Miami or AmiTCP - Which is better?
Christopher P. Mills
Which is better - Miami or AmiTCP?
I am a Network Administrator so no worries about Internet Jargon and
setting it up and stuff, Just wondered if anyone has used both and have
any preferences. For me, speed and memory overheads are more important
than ease of setting up.
Michael M. Rye
On Thu, 02 Jan 1997 16:07:58 +0000 "Christopher P. Mills"
<ch...@gw.rmcs.cranfield.ac.uk> wrote:
>Which is better - Miami or AmiTCP?
Easy - Miami. It's much easier to set up and use. I had it running
within 5 minutes after installing it. If you know anything about
setting up UNIX TCP, AmiTCP is at the same difficulty level. It's
hard to keep track of all those text files that it uses for
configuration. Miami, on the other hand, uses MUI for its GUI, and
all the configuration is handled through that GUI. Since Miami uses
MUI, it uses more memory while you have the GUI open. But, if you
set up Miami to iconify the GUI when you go online, you can then
flush the memory and most or all of the MUI libraries will get out
of memory and you'll get a lot of it back.
[snip]
--
*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*
* -Michael M. Rye- (Jedi) | Amiga 1200, 68030/MMU 50 MHz, 68882 50 MHz, *
* UNIX/C Admin/Design |____ 10 MB RAM, 2.0 Gig HD, SupraFAX 33.6 Kbps ____*
* <mr...@topcity.mn.org> | Amiga 500, AdRAM 540 - 1 Chip/2 Fast, 2 floppies *
*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*
* Thought of the Day: *
* I'm your father, Luke! Give in to the Dark Side, you knob!!! *
*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*
Kenton A. Groombridge
In article <32CBDD...@gw.rmcs.cranfield.ac.uk> "Christopher P. Mills" <ch...@gw.rmcs.cranfield.ac.uk> writes:
> Which is better - Miami or AmiTCP?
>
> I am a Network Administrator so no worries about Internet Jargon and
> setting it up and stuff, Just wondered if anyone has used both and have
> any preferences. For me, speed and memory overheads are more important
> than ease of setting up.
AmiTCP is by far the winner. It doesn't use that nasty Magic User Interface
(MUI), and it is super powerful. Miami can't handle the login script
necessary to connect to my Internet Service Provider (ISP). I have a dozen
phone numbers each with their own IP address, three name servers, and a bunch
of other stuff.
I am not a genious, and I set up AmiTCP 4.0 demo in about 10 minutes, now
I am a registered 4.3 user and very happy with it. Super powerful ARexx
dial scripts. And it doesn't use that NASTY MUI.
Kenton Groombridge
kgro...@hereford.ampr.org
http://www.cec.army.mil/~kgroombr/
A4000/40, registered AmiTCP 4.3, register PPP 1.45, registered AWeb 2.1,
and MUI FREE!
Anbjorn Myren
>
>I am a Network Administrator so no worries about Internet Jargon and
>setting it up and stuff, Just wondered if anyone has used both and have
>any preferences. For me, speed and memory overheads are more important
>than ease of setting up.
For speed, it seems to be the same if you are using AmiTCP or Miami, when it comes
to the data-tranfer speed.
Using MUI might slow down your systen a bit, at least if you have a slow processor.
In my case I've tried AmiTCP-4.0demo and Miami1.0 demo.
The only place I've noticed difference in speed is when I log in.
AmiTCP and PPP1_45 (registered) dials and connects at once and brings up the AmiTCP
demo window ( reminding you about that the full version costs a fortune ), and leave
the window open for about 10 seconds.
Miami seems to use more time connect. It looks like it's doing some
self-configuration or something ( looking for dns entries or something )
I haven't used Miami much, because I've already got AmiTCP set up and working fine
before Miami was available. And since the demo times out after 1 hour use, I did'nt
bother much with it.
But I did register Holger Kruse's PPP device, which speeds up my transfer a lot.
( docs says 40% faster than the shareware version )
For memory usage, I don't know how much memory Miami it self uses, but in order
to use Miami, you must have MUI installed on your system. So I guess that
when you are using MUI + Miami, this will use more memory than AmiTCP will.
For price matters, Miami is cheaper than AmiTCP if you want the full versions.
But AmiTCP4.0demo is fully working for me, and the 10 secs. delay doesnt bother
me much.
For ease of use, Miami is a dream to set up, unlike AmiTCP.
If Miami had been available before I started to set up AmiTCP, I would
have benn using Miami today.
( I used a lot of hours to set up AmiTCP, and downloaded a lots of "Help to set
up Amiga to internet" files from several BBS's before I found what I needed )
Hope this gives you some answers.
But you don't need to take my word for any of this. All you need for testing
these programs are available on Aminet.
A tip though; if you set up your system to use Miami or AmiTCP + PPP1_45 ( or
AmiPPP) and you don't get in touch with your modem, try using a serial.device
replacement. The only devices that make Miami or PPP to allow me to access my modem
is BaudBandit.device or the New8N1.device
If I use serial.device, artser.device or 8N1.device I only get messages like "modem
not responding"
serial.device works fine with Term or NComm.
BTW, there is also TermiteTCP, which I haven't tried. But it is said to be easy
to set up. Don't think it's using MUI.
--
Anbjorn Myren, Norway, A4000/040, IRC: Anmy
Paul Copsey
On Thu, 02 Jan 1997 16:07:58 +0000, Christopher P. Mills <ch...@gw.rmcs.cranfield.ac.uk> wrote:
> Which is better - Miami or AmiTCP?
>
> I am a Network Administrator so no worries about Internet Jargon and
> setting it up and stuff, Just wondered if anyone has used both and have
> any preferences. For me, speed and memory overheads are more important
> than ease of setting up.
In your case, AmiTCP
Paul
Henrik Dissing
On 02-Jan-97 21:30:50, Michael M. Rye wrote:
> On Thu, 02 Jan 1997 16:07:58 +0000 "Christopher P. Mills"
> <ch...@gw.rmcs.cranfield.ac.uk> wrote:
>>Which is better - Miami or AmiTCP?
> Easy - Miami. It's much easier to set up and use. I had it running
> within 5 minutes after installing it. If you know anything about
> setting up UNIX TCP, AmiTCP is at the same difficulty level. It's
> hard to keep track of all those text files that it uses for
> configuration. Miami, on the other hand, uses MUI for its GUI, and
> all the configuration is handled through that GUI.
You clearly didn't read the original article carefully enough:
-- cut --
I am a Network Administrator so no worries about Internet Jargon and
setting it up and stuff, Just wondered if anyone has used both and have
any preferences. For me, speed and memory overheads are more important
than ease of setting up.
-- cut --
> Since Miami uses
> MUI, it uses more memory while you have the GUI open. But, if you
> set up Miami to iconify the GUI when you go online, you can then
> flush the memory and most or all of the MUI libraries will get out
> of memory and you'll get a lot of it back.
Bla, bla, bla :-)
The fact is that Miami has two strong points in the comparison with
AmiTCP: Easier installation and lower price.
