www.oasya.com
Olivier Egger
Oasya SA is happy to announce that Snowie Professional Edition will appear
on the market by June 98. Many people already know Snowie from fibs where it
played many matches a little while ago. The Win95 program contains many
features such as rollouts, analysis of whole matches, import matches played
on the Internet etc. If you want to know more about our product please visit
us at www.oasya.com.
One of the very useful feature of Snowie is the possibility of analyzing
whole matches. If you would like one of your matches be analyzed by Snowie
we offer at the moment the possibility to submit your match played on fibs
(recorded with toggle moves) and we will analyze it for you and either send
the anlaysis back by email or put it on our web-site. For that, just drop by
at the section Backgammon of our web-site.
Regards,
Olivier
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~
Olivier Egger
Oasya SA
Rte d'Yverdon 4
1028 Preverenges
Switzerland
Fibs: oegger
email: eg...@oasya.com
code...@hotmail.com
On Thu, 9 Apr 1998 20:22:05 +0200, "Olivier Egger" <oeg...@lcavsun1.epfl.ch>
wrote:
>Oasya SA is happy to announce that Snowie Professional Edition will appear
>on the market by June 98. Many people already know Snowie from fibs where it
>played many matches a little while ago. The Win95 program contains many
>features such as rollouts, analysis of whole matches, import matches played
>on the Internet etc. If you want to know more about our product please visit
>us at www.oasya.com.
>
>One of the very useful feature of Snowie is the possibility of analyzing
>whole matches. If you would like one of your matches be analyzed by Snowie
>we offer at the moment the possibility to submit your match played on fibs
>(recorded with toggle moves) and we will analyze it for you and either send
>the anlaysis back by email or put it on our web-site. For that, just drop by
>at the section Backgammon of our web-site.
>
>Regards,
>
>Olivier
>
Is this some sort of elaborate hoax ?
I ask for the following reasons:
Firstly, there are people talking about the site before it has been
announced.
Secondly, someone is talking about clearing up a controversy before
one has developed.
Thirdly, if you go to the site mentioned, it really contains very
little information indeed, in fact virtually nothing is added to the
above, and what there is very general.
Fourthly, according to the FIBS reports (mine only go back to 20th
Dec), this new "world class" program has not played a match for over
three months.
Fifthy, according to the fibs report, Snowie has a rating of over
30,000. Not even the most active players average over 2000
experience points a month. This implies that Snowie had been playing
for around 15 months before it "retired".
It only take a couple of weeks to produce a decent front end, and
there is evidence that the game engine behind Snowie has been working
for a long time.
Are we to believe that someone develops a world class BG program, and
makes no attempt to market it for over 18 months.
Why is there no trial version available ?
How about a sample of a match analysis posted to the group ?
Surely it's not a secret what these will look like.
Of course I could just be being paranoid (I'll start thinking the dice
are loaded against me in a minute), and if I am, I appologize, but
perhaps a little more info, particularly such things as price, and why
the product was "announced" two months before it will be available,
despite the technology apparantly being well established.
--
Chuck Evans
-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/ Now offering spam-free web-based newsreading
Patti Beadles
In article <6glpfs$po$1...@nnrp1.dejanews.com>, <code...@hotmail.com> wrote:
>Fourthly, according to the FIBS reports (mine only go back to 20th
>Dec), this new "world class" program has not played a match for over
>three months.
True. But SnowWhite has been around FIBS for a very long time.
>Fifthy, according to the fibs report, Snowie has a rating of over
>30,000. Not even the most active players average over 2000
>experience points a month. This implies that Snowie had been playing
>for around 15 months before it "retired".
Yeah, that sounds about right, judging by my memory.
>It only take a couple of weeks to produce a decent front end, and
>there is evidence that the game engine behind Snowie has been working
>for a long time.
Say WHAT?!?
To develop a simple "play against the engine" front end might only
take a few weeks. But to do a full-featured program that handles
replays, rollouts, etc., is easy to use, and has gone through a good
QA cycle would take several months.
>Are we to believe that someone develops a world class BG program, and
>makes no attempt to market it for over 18 months.
I believe that SnowWhite has gone through a tuning process, just as
the other programs have. TD Gammon was around for years before IBM
decided to make it available.
>How about a sample of a match analysis posted to the group ?
Oh, please. There's one on the website.
>Of course I could just be being paranoid (I'll start thinking the dice
>are loaded against me in a minute), and if I am, I appologize, but
>perhaps a little more info, particularly such things as price, and why
>the product was "announced" two months before it will be available,
>despite the technology apparantly being well established.
I think you're being paranoid.
