increased overhead and latency. Not good.
If the larger OS is as efficiently designed and coded as the smaller one,
then it contains more functionality. But is it modular? Why carry around
all that baggage if all you need is a couple hundred KB of functionality.
You don't drive around in your pickup truck with a load full of sand
everywhere you go. Although, you don't take all the seats out of your
minivan and drive around that way either. Hmm, bad analogy. :-)
Felix S. Gallo
Tom Sheppard writes:
>[I, Felix, write:]
>>Not necessarily. An OS that takes 1.5M and an OS that takes 256K are
>>remarkably similar-looking if you have 32M of RAM, all else being equal.
>
>1.5 MB OS contains approximately 400 k instructions to implement some
>units of functionality. A 256 KB OS only has 65 k instructions. If "all
>else being equal" and that means functionality too, then the large OS
>executes more instructions to accomplish the same work. That means
>increased overhead and latency. Not good.
That's not what I meant by "all else being equal," but you're right. The
issue is how important it is that you're right. QNX does indeed have
decreased overhead, but as I point out, that fact is becoming
unimportant; and the argument "but you can save money that way"
doesn't work in any case because the incremental cost of QNX over
those solutions dwarfs the incremental whole-system savings of the
extra RAM (for example) at the moment, in 10k quantity compared
with, say, vxworks. So while it's true in theory that QNX is faster,
the difference that used to exist between 1 sec and .2 sec is less
important now that it's the difference between 10 ns and .2 ns, and
the difference between 512K and 2M is in the noise now that
32M RAM comes free in a box of cereal.
>If the larger OS is as efficiently designed and coded as the smaller one,
>then it contains more functionality. But is it modular? Why carry around
>all that baggage if all you need is a couple hundred KB of functionality.
The field of couple-hundred-KB systems is shrinking very quickly,
at least as a percentage of total system opportunity. The old
necessities were RS232 and some LEDs; the new necessity is
TCP/IP, httpd, and possibly a vm. But I could be saying this just
because that's all I know. :)
Anyway, go QNX!, dammit. Stop teasing me with a superior OS
and dev sys that's out of reach for consumer electronics system
designers, though.
Felix
Chris "Ng" Jones
On Mon, 16 Nov 1998, Igor Kovalenko wrote:
> What is curious, is how QSSL may treat this new OS. It actually may
> become a direct competitor, unless some functionality will be
The company writing AmigaOS5 is Amiga Inc. (www.amiga.com). They are
taking a similar approach to 3Dfx, they will not be selling anything
themselves, but licensing to other companies. The OS is being designed
with a range of applications in mind, from embedded applications, to PDAs,
to low end computers and probably high end workstations.
If the OS is anything like it was before (and it certainly should be, Dr
Allan Havemose (sp?) worked at Commodore in the good old days), you can
expect several things from it. It will be easy to use, but will also offer
great power through shells, it will offer very flexible system wide IPC
through a scripting language (REXX is a likely favorite). It will gobble
up multimedia like nothing else and it will scale well.
From what I have heard, it has the potential to actually be an OS capable
of being all things to all people. Hurrah! (And dont say anything about
NT, that is an OS that is all bugs for all peopl ;-)
> intentionally restricted or Amiga will run only on a totally different
> platform. What about runtime prices? They can't be as high as for QNX
> for Amiga market, but otherwise QSSL will have to drop runtime prices
> too... or it will compete against itself like with QNX Windows & Photon?
Since Amiga Inc are not making machines, the target platforms are not yet
entirely clear, but presumably any platform QNX runs on now, AmigaOS will
run on, with more to be added.
As well as partnering with QNX, Amiga Inc have partnered with an as yet
unamed CPU maker to produce a stunning new CPU like nothing before. Strong
rumours are circulating that this will be TransMeta, in which case, given
their engineering employee list, you can expect something wonderful.
Obviously, many of the low end machines will be running less powerful
processors, but I doubt that the OS will be marketed (at least initially)
to compete with QSSL. Then again, it really depends on which is better for
the job required.
Lots of information can be found at:
http://www.cucug.org/amiga.html
http://www.amiga.org/
http://www.amiga.com
http://www.amiga.de
Expect to hear more in the coming months - the developer version of OS5 is
expected within 6 months.
--
Chris Jones
Black-Sun Software
ch...@black-sun.co.uk
www.black-sun.co.uk
Team *AMiGA*
Chris "Ng" Jones
On Mon, 16 Nov 1998, W.Stokes wrote:
> The Amiga platform is dead.
Error. Try doing a web search for 'Amiga'. You will be surprised by the
number of results you see. Read a few back issues of Wired, several
entries, go to NASA and ask to see the controller machines for the Delta
5, go to a US local cable channel and ask to see their 'Toaster' (try
India too for that matter).
Lack of retail availability doest not a dead machine make.
> Its main market in the end was video and that has long dried up.
Really? People stopped using video? I'd heard that "video killed the radio
star", but I havent heard of anything killing video yet!
> To tie their name to an out of date GAME machine is really a smart
> move.