As far as speed is concerned they're pretty much the same, but when it
comes to memory overhead my private tests showed that AmiTCP +
ppp.device take less than Miami in the standard configuration, even if
MUI is loaded in both set-ups. Or put in another way: AmiTCP +
ppp.device + IBrowse take up considerably less memory than Miami +
IBrowse.
--
Henrik Dissing E-mail: hen...@post4.tele.dk
Tranbjerg, Denmark FidoNet: 2:238/24.6
Michael B. Smith
In article <32CBDD...@gw.rmcs.cranfield.ac.uk> "Christopher P. Mills" <ch...@gw.rmcs.cranfield.ac.uk> writes:
>
> setting it up and stuff, Just wondered if anyone has used both and have
> any preferences. For me, speed and memory overheads are more important
> than ease of setting up.
Then go with I-Net 225. With Quick_Prefs (our setup tool) it's also a
snap to install.
--
// Michael B. Smith
\X/ m...@adastra.cvl.va.us
Tobias Geijersson
Christopher P. Mills <ch...@gw.rmcs.cranfield.ac.uk> wrote at 02-Jan-97
17:07:58 about 'Miami or AmiTCP - Which is better?':
>Which is better - Miami or AmiTCP?
>I am a Network Administrator so no worries about Internet Jargon and
>setting it up and stuff, Just wondered if anyone has used both and have
>any preferences. For me, speed and memory overheads are more important
>than ease of setting up.
I think AmiTCP suits you best (but I have only tried the demoversion of
Miami)!
If you have some experience with Unix you'll probably have no problem at all
configure it.
The transfer speed are the same with both AmiTCP and Miami.
Miami uses MUI and if you don't use any other MUI application at the same time
(like IBrowse) it will use more mem than AmiTCP (which has no gui when
connecting).
Everyone says that Miami is easy to configure but I found AmiTCP easier when I
tried to add deamons, gateway and stuff (gave up trying with Miami).
btw. AmiTCP 4.2 is easier to setup than the 4.0 demo version.
--
__
__ /// Tobias Geijersson \ mailto:tge...@tufvan.hv.se or use
\\X// Mathematic & Computer science \ mailto:tobias.g...@swipnet.se
\X/ Växjö University College, Sweden \ http://www.tufvan.hv.se/~tgedi94
Johnathan Gapen
On 03 Jan 97 21:51:25 +0100, Tobias Geijersson <tobias.g...@swipnet.se> wrote:
>
>Everyone says that Miami is easy to configure but I found AmiTCP easier when I
>tried to add deamons, gateway and stuff (gave up trying with Miami).
How hard can it be? You go to the InetD entry on the Database page, and
enter the same info you'd enter into the AmiTCP:db/inetd.conf file.
Sometimes, you have to enter a line that would have gone into
AmiTCP:db/services in Miami's services list, also on the Database page.
That's it, you're done, and you don't even have to re-start Miami.
--
Jonathan Gapen <innu...@execpc.com>
Some lead. Others follow. I'm lost.
Jeffrey W Grzanich
In article <mbs....@adastra.cvl.va.us>,
Michael B. Smith <m...@adastra.cvl.va.us> wrote:
>In article <32CBDD...@gw.rmcs.cranfield.ac.uk>
"Christopher P. Mills" <ch...@gw.rmcs.cranfield.ac.uk> writes:
>>
>> I am a Network Administrator so no worries about Internet Jargon and
>> setting it up and stuff, Just wondered if anyone has used both and have
>> any preferences. For me, speed and memory overheads are more important
>> than ease of setting up.
>
>snap to install.
>--
I-Net costs more, but does it offer performance gains as well? Anybody
done a complete comparision of the two products side by side?
> // Michael B. Smith
>\X/ m...@adastra.cvl.va.us
--
************************************************************
*** grza...@sover.net (Jeffrey W. Grzanich) ***
************************************************************
Tobias Geijersson
Johnathan Gapen <innu...@earth.execpc.com> wrote at 04-Jan-97 03:45:33 about
'Re: Miami or AmiTCP - Which is better?':
>On 03 Jan 97 21:51:25 +0100, Tobias Geijersson <tobias.g...@swipnet.se>
>wrote:
>>
>>Everyone says that Miami is easy to configure but I found AmiTCP easier when
>>I tried to add deamons, gateway and stuff (gave up trying with Miami).
> How hard can it be? You go to the InetD entry on the Database page, and
>enter the same info you'd enter into the AmiTCP:db/inetd.conf file.
>Sometimes, you have to enter a line that would have gone into
>AmiTCP:db/services in Miami's services list, also on the Database page.
>That's it, you're done, and you don't even have to re-start Miami.
I (blush) know that know, I played around with MIami last night and
found that out, it's as simple as in AmiTCP.
I'm sorry to have spread wrong information about Miami!
Paul Copsey
On 04 Jan 1997 02:45:33, Johnathan Gapen <innu...@earth.execpc.com> wrote:
> On 03 Jan 97 21:51:25 +0100, Tobias Geijersson <tobias.g...@swipnet.se> wrote:
> >
> >Everyone says that Miami is easy to configure but I found AmiTCP easier when I
> >tried to add deamons, gateway and stuff (gave up trying with Miami).
>
> How hard can it be? You go to the InetD entry on the Database page, and
> enter the same info you'd enter into the AmiTCP:db/inetd.conf file.
> Sometimes, you have to enter a line that would have gone into
> AmiTCP:db/services in Miami's services list, also on the Database page.
> That's it, you're done, and you don't even have to re-start Miami.
I have to agree with Tobias, I'm pretty comfortable setting up and
reconfiguring AmiTCP, but I found Miami trickier and slower to
configure because I had to open the GUI find the right page, and then
type the info in. I guess given time I'd get more used to Miami, and
be able to do it faster, but I doubt I'd ever be able to do it as fast
as I can with AmiTCP
Paul
busse
Henrik Dissing wrote:
>
> On 02-Jan-97 21:30:50, Michael M. Rye wrote:
>
> > <ch...@gw.rmcs.cranfield.ac.uk> wrote:
>
> >>Which is better - Miami or AmiTCP?
>
> > within 5 minutes after installing it. If you know anything about
> > setting up UNIX TCP, AmiTCP is at the same difficulty level. It's
> > hard to keep track of all those text files that it uses for
> > configuration. Miami, on the other hand, uses MUI for its GUI, and
> > all the configuration is handled through that GUI.
>
> You clearly didn't read the original article carefully enough:
>
> -- cut --
> setting it up and stuff, Just wondered if anyone has used both and have
> any preferences. For me, speed and memory overheads are more important
> than ease of setting up.
>
> > Since Miami uses
> > MUI, it uses more memory while you have the GUI open. But, if you
> > set up Miami to iconify the GUI when you go online, you can then
> > flush the memory and most or all of the MUI libraries will get out
> > of memory and you'll get a lot of it back.
>
> Bla, bla, bla :-)
>
> The fact is that Miami has two strong points in the comparison with
> AmiTCP: Easier installation and lower price.