There are actually several neural nets around that play more or less
at the same level. Jellyfish is one. TD Gammon is another, SnowWhite
is a third. Yabot is certainly in approximately the same ballpark, as
is Motif.
Putting out commercial shrink-wrap software requires a lot of work,
including marketing, production, distribution, and a support
infrastructure. I can see someone developing an engine as an academic
exercise and then not going through the production process for a long
time (if ever).
-Patti
--
Patti Beadles |
pat...@netcom.com/pat...@gammon.com |
http://www.gammon.com/ | If it wasn't for the last minute
or just yell, "Hey, Patti!" | I'd never get anything done!
Kit Woolsey
code...@hotmail.com wrote:
: On Thu, 9 Apr 1998 20:22:05 +0200, "Olivier Egger" <oeg...@lcavsun1.epfl.ch>
: wrote:
: >Oasya SA is happy to announce that Snowie Professional Edition will appear
: >on the market by June 98. Many people already know Snowie from fibs where it
: >played many matches a little while ago. The Win95 program contains many
: >features such as rollouts, analysis of whole matches, import matches played
: >on the Internet etc. If you want to know more about our product please visit
: >us at www.oasya.com.
: >
: >One of the very useful feature of Snowie is the possibility of analyzing
: >whole matches. If you would like one of your matches be analyzed by Snowie
: >we offer at the moment the possibility to submit your match played on fibs
: >(recorded with toggle moves) and we will analyze it for you and either send
: >the anlaysis back by email or put it on our web-site. For that, just drop by
: >at the section Backgammon of our web-site.
: >
: >Regards,
: >
: >Olivier
: >
: Is this some sort of elaborate hoax ?
Definitely not.
: I ask for the following reasons:
: Firstly, there are people talking about the site before it has been
: announced.
: Secondly, someone is talking about clearing up a controversy before
: one has developed.
: Thirdly, if you go to the site mentioned, it really contains very
: little information indeed, in fact virtually nothing is added to the
: above, and what there is very general.
The site does have some examples.
: Fourthly, according to the FIBS reports (mine only go back to 20th
: Dec), this new "world class" program has not played a match for over
: three months.
So what? Probably means that Olivier wasn't able to find a free place to
keep Snowie on line.
: Fifthy, according to the fibs report, Snowie has a rating of over
: 30,000. Not even the most active players average over 2000
: experience points a month. This implies that Snowie had been playing
: for around 15 months before it "retired".
True -- probably longer than that if my memory is correct.
: It only take a couple of weeks to produce a decent front end, and
: there is evidence that the game engine behind Snowie has been working
: for a long time.
Only a couple of weeks? Are you kidding? I'm not a professional
programmer, but from what I know producing a good front end which will
make the program a good usable produce is quite a difficult task.
: Are we to believe that someone develops a world class BG program, and
: makes no attempt to market it for over 18 months.
Of course. When Olivier first started working on Snowie I imagine he had
no intentions of marketing it -- he was doing it for intellectual
satisfaction. As it is, the decision to market it is a questionable one,
considering the limited market and the firmly established competitor
Jellyfish. While I wish them the best of luck, I'm far from certain it
will turn out to be a profitable endeavor.
: Why is there no trial version available ?
Snowie is in the beta test stage right now. I'm expecting to receive a
beta test version any day, and probably beta-testers in Europe have
already received theirs and are working with them.
: How about a sample of a match analysis posted to the group ?
Take another look at the web page. There are several such match analyses
there.
Kit
: Surely it's not a secret what these will look like.
: Of course I could just be being paranoid (I'll start thinking the dice
: are loaded against me in a minute), and if I am, I appologize, but
: perhaps a little more info, particularly such things as price, and why
: the product was "announced" two months before it will be available,
: despite the technology apparantly being well established.
: --
Carl Larson
Kit Woolsey wrote in message ...
>
>Of course. When Olivier first started working on Snowie I imagine he had
>no intentions of marketing it -- he was doing it for intellectual
>satisfaction. As it is, the decision to market it is a questionable one,
>considering the limited market and the firmly established competitor
>Jellyfish. While I wish them the best of luck, I'm far from certain it
>will turn out to be a profitable endeavor.
>
>
It's got me more than interested. I'm a huge Jellyfish fan, but it lacks
a number of obvious features and the creator is slow in adding features
and delivering updates.
I also belive Jellyfish has missed an opportunity by not packaging their
program for a wider, more general market. Mainstream bridge and chess
programs are obviously targeted and sold to the general public at a
profit. Pros tend to have the more specialized programs. Maybe with what
looks to be better graphics and interface Snowie could be sold to a mass
market publisher (although a good tutor would have to be added).