Another HUGE misconception, the Amiga is not a games machine, that was
just poor marketing by Commodore, the Amiga was engineered with more than
games in mind. After all, how many games machines need pre-emptive
round-robin scheduling. The reason for using QNXs kernel is for reasons of
development time. Im sure Amiga would love to recraft the original m68k
kernel for x86, PPC, MIPS, Alpha and the under-wraps CPU it will be
targeted for, but because of time consraints they have chosen QNX.
> Where can I download the Amiga/QNX version of Marble Madness?
If you want to play it on your loverly Wintel box, you can. Download
WinUAE (Windows Amiga emulator) and then download Marble Madness. Problem
solved.
> I hope they port their DOOM version over too!!!
Well, since the DOOM source is widely available, it wouldnt surprise me.
Or, you could buy Myst, or Quake or (soon anyway) Quake 2 for WinUAE.
Please try to keep your rampant naysaying to yourself in future!
Ciro Scognamiglio
> Hi,
> On Mon, 16 Nov 1998, Igor Kovalenko wrote:
> If the OS is anything like it was before (and it certainly should be, Dr
> Allan Havemose (sp?) worked at Commodore in the good old days), you can
> expect several things from it. It will be easy to use, but will also offer
> great power through shells, it will offer very flexible system wide IPC
> through a scripting language (REXX is a likely favorite). It will gobble
> up multimedia like nothing else and it will scale well.
Actually AmigaOS can do all the things you said (I'm an amiga user BTW :))
The real problem with Classic Amigas (i.e. actual ones) is processor power
and some important features missing from the os (memori protection in primis)
but you must notice that it's several years that the market of this machine
is stuck.
Only third parties are making hw and sw (and of course users, just have a look
to aminet, the biggest amiga archive), Support from the community is strong
(because the amiga community itself is strong).
> From what I have heard, it has the potential to actually be an OS capable
> of being all things to all people. Hurrah! (And dont say anything about
> NT, that is an OS that is all bugs for all peopl ;-)
As I said it already have the capabilities, the problem is that ppl doesn't
know anything about it (and they buy what they see more often around...despite
the quality).
> > intentionally restricted or Amiga will run only on a totally different
> > platform. What about runtime prices? They can't be as high as for QNX
> > for Amiga market, but otherwise QSSL will have to drop runtime prices
> > too... or it will compete against itself like with QNX Windows & Photon?
> Since Amiga Inc are not making machines, the target platforms are not yet
> entirely clear, but presumably any platform QNX runs on now, AmigaOS will
> run on, with more to be added.
Yes AI will not make hw for now but they'll push an ideal hw (i.e. the MMC
(Mega Monster Chip ;)) for theyr os, but 3rd parties are free to choose
theyr own custom hw.
> As well as partnering with QNX, Amiga Inc have partnered with an as yet
> unamed CPU maker to produce a stunning new CPU like nothing before. Strong
> rumours are circulating that this will be TransMeta, in which case, given
> their engineering employee list, you can expect something wonderful.
QNX is an AI partner because their OS is based on some Amiga structure,
I've heard many of QNX people are/were amigans, the actual QNX kernel
is based on AmigaOS ad the OS itself has the same feeling/philosophy.
About TransMeta, they are only romours, no official statement has been
made by AI only because they consider the OS apriority.
> Obviously, many of the low end machines will be running less powerful
> processors, but I doubt that the OS will be marketed (at least initially)
> to compete with QSSL. Then again, it really depends on which is better for
> the job required.
> Expect to hear more in the coming months - the developer version of OS5 is
> expected within 6 months.
Actually the OS5dev had many delays, and maybe the 6 months will be 12 (hope
less
though) but you know, to make things well time is the key :)
Ciro.
--
--
cs55...@dsi.unimi.it | IRC nick: _moray
mo...@galactica.it | ICQ UIN# 16299769
mor...@geocities.com | Amiga user and ATO member
Writing from the computer science department of Milan (DSI)
home page at http://www.geocities.com/SiliconValley/Park/4710
--
Ciro Scognamiglio
480,76482,76485-76486,76488-76490,76493-76495,76497-76499,76501,76504,76506-76509,76511-76514,76517,76519-76520,76522-76525,76527-76531,76533,76535-76539,76541-76542,76544,76546-76551,76556,76558,76560,76563-76565,76567-76568,76573,76575,76577-76581,76583,76585-76588,76592-76595,76599-76601,76606,76608-76609,76612-76613,76615-76619,76621-76625,76627-76630,76632-76633,76637-76638,76640-76643,76646,76649-76650,76652-76653,76655,76657,76659,76661,76663-76665,76667-76668,76670-76672,76676-7668
> If this wasnt the month of November I would have to say this is an
> April fools joke!
> The Amiga platform is dead.
> To tie their name to an out of date GAME machine is really a smart
> move.
> I hope they port their DOOM version over too!!!
The Amiga it's so dead that when the source code of Doom was realised
by ID on the net the "dead" amiga programmers made 3 (yea three) different
conversions of doom in just a couple of days...