>
> As far as speed is concerned they're pretty much the same, but when it
> comes to memory overhead my private tests showed that AmiTCP +
> ppp.device take less than Miami in the standard configuration, even if
> MUI is loaded in both set-ups. Or put in another way: AmiTCP +
> ppp.device + IBrowse take up considerably less memory than Miami +
> IBrowse.
>
> --
> Henrik Dissing E-mail: hen...@post4.tele.dk
> Tranbjerg, Denmark FidoNet: 2:238/24.6
server? For instance like Netscape on the IBM? I'm new at setup for
Amiga and trying to find what is needed to connect, first of all.
Dolores
J. YURICK
: On 03 Jan 97 21:51:25 +0100, Tobias Geijersson <tobias.g...@swipnet.se> wrote:
: >
: >Everyone says that Miami is easy to configure but I found AmiTCP easier when I
: >tried to add deamons, gateway and stuff (gave up trying with Miami).
MIAMI is much easier than AmiTCP. Hands down!
--
Michael B. Smith
In article <5akq7p$d...@granite.sover.net> grza...@granite.sover.net (Jeffrey W Grzanich) writes:
> In article <mbs....@adastra.cvl.va.us>,
> Michael B. Smith <m...@adastra.cvl.va.us> wrote:
> >Then go with I-Net 225. With Quick_Prefs (our setup tool) it's also a
> >snap to install.
> >--
> Can someone answer what I-Net 225 has over AmiTCP 4.3 Pro? I know
> I-Net costs more, but does it offer performance gains as well? Anybody
> done a complete comparision of the two products side by side?
A complete list of everything included in I-Net 225 is available
at http://www.iworks.com/ (well, the names of the clients and
servers).
--
Tobias Geijersson
busse <bu...@athenet.net> wrote at 05-Jan-97 02:57:12 about 'Re: Miami or
AmiTCP - Which is better?':
>server? For instance like Netscape on the IBM? I'm new at setup for
>Amiga and trying to find what is needed to connect, first of all.
>Dolores
Miami and AmiTCP are TCP/IP stacks, you need a TCP/IP stack to connect your
computer to the Internet.
IBrowse (AWeb, Voyager, ALynx, AMosaic...) is a browser (like Netscape on
other platforms) and requires a TCP/IP stack to be able to browse on Internet.
For more information try the faqs that are regulary posted to this group.
Tobias Geijersson
Paul Copsey <pa...@white-star.com> wrote at 04-Jan-97 23:00:45 about 'Re: Miami
or AmiTCP - Which is better?':
>> <tobias.g...@swipnet.se> wrote:
>> >
>> >Everyone says that Miami is easy to configure but I found AmiTCP easier
>> >when I tried to add deamons, gateway and stuff (gave up trying with
>> >Miami).
>>
>> enter the same info you'd enter into the AmiTCP:db/inetd.conf file.
>> Sometimes, you have to enter a line that would have gone into
>> AmiTCP:db/services in Miami's services list, also on the Database page.
>> That's it, you're done, and you don't even have to re-start Miami.
>I have to agree with Tobias, I'm pretty comfortable setting up and
>reconfiguring AmiTCP, but I found Miami trickier and slower to
>configure because I had to open the GUI find the right page, and then
>type the info in. I guess given time I'd get more used to Miami, and
>be able to do it faster, but I doubt I'd ever be able to do it as fast
>as I can with AmiTCP
>Paul
And if you have done anything with Linux (or any Unix) you'll (IMHO) like
AmiTCP over Miami even more.
Adam Szymczak
On 04-Jan-97 17:00:45, Paul Copsey (pa...@white-star.com) wrote:
>On 04 Jan 1997 02:45:33, Johnathan Gapen <innu...@earth.execpc.com> wrote:
>> On 03 Jan 97 21:51:25 +0100, Tobias Geijersson
>> <tobias.g...@swipnet.se> wrote:
>> >
>> >Everyone says that Miami is easy to configure but I found AmiTCP easier
>> >when I tried to add deamons, gateway and stuff (gave up trying with
>> >Miami).
>>
>> How hard can it be? You go to the InetD entry on the Database page, and
>> enter the same info you'd enter into the AmiTCP:db/inetd.conf file.
>> Sometimes, you have to enter a line that would have gone into
>> AmiTCP:db/services in Miami's services list, also on the Database page.
>> That's it, you're done, and you don't even have to re-start Miami.
>I have to agree with Tobias, I'm pretty comfortable setting up and
>reconfiguring AmiTCP, but I found Miami trickier and slower to
>configure because I had to open the GUI find the right page, and then
>type the info in. I guess given time I'd get more used to Miami, and
>be able to do it faster, but I doubt I'd ever be able to do it as fast
>as I can with AmiTCP
Miami trickier and slower to configure? Let's see what you have to do with
AmiTCP: Run an editor, find the file and load it in, make your changes,
save the file, quit the editor. Miami: Run Miami/Open the GUI, select
Database, select the appropriate db file, make your changes, save settings,
quit Miami/close GUI. Not much different in the two processes.
Besides, you can import AmiTCP:db files into Miami. Everything is copied
accept passwords (see the Miami docs for the reasons why).
Adam Szymczak
Email: szy...@server.uwindsor.ca
WWW: http://supernova.uwindsor.ca/~szymcza
--
We shall shortly be landing. Please return your stewardess to the upright position.
nepaG nahtanoJ
On 06 Jan 97 00:44:21 +0100, Tobias Geijersson <tobias.g...@swipnet.se> wrote:
>
>And if you have done anything with Linux (or any Unix) you'll (IMHO) like
>AmiTCP over Miami even more.
Heh. I've installed Linux on a number of machines, installed Solaris on
a SPARCstation and act as a sorta-administrator on it. Even so, AmiTCP hit
the bit-bucket with as much speed as an old '030 can muster as I replaced it
with Miami. Bottom line: AmigaOS is not UNIX.
Greg Tallent
Way back on 04-Jan-97 14:49:37 J. YURICK said: Re: Miami or AmiTCP - Which is
better?
>: On 03 Jan 97 21:51:25 +0100, Tobias Geijersson
>: <tobias.g...@swipnet.se> wrote:
>: >
>: >Everyone says that Miami is easy to configure but I found AmiTCP easier
>: >when I tried to add deamons, gateway and stuff (gave up trying with
>: >Miami).
>MIAMI is much easier than AmiTCP. Hands down!
And TermiteTCP is even easier!!
>Joe Yurick
>jyu...@tdrss.wsc.nasa.gov
>jyu...@dante.nmsu.edu
<TSB>
Greg Tallent |Amiga2000/040/33mhz/3.1 32 megs ram,2 gig/VideoToaster3.1|
gr...@mcn.org |Amiga 500/020/14mhz/3.1 8 megs ram,2.5 gig/SupraFax 28.8|
In Northern Calif. |Termite Tcp,IBrowse,AWeb II,Miami 1.1,DirWork 2.1,ImageFX|
Living on 750 Acres|AdPro,DPaint V,Brilliance: P133/64 Megs ram,9 gig hd's |
Fun,FUN,FUN!! NOT! |LightWave: 486/120/32 Megs ram 2 gig hd's/LightWave |
SKIER: Someone who pays an arm and a leg to break them.