A quality backgammon program marketed to the mainstream public would
certainly help the game by attracting more players.
Carl Larson
Kevin Bastian
code...@hotmail.com wrote in article <6glpfs$po$1...@nnrp1.dejanews.com>...
<snip>
> Not even the most active players average over 2000
> experience points a month.
>
<snip>
> --
> Chuck Evans
>
> -----== Posted via Deja News, The Leader in Internet Discussion ==-----
> http://www.dejanews.com/ Now offering spam-free web-based newsreading
>
Patti's already addressed most of the major issues in your post (which I
agree are off the mark). One minor detail struck me, although it really has
little to do with the theme of your post: your assertion that 2000
experience/month is the upper limit for a bot. That didn't feel AT ALL
correct (heck, I keep my own records and from 2/16-3/15 I played 1060...and
I'm predominantly human), so I checked the latest FIBS rating report, and
the 4 most active players -- all bots -- averaged 3000-5000/month (the
report is for 1/2 month -- 3/16-4/1 -- so I'm multiplying by 2):
*** TOP 25 *** Most Active Players
## Name-------------------- : Experience Change
1 MonteCarlo : 85372 +2515
2 harmony : 27377 +1981
3 yabe : 14068 +1740
4 yabs : 12663 +1609
As you can see, based on MonteCarlo's rate during this period, a bot could
easily get 30,000 experience in 6 months. I suspect a search of past
reports would show that even 5000/month is an underestimate of what a bot
can do (and has done). We've seen some very active bots come and go over
the past few years.
code...@hotmail.com
References: <pattibE...@netcom.com>
On Fri, 10 Apr 1998 20:11:59 GMT, pat...@netcom.com (Patti Beadles) wrote:
Glad to be reassured that it's not a hoax.
It was the odd order of messages that really made me wonder, and I had
to ask, 'cause I was just about to buy the JF tutorial and didn't want
to wait another few months for no reason.
>>It only take a couple of weeks to produce a decent front end, and
>>there is evidence that the game engine behind Snowie has been working
>>for a long time.
>
>Say WHAT?!?
You heard.
>To develop a simple "play against the engine" front end might only
>take a few weeks. But to do a full-featured program that handles
>replays, rollouts, etc., is easy to use, and has gone through a good
>QA cycle would take several months.
Sorry, you're wrong. I know from experience. It only took me four days
to do a play against an engine with full color configuration, gliding
pieces (variable speed), replay (variable speed, both directions),
annotation, logging. Of course, the 'engine' was just "make the first
legal move", but at least I could beat it <g>.
Admittedly QA is another matter.
>>How about a sample of a match analysis posted to the group ?
>
>Oh, please. There's one on the website.
Please what?
The web site is *very* poor.
It contains very little that was not in the announcement on this NG,
and is spread over several pages.
It also seems to be hosted in Europe, so it was very slow loading, and
I must admit that there may have been a page I didn't get to, but I
was becoming frustrated having spent ten minutes and not found out
anything new.
It really is very bad design, for example, to have a link to a page
saying something like "ordering", only to have a long wait before
being told that it's not available yet (which I already knew, I was
hoping for the elusive price/delivery details).
In fact the whole site looked more like a marketing guy's rough idea,
before anyone had worked out how to do it yet, and filled in the
details.
Sorry if this sounds harsh, but I'm in the position of having to wait
maybe three-four months (with delays, waiting for feedback), before
shelling out an unknown quantity of bucks.
I know there's a very strong "take it or leave it" attitude on this
NG, but in the real world, you have to be aware of your customer's
needs and expectations.
>I think you're being paranoid.
But hoaxes do happen <g>, and you're obviously party to a lot more
info on this subject than I am.
Anyway, thanks for the response, I guess I'll wait awhile before
making a purchasing decision. How frustrating!
John Goodwin
>
>>How about a sample of a match analysis posted to the group ?
>
>Oh, please. There's one on the website.
>
How do you get to it? I tried to have a look at this site, and found
it slow, and lacking in any real detail.
I was particularly looking for the match analysis, but could only get
to some rather fuzzy representations of dialogue boxes.
Which link do you follow to get to a sample analysis.
Thankyou for any help.
JG
Patti Beadles
In article <352f1cd1....@news.demon.co.uk>,
John Goodwin <J...@opticon.demon.co.uk> wrote:
>How do you get to it? I tried to have a look at this site, and found
>it slow, and lacking in any real detail.
Click on backgammon, and you'll find matches.