Just have a look around on the net and try to spot out what an Amiga is
before talking...
Sorry guys but I hate ppl talking too much (expecially bad) about topics
they completely ignore.
Dave Baerg
Mitchell Schoenbrun wrote in message <72vefd$vi$1...@supernews.com>...
>Andrew Thomas <and...@cogent.ca> wrote:
>
>> They were complete, non-expiring licenses. As far as I know, they
>> included Photon development as well. It was worth A LOT of money.
>
>This could be so, but it would be unlike anything QSSL has ever done.
I would suggest looking at Dr. Dobbs Journal #292 - December 1998 (page 71)
The ad sure seems to show a *FREE* QNX Eval System (not just the disk). I
can only hope that this is the same
[snip]
"complete, non-expiring licenses"
[end snip]
version Andrew was talking about. If it is, this is a step in the right
direction.
Dave Baerg (dba...@ise.bc.ca) remove the x to reply
Software Developer
Mitchell Schoenbrun
> Mitchell Schoenbrun wrote in message <72vefd$vi$1...@supernews.com>...
>>Andrew Thomas <and...@cogent.ca> wrote:
>>
>>> They were complete, non-expiring licenses. As far as I know, they
>>> included Photon development as well. It was worth A LOT of money.
>>
>>This could be so, but it would be unlike anything QSSL has ever done.
First an update. A student at Waterloo wrote to me and indicated
that the licenses were 1 year expriring.
> I would suggest looking at Dr. Dobbs Journal #292 - December 1998 (page 71)
> The ad sure seems to show a *FREE* QNX Eval System (not just the disk). I
> can only hope that this is the same
> [snip]
> "complete, non-expiring licenses"
> [end snip]
> version Andrew was talking about. If it is, this is a step in the right
> direction.
Let's be rational. This is probably their CD with some short term
"EVAL" licenses. If not, I want to know where I can get some.|-).
Cam Mitchner
Mitchell Schoenbrun wrote:
>
> Andrew Thomas <and...@cogent.ca> wrote:
>
> > They were complete, non-expiring licenses. As far as I know, they
> > included Photon development as well. It was worth A LOT of money.
>
> This could be so, but it would be unlike anything QSSL has ever done.
>
> --
> Mitchell Schoenbrun --------- masc...@pobox.com
Yes, but it is very like what Metrowerks did at QNX'98!
Tom Sheppard
<f...@ultranet.com> wrote:
...
>So while it's true in theory that QNX is faster,
>the difference that used to exist between 1 sec and .2 sec is less
>important now that it's the difference between 10 ns and .2 ns, and
>the difference between 512K and 2M is in the noise now that
>32M RAM comes free in a box of cereal.
I thinks it's the old case of functionality expanding to consume all
available memory and CPU cycles.
The most recent project I was involved in had people examining the
instruction set to see if one instruction executed faster than another.
They were down to counting _cycles_ where the difference between 10 ns and
0.2 ns was indeed significant.
The same with memory... there's never enough memory because the sales
folks keep promising functionality to be supplied in existing equipment
which cannot have memory exanded because it's soldered on the board.
Something about component height, vibration susceptibility, that sort of
thing.
Granted, most projects aren't in those dire straights. It just seems that
no matter how fast CPUs and memory gets, or how cheap the components are,
it only seems to last so long and then it's not enough again.
The cost of the license is one of many factors to consider in designing
and building real-time systems. QSSL has chosen a pricing model which they
believe is sustainable and the value for the money is appropriate for some
people and projects.
I'd like to own a Ferrari but it doesn't fit my budget. Still, it's nice
to know that someone out there is making them available should my
situation change. :-)
Mario Charest
>Since Amiga Inc are not making machines, the target platforms are not yet
>entirely clear, but presumably any platform QNX runs on now, AmigaOS will
>run on, with more to be added.
>
was with the hardware.
Shaun C. Murray
Scognamiglio) wrote:
>
>QNX is an AI partner because their OS is based on some Amiga structure,
>I've heard many of QNX people are/were amigans, the actual QNX kernel
>is based on AmigaOS ad the OS itself has the same feeling/philosophy.
No. QNX existed quite a few years before AmigaOS. Whereas it shares a
somewhat philosophical idea, the implementation is quite seperate.
AmigaOS isn't a microkernal for one.
--
Shaun
s...@enterprise.net PGP Key available
Chris "Ng" Jones
Ever since Amiga Inc announced they were only making the OS, there has
been a considerable amount of debate within the Amiga community as to what
makes an Amiga, is it the OS, the h/ware, or both.
My take on it is that originally, hardware specs were poor. They were way
in advance of the IBM compatibles of the time, but still not wonderful, so
the programmers made use of the hardware directly ("banging the hardware")
to maximise its performance by circumventing the OS' APIs for the
hardware. These days, computing power is not quite so
hard to come by and you would never get someone directly banging a VooDoo2
in assembler, you would use Glide, so whereas initially, the Amiga was
what it was because of the OS and the hardware, I am now convinced that it
is going to be the OS alone that makes or doesnt make it an Amiga.