Paul Copsey
On 06 Jan 1997 16:37:31, Adam Szymczak <szy...@server.uwindsor.ca> wrote:
> On 04-Jan-97 17:00:45, Paul Copsey (pa...@white-star.com) wrote:
> >I have to agree with Tobias, I'm pretty comfortable setting up and
> >reconfiguring AmiTCP, but I found Miami trickier and slower to
[snip]
> Miami trickier and slower to configure? Let's see what you have to do with
> AmiTCP: Run an editor, find the file and load it in, make your changes,
> save the file, quit the editor. Miami: Run Miami/Open the GUI, select
> Database, select the appropriate db file, make your changes, save settings,
> quit Miami/close GUI. Not much different in the two processes.
I agree, but a lot of people, me included, can type faster than they
can work a gui, plus I don't have to run the editor, then edit the
file, I usually load it in one go, ed et[tab]*.conf[ret] pulls in my
conf files in one go or another [tab] (I use Kingcon) lets me select
from a short menu
> Besides, you can import AmiTCP:db files into Miami. Everything is copied
Maybe, but adding and changing is what I do more of rather than the
wholesale installation.
I've got Miami setup fine, just don't really like or feel comfortable
using it (plus my ISP seems to go tits up everytime I try Miami :-( )
Paul
Paul Copsey
On 05 Jan 1997 23:44:21, Tobias Geijersson <tobias.g...@swipnet.se> wrote:
> Paul Copsey <pa...@white-star.com> wrote at 04-Jan-97 23:00:45 about 'Re: Miami
> >be able to do it faster, but I doubt I'd ever be able to do it as fast
> >as I can with AmiTCP
> And if you have done anything with Linux (or any Unix) you'll (IMHO) like
> AmiTCP over Miami even more.
Where do you think I had my first real play with a TCP setup? ;-)
Paul
Tobias Geijersson
Adam Szymczak <szy...@server.uwindsor.ca> wrote at 06-Jan-97 17:37:31 about
'Re: Miami or AmiTCP - Which is better?':
>>> <tobias.g...@swipnet.se> wrote:
>>> >
>>> >Everyone says that Miami is easy to configure but I found AmiTCP easier
>>> >when I tried to add deamons, gateway and stuff (gave up trying with
>>> >Miami).
>>>
>>> and
>>> enter the same info you'd enter into the AmiTCP:db/inetd.conf file.
>>> Sometimes, you have to enter a line that would have gone into
>>> AmiTCP:db/services in Miami's services list, also on the Database page.
>>> That's it, you're done, and you don't even have to re-start Miami.
>>I have to agree with Tobias, I'm pretty comfortable setting up and
>>reconfiguring AmiTCP, but I found Miami trickier and slower to
>>configure because I had to open the GUI find the right page, and then
>>type the info in. I guess given time I'd get more used to Miami, and
>>be able to do it faster, but I doubt I'd ever be able to do it as fast
>>as I can with AmiTCP
>Miami trickier and slower to configure? Let's see what you have to do with
>AmiTCP: Run an editor, find the file and load it in, make your changes,
>save the file, quit the editor. Miami: Run Miami/Open the GUI, select
>Database, select the appropriate db file, make your changes, save settings,
>quit Miami/close GUI. Not much different in the two processes.
The Editor is almost always running, the files in db/ are named the same as
/etc in Linux (easy to remember their name and function), to save just
RAmiga+W, leave the editor running (as usally).
I like pure textfiles as configfiles (easy to cut/copy/paste and I can use
my favorite editor as configurator).
I like the similaries with Linux (easy to make the same changes on both
systems, cut/copy/paste)
I like the developer files with AmiTCP, where do I find the Miami developer
package???
>Besides, you can import AmiTCP:db files into Miami. Everything is copied
>accept passwords (see the Miami docs for the reasons why).
Cool, I can make the changes in my favorite aditor.
>Adam Szymczak
>--
>We shall shortly be landing. Please return your stewardess to the upright
>position.
AmiTCP or Miami, it depends on who you ask and because there is no differ in
data transfer speed I will stay with AmiTCP.
Joseph Waldvogel
>Way back on 04-Jan-97 14:49:37 J. YURICK said: Re: Miami or AmiTCP - Which is
>better?
>>: On 03 Jan 97 21:51:25 +0100, Tobias Geijersson
>>: <tobias.g...@swipnet.se> wrote:
>>: >
>>: >Everyone says that Miami is easy to configure but I found AmiTCP easier
>>: >when I tried to add deamons, gateway and stuff (gave up trying with
>>: >Miami).
>>MIAMI is much easier than AmiTCP. Hands down!
> And TermiteTCP is even easier!!
I've tried TermiteTCP and and Miami is easier to setup and faster then
TermiteTCP is, and has a great many more advanced features the
TermiteTCP is lacking. Miami also has a far better GUI then
TermiteTCP. Again easier to use and setup, Configure.
Joeayo
Yes but Miami will run virtually everything made for Ami-TCP and you set
your servs, DB files, passwords, groups, inetd all in a GUI window so one
simple preloaded config file holds the same thing as Ami-TCP's 500
seperate files, in general operation, it can handle all the servers and
not hit the hard drive over and over for config files, the performance is
excellent, feel almost like your amitcp is running out of a ram disk by
comparison, no need to look for hours thru hundreds of config files,
everything is so simple sitting right in front of you that setting up a
ftp server or web server is dramatically simplified and hours are taken
off of the configing process with much fewer bugs (so simple there is
actually less chance for mistakes!!) Also my 8 meg of ram computer has no
problem running Miami, 4 telnet nodes, 4 dailup lines, FTP, IRC interface
and it still has about 5.5 megs ram free with it all active!
Joeayo
AmiTCP is a total nightmare. I tried for 3 years to get amitcp to work
using ami_ppp and everything I could think of, using advise from experts,
I will never use it or even consider it again. THen along came Miami, it
worked on the very first try, then web browsers worked, FTP, IRC, I simply
couldnt believe it and what was so remarkable is it worked so well,
Actually by comparison, Windows 95 sitting next to my Amiga, I have socket
problems, cant get netscape to work, cant get mirc to work, etc etc (I
think I stumpled onto a win95 bug or something) and yet now everything
just works and works so well on the Amiga. Within 4 hours after trying
Miami I actually had a working FTP daemon, WWW server, telnet, CLChat IRC
server clone all working perfectly, I am still done over at what an
excellent, well thought out program miami is and Hodger is a genious
programmer. Anytime I e-mailed a question to him, he knew exactly what it
was and responded overnight, the guy is constantly updating and perfecting
it, closely talking with registered users in the newsgroup, you feel
almost like he is around the corner ready to personally fix your problems,
and now miami supports the next level of T/TCP and multicasts which
Ami-TCP hasnt thought of and is pushing the limit of being the most up to
date of all TCP stacks for any system. Miami is probably the most
satisfying and rewarding piece of software I have ever registered or
purchased on both my 200 mhz pentium or Amiga combined!