-Patti
--
Patti Beadles | Not just your average
pat...@netcom.com/pat...@gammon.com | degenerate gambling adrenaline
http://www.gammon.com/ | junkie software geek leatherbyke
or just yell, "Hey, Patti!" | nethead biker.
Patti Beadles
In article <6gn62h$re2$1...@nnrp1.dejanews.com>, <code...@hotmail.com> wrote:
>Sorry, you're wrong. I know from experience. It only took me four days
>to do a play against an engine with full color configuration, gliding
>pieces (variable speed), replay (variable speed, both directions),
>annotation, logging. Of course, the 'engine' was just "make the first
>legal move", but at least I could beat it <g>.
>Admittedly QA is another matter.
A commercially viable product? With an installer, documentation, etc?
Full-featured? which performs at an acceptable performance level on
any reasonable platform?
What you're describing to me doesn't sound like a full-featured product
at all, just the bare bones of an interface. And yeah, I can see
something like that being workable in only a week of work.
But I'll certainly agree with you that the Oasya web site leaves a lot
to be desired.
Rodrigo Andrade
> I also belive Jellyfish has missed an opportunity by not packaging their
> program for a wider, more general market. Mainstream bridge and chess
> programs are obviously targeted and sold to the general public at a
> profit. Pros tend to have the more specialized programs. Maybe with what
> looks to be better graphics and interface Snowie could be sold to a mass
> market publisher (although a good tutor would have to be added).
The main difference between chess and backgammon softwares is that chess
softwares have a damn wider market than backgammon does. A high-end chess
software (say Fritz, or Rebel) costs the same price of Jellyfish Tutor. But
you can also find cheaper software (say, $50) that appeal to the masses, like
Power Chess or Chessmaster.
I think it's just not worthy the investment to make a Jellyfish that says
"White doubles" w/ 20 different voices, and uses 10,000 different 2D and 3D
backgammon sets, and starts w/ a beautifully rendered brief introduction w/
Bill Robertie's voice telling the history of backgammon. I mean, that's all
just crap, and I bet serious players don't give a damn about hearing the dice
rolling different in every roll, playing on a 3D marble board w/ Egyptian
themes on the background...
Rodrigo
Rodrigo Andrade
Oh, and by the way, chess is the 2nd most popular game on the planet (after
soccer... yuk!!!!), and the FIDE (French abbreviation for Int'l Federation of
Chess), is the 3rd biggest sport federation on the planet (after the olympic
commitee, and FIFA, the soccer federation).
Rodrigo
FERRANTE
On Sat, 11 Apr 1998 22:52:57 GMT, pat...@netcom.com (Patti Beadles)
wrote:
>In article <6gn62h$re2$1...@nnrp1.dejanews.com>, <code...@hotmail.com> wrote:
>>Sorry, you're wrong. I know from experience. It only took me four days
>>to do a play against an engine with full color configuration, gliding
>>pieces (variable speed), replay (variable speed, both directions),
>>annotation, logging. Of course, the 'engine' was just "make the first
>>legal move", but at least I could beat it <g>.
>
>>Admittedly QA is another matter.
>
>A commercially viable product? With an installer, documentation, etc?
>Full-featured? which performs at an acceptable performance level on
>any reasonable platform?
>
>What you're describing to me doesn't sound like a full-featured product
>at all, just the bare bones of an interface. And yeah, I can see
>something like that being workable in only a week of work.
>
>
>But I'll certainly agree with you that the Oasya web site leaves a lot
>to be desired.
And the product may also leave a lot to be desired. I certainly hope
that if this product is indeed as good as they want people to believe,
that a demo will be available. People can "try before they buy" and
thereby eliminate the possibility of wasting good money.
Mark Ferrante
>
>-Patti
"Man is neither beast nor angel and as misfortune would have it, he who wishes to become the angel, becomes the beast..."--Pascal
Julian Haley
Regardless whether it is good or not, someone is bound to start whining
about fixed dice when they lose to it a couple of times<grin>
Claes Thornberg
code...@hotmail.com writes:
>
> Are we to believe that someone develops a world class BG program, and
> makes no attempt to market it for over 18 months.
>
Yup, mloner/loner has played on the net for more than two years if my
memory serves me right. And that program, as far as playing goes, is
one of the finest, if not the finest bg-program available, IMHO. And
you can't purchase it, you can only play it on gamesgrid as far as i
know.
Regards,
Claes Thornberg
--
______________________________________________________________________
Claes Thornberg Internet: cla...@it.kth.se
Dept. of Teleinformatics URL: NO WAY!
KTH/Electrum 204 Voice: +46 8 752 1377
164 40 Kista Fax: +46 8 751 1793
Sweden