Joeayo
Termite is very simple and easy however if you ran a bbs or wanted
something like your own ftp site (while you are online hosted right there
in your computer) then termite is totally inadequate. Termite is like
using America Online, its a client only TCP. Miami is very easy, clean
but has the power to compete with AmiTCP and now the newer version
supports features that AmiTCP doesnt, like the new T/TCP which can double
speed of web serving, multicasting which allows you to run such streaming
internet things like these web pages where you can hear the radio live
streaming thru the IRC, CUseeme serving, etc, its totally pushing the
envelop with the top Unix routers and at the rate the author is improving
the miami beta test versions, it will be a complete 1 GUI router,
replacing the need for a "unix box".
Greg Tallent
Way back on 07-Jan-97 19:26:35 Joseph Waldvogel said: Re: Miami or AmiTCP -
Which is better?
>>Way back on 04-Jan-97 14:49:37 J. YURICK said: Re: Miami or AmiTCP - Which
>>is better?
>>>: On 03 Jan 97 21:51:25 +0100, Tobias Geijersson
>>>: <tobias.g...@swipnet.se> wrote:
>>>MIAMI is much easier than AmiTCP. Hands down!
>> And TermiteTCP is even easier!!
>I've tried TermiteTCP and Miami is easier to setup and faster then
>TermiteTCP is, and has a great many more advanced features the
>TermiteTCP is lacking. Miami also has a far better GUI then
>TermiteTCP. Again easier to use and setup, Configure.
Well for me, Termite is way faster than Miami. Everybody has different
Amiga setups, But on my system(500/020) Miami slows the system down to the
point that the mouse starts to jerk around when set to 38,400. Termite, On
the other hand, Is set to 115,200 and STILL doesn't slow down the system like
Miami does. For some reason, If I use Miami with Thor, Getting E-Mail can
take me about 20 minutes to download 12 messages, With Termite it takes me
less than 1 minute. I've narrowed this problem down to Miami and can't seem
to find a problem with the config.
I have never had a problem running Termite with any Web Browser that doesn't
need AmiTCP. If the program I'm running needs AmiTCP then I'll run it with
Miami instead, But my preference is still Termite.
Also it took me about 2 minutes to configure Termite and I've never had a
problem since then. Yes, Termite is lacking advanced features. But like I
say, I haven't noticed any problems. If Miami wasn't such a CPU pig, I'd use
it instead of TermiteTCP.
<TSB>
Greg Tallent |Amiga2000/040/33mhz/3.1 32 megs ram,2 gig/Toaster3.1 |
gr...@mcn.org |Amiga 500/020/14mhz/3.1 8 megs ram,2.5 gig/SupraFax28.8|
In Northern Calif. |TermiteTCP,IBrowse,AWeb II,Miami 1.1,DirWork 2.1,ImageFX|
Living on 750 Acres|P133/64 Megs ram,9 gig hd's, Lightwave, WinNT 4.0 |
Fun,FUN,FUN!! NOT! |486/120/32 Megs ram 2 gig hd's, LightWave, Win95 |
POLITICS: The entertainment branch of industry.
J. YURICK
First off, I haven't used Termite, so I'm not putting it down in anyway...
Greg Tallent (gr...@mcn.org) wrote:
: But on my system(500/020) Miami slows the system down to the
: point that the mouse starts to jerk around when set to 38,400.
Wow! I have an A2000/68000, and my serial port is set for 19200, and I
don't see the problems you mentioned. If anything my system would be
much slower than yours, but it runs fine (as fine can be on an
unaccelerated A2000.)
: Also it took me about 2 minutes to configure Termite and I've never had a
: problem since then.
The same holds true with my setup of Miami.
: If Miami wasn't such a CPU pig,
If anything is a pig, it's @#$!@ MUI. *BLECH!* The Amiga
deserves better.
As I looked over the original post (again), I realized that Miami and
Termite may not be what the original poster requires. (Ease of setup was
not a requirement.)
Joeayo
GO to databases, select services, paswords, groups, etc from Ami-TCP
directory and hit the right mouse button over the item, you can then
import ALL settings from Ami-TCP and everyone once added is completely
updatable and editable from there.
Rask Ingemann Lambertsen
Den 04-Jan-97 03:45:33 skrev Johnathan Gapen fřlgende om "Re: Miami or AmiTCP
- Which is better?":
> On 03 Jan 97 21:51:25 +0100, Tobias Geijersson
> <tobias.g...@swipnet.se> wrote:
>>
>>I tried to add deamons, gateway and stuff (gave up trying with Miami).
> How hard can it be? You go to the InetD entry on the Database page, and
> enter the same info you'd enter into the AmiTCP:db/inetd.conf file.
> Sometimes, you have to enter a line that would have gone into
> AmiTCP:db/services in Miami's services list, also on the Database page.
> That's it, you're done, and you don't even have to re-start Miami.
Guess what, you don't have to restart AmiTCP/IP either. You don't even have to
restart inetd. How hard can it be?
Regards,
/ŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻTŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻ\
| Rask Ingemann Lambertsen | E-mail: ra...@kampsax.dtu.dk |
| Amiga GNU CC README maintainer | WWW: http://www.gbar.dtu.dk/~c948374/ |
| Registered Phase5 developer | "ThrustMe" on XPilot, ARCnet and IRC |
| Whoever said you can't multitask a Mac didn't own an Amiga |
Patrick Sheffield
>The fact is that Miami has two strong points in the comparison with
>AmiTCP: Easier installation and lower price.
>As far as speed is concerned they're pretty much the same, but when it
>comes to memory overhead my private tests showed that AmiTCP +
>ppp.device take less than Miami in the standard configuration, even if
>MUI is loaded in both set-ups. Or put in another way: AmiTCP +
>ppp.device + IBrowse take up considerably less memory than Miami +
>IBrowse.
And why has no one ever mentioned this: The main reason I haven't purchased
Miami is that it has a memory leak. A major one. If I leave it running, tho
disconnected, it will eventually consume all available memory until the
system crashes. It returns it all once I quit, but while it is running, it
will continually consume memory.
I don't think this is a bug in my setup since, as I pointed out, it does
release all the memory once I quit. I have only noticed this when it runs
when I am disconnected. I don't think it consumes while it is connected
(maybe it has too much to do so it doesn't sit around eating...;-) )
Patrick Sheffield
Patrick Sheffield
>: >Everyone says that Miami is easy to configure but I found AmiTCP easier
>: >when I tried to add deamons, gateway and stuff (gave up trying with
>: >Miami).
>MIAMI is much easier than AmiTCP. Hands down!
One other possible consideration... Miami ONLY works with serial devices.
AmiTCP works with ANY sana2 device. So if you were considering any ethernet
or PLIP or AmigaLink, etc, Miami isn't made for that.
Patrick Sheffield
Patrick Sheffield
Mitch Thompson wrote:
>Hello, Patrick Sheffield, on 24-Jan-97 17:39:30 you posted a message in
>comp.sys.amiga.networking concerning Re: Miami or AmiTCP - Which is better?,
>in which you stated:
>>One other possible consideration... Miami ONLY works with serial devices.
>>AmiTCP works with ANY sana2 device. So if you were considering any ethernet
>>or PLIP or AmigaLink, etc, Miami isn't made for that.
>>Patrick Sheffield
>That's no longer true. The current beta copies support SANA devices.
> From the Miami window, go to the /Interface/ section, and click on the
> /Interface type/ button. You have the options:
>serial (PPP/CSLIP)
>SANA II Point-to-Point
>SANA-II EtherNET
>SANA-II ArcNET
>SANA-II Other bus/ring
So 1.0.2 is /not/ the most recent rev, eh? Guess I'd better find that Miami
http address again...
Patrick Sheffield
J. YURICK
Patrick Sheffield (pshef...@earthlink.net) wrote:
: So 1.0.2 is /not/ the most recent rev, eh? Guess I'd better find that Miami
: http address again...
: Patrick Sheffield
Nope, that's an old version. 1.9.3 is the latest to date.
J. YURICK
Patrick Sheffield (pshef...@earthlink.net) wrote:
: One other possible consideration... Miami ONLY works with serial devices.
: AmiTCP works with ANY sana2 device. So if you were considering any ethernet
: or PLIP or AmigaLink, etc, Miami isn't made for that.
: Patrick Sheffield
I don't think so. Miami comes with SANA2 support.
Eka Durmin
Ahem... and what about the speed... which is faster when using SLIP/PPP?
Regards,
--
__|||||__
(o o)
+-----------------------oOO-(_)-OOo------------------------+
| email: e...@stack.nl (MIME/PGP welcome!) |
| www: http://www.stack.nl/~eka/ (Visit it!) |
+----------------------------------------------------------+
A Welcoming Smile
A Sympathetic Ear
A Kind Word
Can Show God's Love
John S. Burger
On or about 24-Jan-97 10:39:30 Patrick Sheffield typed the following words
about "Re: Miami or AmiTCP - Which is better?". My reply is thus...
Hi Patrick,
>>: >Everyone says that Miami is easy to configure but I found AmiTCP
>>: >easier when I tried to add deamons, gateway and stuff (gave up trying
>>: >with Miami).
>>MIAMI is much easier than AmiTCP. Hands down!
PS> One other possible consideration... Miami ONLY works with serial
PS> devices. AmiTCP works with ANY sana2 device. So if you were
PS> considering any ethernet or PLIP or AmigaLink, etc, Miami isn't made
PS> for that.
Apparently you havn't looked a Miami lately!!!
PS> Patrick Sheffield
--
-= John =- jsbu...@xmission.com BIX: jburger
Amiga 2000 Via
68030 25MHz Amateur Radio KB0ES T
2+44 MB RAM // H
850 MB HD John S. Burger Hooper, UT \X/ O
Iomega ZIP R
Toshiba 6.7x CD This message was composed on... 2.5 Carme
27-Jan-97 12:01:55 MST
--
Shell to DOS....Come in, DOS.....Do you Copy?
Adam Szymczak
On 24-Jan-97 12:28:16, Patrick Sheffield (pshef...@earthlink.net) wrote:
>>The fact is that Miami has two strong points in the comparison with
>>AmiTCP: Easier installation and lower price.
>>As far as speed is concerned they're pretty much the same, but when it
>>comes to memory overhead my private tests showed that AmiTCP +
>>ppp.device take less than Miami in the standard configuration, even if
>>MUI is loaded in both set-ups. Or put in another way: AmiTCP +
>>ppp.device + IBrowse take up considerably less memory than Miami +
>>IBrowse.
Define "considerably less"? It is this kind of bogus and hollow statements
that fuel the AmiTCP vs. Miami debate. First, the guy states that it takes
less (how much less) and then states that it takes "considerably less"?
Makes you wonder.
>And why has no one ever mentioned this: The main reason I haven't purchased
>Miami is that it has a memory leak. A major one. If I leave it running, tho
>disconnected, it will eventually consume all available memory until the
>system crashes. It returns it all once I quit, but while it is running, it
>will continually consume memory.
Maybe no one has mentioned it because it has NOT happened to them. I have
left Miami running while offline and have not ever noticed it "consuming"
memory nor crashing. Have you notified Holger about your "memory leak"
problem?
>I don't think this is a bug in my setup since, as I pointed out, it does
>release all the memory once I quit. I have only noticed this when it runs
>when I am disconnected. I don't think it consumes while it is connected
>(maybe it has too much to do so it doesn't sit around eating...;-) )
Alternatively, it could be a MUI related process that Miami has called.
Adam Szymczak
--
"See-toe, Reach-oh, Malto-Ray." -- SoulHunter
Neil Bothwick
Hi there Eka Durmin, I believe it was you who said:
> Ahem... and what about the speed... which is faster when using SLIP/PPP?
There's no real difference. The limiting factor is the speed of your
link, and the implementation fo SLIP/PPP you use (as in the speed
restriction on the unregistered ppp.device).
Neil
--
Neil Bothwick, Warrington, England - CU Amiga CD compiler
Connected via Wirenet - The UK's first Amiga-only internet access provider
Mail: in...@wirenet.u-net.com WWW: http://www.u-net.com/~wirenet/
--
ALZHEIMER.COM found . . . Out of . . . something . .
Storm
Adam Szymczak (szy...@server.uwindsor.ca) wrote:
: On 24-Jan-97 12:28:16, Patrick Sheffield (pshef...@earthlink.net) wrote:
: >>As far as speed is concerned they're pretty much the same, but when it
: >>comes to memory overhead my private tests showed that AmiTCP +
: >>ppp.device take less than Miami in the standard configuration, even if
: >>MUI is loaded in both set-ups. Or put in another way: AmiTCP +
: >>ppp.device + IBrowse take up considerably less memory than Miami +
: >>IBrowse.
: Define "considerably less"? It is this kind of bogus and hollow statements
: that fuel the AmiTCP vs. Miami debate. First, the guy states that it takes
: less (how much less) and then states that it takes "considerably less"?
: Makes you wonder.
Figures have been posted, many months ago, when Miami was first released.
Search on DejaNews if you really want to know, or do the tests yourself.
As far as I can remember, it was in the vicinity of 150-200k extra memory
used by Miami+IBrowse.
-- ______________________________
\_/ "\/\/\__"\/ "\/ "\/\__"\_/
Storm / Cydonia / / / / / / / / / / / / ' / Packing class
/ /\/> / / / / / / / / / /__ & kicking arse!
(coder) \__/ \_/\__/\__/\/\/\/\/\/ \/
Jonas Thorell
Storm (st...@ar.ar.com.au) wrote:
>Figures have been posted, many months ago, when Miami was first released.
>Search on DejaNews if you really want to know, or do the tests yourself.
>As far as I can remember, it was in the vicinity of 150-200k extra memory
>used by Miami+IBrowse.
Oh, that much. Gee, much to complain about. In order to use the internet
on an Amiga, realistically speaking you'll need atleast 6 megs of memory
(if you don't want to go insane during the process) and is 200k anything
to make a fuss over then?
---
Jonas Thorell |Internet: jon...@bahnhof.se |"Animals are our
Skyttbacksvägen 11 |Fidonet: 2:206/124.7 |to wear, eat and
740 34 Skyttorp |Voice: +46-18-352444 | experiment on"
Sweden | |- Cat rapes dog
---
Lost .sig. Reward offered...
Mike Dijkema
In article <964.6970...@bahnhof.se> Jonas Thorell <jon...@bahnhof.se> writes:
>From: Jonas Thorell <jon...@bahnhof.se>
>Subject: Re: Miami or AmiTCP - Which is better?
>Date: 31 Jan 97 02:30:34 +0100
>Storm (st...@ar.ar.com.au) wrote:
Miami is faster and easier to configure.
Mike.
Neil Bothwick
Hi there Mike Dijkema, I believe it was you who said:
> Miami is faster and easier to configure.
It may be easier to configure for many people, but can you say how
much faster? All the tests I have done and read show insignificant
speed differences between Miami and AmiTCP+ppp.device.
Neil
--
Neil Bothwick, Warrington, England - CU Amiga CD compiler
Connected via Wirenet - The UK's first Amiga-only internet access provider
Mail: in...@wirenet.u-net.com WWW: http://www.u-net.com/~wirenet/
--
Two most common elements: Hydrogen and Stupidity
Jonas Thorell
Mike Dijkema (dijk...@bureau.rug.nl) wrote:
>Miami is faster and easier to configure.
Most definatly. The best money I ever spent on computer software.
I'm not so sure it's faster than AmiTCP and ppp.device though.
With unregistered ppp.device, without a doubt. With a registered
ppp.device I would call it even.
---
Jonas Thorell |Internet: jon...@bahnhof.se |"Animals are our
Skyttbacksvägen 11 |Fidonet: 2:206/124.7 |to wear, eat and
740 34 Skyttorp |Voice: +46-18-352444 | experiment on"
Sweden | |- Cat rapes dog
---
How did a fool and his money get together in the first place?
Greg Tallent
Way back on 31-Jan-97 16:43:17 Neil Bothwick said: Re: Miami or AmiTCP - Which
is better?
>Hi there Mike Dijkema, I believe it was you who said:
>> Miami is faster and easier to configure.
>It may be easier to configure for many people, but can you say how
>much faster? All the tests I have done and read show insignificant
>speed differences between Miami and AmiTCP+ppp.device.
Has anobody thought of adding TermiteTCP to this mix? I know it doesn't
support as much stuff as Miami and AmiTCP, But with the stuff it's compatible
with, It runs circles around Miami AND AmiTCP.
On my 020, Termite runs at 115,200 with a medium system load, Whereas
Miami at 38,400 is such a load that I can barely move the mouse pointer.
Running it at 57,600 is not much fun for me.
Also Termite uses less memory than either of the other two, Kind of a moot
point when you have 8 Megs of ram, But still useful.
And Termite isn't that hard to configure either, It's not automatic like
Miami but it's a heck of a lot easier than AmiTCP!
Using Thor, When I get New Mail, I average 4000cps according to Thor, Miami
on the other hand, Averages about 2100cps. You might want to take this with a
rather large grain of salt, Though.
>Neil
<TSB>
Greg Tallent |Amiga2000/040/33mhz/3.1 32 megs ram,2 gig/Toaster3.1 |
gr...@mcn.org |Amiga 500/020/14mhz/3.1 8 megs ram,2.5 gig/Supra28.8|
In Northern Calif. |TermiteTCP,IBrowse,AWeb II,Miami 1.1,DirWork 2.1 |
Living on 750 Acres|P133/64 Megs ram,9 gig hd's, LightWave, WinNT 4.0 |
Fun,FUN,FUN!! NOT! |AMD P133/32 Megs ram 2 gig hd's, LightWave, Win95 |
Member: International Brotherhood of Tagline Thieves!
Olivier Anh
Mike Dijkema (dijk...@bureau.rug.nl) was telling us :
>Miami is faster and easier to configure.
Faster ? how much ? (I mean with miami 2.0)
Easier to configure : absolutely right.
Read you
Oliver
---
/ Olivier Anh \
| -:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:- |
| Email : a...@inforoute.cgs.fr |
\ Homepage : www.mygale.org/06/anh /
Olivier Anh
Hey I just download Miami 2.0. It's very cool. I can't test the SANA II port.
Did already someone ?
I think that now, AMITCP and Miami are both equal.
Patrick Sheffield
>And why has no one ever mentioned this: The main reason I haven't
>purchased Miami is that it has a memory leak. A major one. If I leave it
>running, tho disconnected, it will eventually consume all available
>memory until the system crashes. It returns it all once I quit, but while
>it is running, it will continually consume memory.
>Patrick Sheffield
Well, The new version is fixed, no memory leaks. I bought it and I am
happy.
Patrick Sheffield
Josh Wingell
Neil Bothwick wrote:
>
> Hi there Mike Dijkema, I believe it was you who said:
>
>
> much faster? All the tests I have done and read show insignificant
> speed differences between Miami and AmiTCP+ppp.device.
It depends on what you are doing. Miami supports T/TCP, which speeds
up browser connections to the WWW. Of course, you need a browser that
supports it...like Voyager :)
Miami also has ping flood protection (it stops returning them if there
is a flood). So, that could speed up your connection if someone really
hated you ;)
As for regular connections? Well...I don't think Miami is faster...but
it certainly isn't slower. And it is based on a newer version of BSD
than AmiTCP is. That is a good thing, IMO.
And Miami is being actively developed. I think AmiTCP might just be
inactively developed :) Has anyone heard anything from NSDi recently?
Josh
--
Amiga///Joshua B. Wingell (Shape)\ A3000T/060/50MHz/50MB *AmigaOS v3.1*
/// Home: win...@ziplink.net \ 3GB HD/SyQuest EZ135/NEC3Xi CD-ROM
\\\/// AmigaOS/Linux68k/MacOS \ GVP ioExtender/USR Sportster 33.6!
\XX/http://www.ziplink.net/~wingell\ Piccolo SD64 4MB / Nanao TX*C7S
"Spoon!" / The views expressed here are not necessarily the \ "Spoon!"
-The Tick\ views of Digital Equipment Corporation. /-The Tick
Chris Wiles
>
> And Miami is being actively developed. I think AmiTCP might just be
> inactively developed :) Has anyone heard anything from NSDi recently?
Yes, us! We are working with them at the moment.
Chris.
--
-------------------------------------------------------------------------
ACTIVE SOFTWARE | Tel - +44 (0)1325 352260
P.O. Box 151, | E-Mail - act...@enterprise.net <Active Software>
Darlington, | E-Mail - ch...@active2.demon.co.uk <Chris Wiles>
County Durham, | Contact- Chris Wiles <Managing Director>
DL3 8YT, ENGLAND | Hours - 9.30am - 6pm GMT Monday to Friday
gregh
To be honest with you, I think the ONLY true answer to this question is:
The one that is better is the one you LIKE the most. Other than that, at
this moment, it's apples and oranges to me. ;-}
------------------------------------------------------------------------
|Sysop of Amiga's Sci-Fi BBS gr...@fl.net.au Chow Chow lover! |
|Are you old when you enjoy a good headbanger record with others over 40?|
------------------------------------------------------------------------
Uwe Sigurd Valentin Kubosch
Chris Wiles <enqu...@active2.demon.co.uk> writes:
> > And Miami is being actively developed. I think AmiTCP might just be
> > inactively developed :) Has anyone heard anything from NSDi recently?
>
> Yes, us! We are working with them at the moment.
So, do you know what's going on? Is there a new version in the
horizon?
Do you know why they _NEVER_ answer email? Much like phase5, BTW.
What are you working on with them?
donV
--
Uwe S.V. Kubosch Milron Data Crusaders Productions
Stavikbakken 3 Johan Svensensv. 3a Stasjonsveien 56
N-1472 Fjellhamar N-1472 Fjellhamar N-2010 Strømmen
Norway Norway Norway
Phone:+47 67 90 11 95 Phone:+47 6790 9524 Phone:+47 63 80 33 24
Phone:+47 92 20 60 46 Phone:+47 63 80 00 53 (ISDN)
Web:<http://www.ifi.uio.no/~uwek/> Fax :+47 63 80 00 54 (ISDN)
Crusaders:<http://www.crusaders.no/> BBS :+47 22 10 46 46
Chris Wiles
Hi,
> So, do you know what's going on? Is there a new version in the
> horizon?
Well, sort of :)
We have been working with NSDi, since last Summer, to "upgrade" AmiTCP
(DialUP/Internet version) with a GUI based system for ease-of-use (this
is mainly aimed towards new users, novices/beginners etc.
The product is called NetConnect and will be out Real Soon Now (tm),
trademark owned by Olli Wagner ;)
> Do you know why they _NEVER_ answer email? Much like phase5, BTW.
Depends what you ask them :) AFAIK they all work for Nokia in Finland,
working long hours etc.
> What are you working on with them?
People who have seen our adverts for NetConnect (in the UK Amiga press)
say "why aren't you using Miami?". Well, imho, AmiTCP v4.5 DialUP is
a nice piece of software, easy to install/configure (GUI driven) and
uses far less memory than Miami. Also, and this is true as I have asked
many of our customers about this, there are quite a few new users/beginners
using Miami who think it is "untouchable" - ie. there are so many options
within the GUI it is very difficult to know what buttons does what! :)
AmiTCP does have a future but in different roles (as part of the NetConnect
product etc).
Chris.
Uwe Sigurd Valentin Kubosch
Chris Wiles <act...@enterprise.net> writes:
> > So, do you know what's going on? Is there a new version in the
> > horizon?
> Well, sort of :)
> We have been working with NSDi, since last Summer, to "upgrade" AmiTCP
> (DialUP/Internet version) with a GUI based system for ease-of-use (this
> is mainly aimed towards new users, novices/beginners etc.
> The product is called NetConnect and will be out Real Soon Now (tm),
> trademark owned by Olli Wagner ;)
Sounds neat. With 4.3 I grew very comfortable with the setup,
but a GUI for setup would be good.
> > Do you know why they _NEVER_ answer email? Much like phase5, BTW.
> Depends what you ask them :) AFAIK they all work for Nokia in Finland,
> working long hours etc.
As faar as I can tell, my questions have been good questions, and
I have bought Amitcp so I think I should get an answer. Anyway,
not your problem.
> > What are you working on with them?
> People who have seen our adverts for NetConnect (in the UK Amiga press)
> say "why aren't you using Miami?". Well, imho, AmiTCP v4.5 DialUP is
> a nice piece of software, easy to install/configure (GUI driven) and
> uses far less memory than Miami. Also, and this is true as I have asked
> many of our customers about this, there are quite a few new users/beginners
> using Miami who think it is "untouchable" - ie. there are so many options
> within the GUI it is very difficult to know what buttons does what! :)
4.5??? Is that a coming version or is it out? In both cases
some propagande is needed. Make them advertise! Estimated
release dates!
> AmiTCP does have a future but in different roles (as part of the NetConnect
> product etc).
??? Have you bought Amitcp or what. Sorry, I didn't quite get
the meaning of that. i agree that Amitcp has a future since it
is the only tcp stack that supports multiple devices.
Matthew Hunter
On 7 Feb 97 10:15:40 GMT+0000, Chris Wiles <act...@enterprise.net>
wrote:
>People who have seen our adverts for NetConnect (in the UK Amiga press)
>say "why aren't you using Miami?". Well, imho, AmiTCP v4.5 DialUP is
>a nice piece of software, easy to install/configure (GUI driven) and
What about those users who have ethernet connections, and simply
want an upgraded AmiTCP STACK, not necessarily an upgraded GUI? I
bought 4.3, but now that a sana-II Miami is out and has apparantly a
much more advanced TCP/IP implementation, why should I stay with
AmiTCP?
--
Matthew Hunter (mhun...@andrew.cmu.edu) (Amiga 4000/030)
http://www.contrib.andrew.cmu.edu/~mhunter -- My Homepage
http://shadow.res.cmu.edu/WheelOfTime/ for the Wheel of Time FAQ
Markus Dages
Olivier Anh (a...@inforoute.cgs.fr) wrote:
: Mike Dijkema (dijk...@bureau.rug.nl) was telling us :
:
: >Miami is faster and easier to configure.
: Faster ? how much ? (I mean with miami 2.0)
:
: Easier to configure : absolutely right.
I think it`s dependent on what you are doing with TCP/IP. If you are
only connect one machine to a TCP/IP network, miami will be the better
choice, because it`s easier to configure and under some circumstances
a little bit faster.
But if you plan to set up your own network or a gateway with more
than one connection, than AmiTCP is the best choice for Amiga
computers.
regards,
Markus Dages
Allen Wittenauer
Matthew Hunter wrote in comp.sys.amiga.networking about "Re: NSDi News?":
> >Well, imho, AmiTCP v4.5 DialUP is a
> >nice piece of software, easy to install/configure (GUI driven) and
>
> What about those users who have ethernet connections, and simply want an
> upgraded AmiTCP STACK, not necessarily an upgraded GUI?
I definitely second Matthew's concerns... One of my biggest gripes is that
AmiTCP doesn't come with lpr support natively and various other "niceties"
that TCP stacks for other platforms support (as well as other TCP stacks
for the Amiga, from my understanding).
If all of the newer versions of AmiTCP are only going to benefit PPP
and SLIP users with their setups, then I would say that it is time for
some of us to start shopping around for a new stack.
Zap
>Matthew Hunter wrote in comp.sys.amiga.networking about "Re: NSDi News?":
>> >Well, imho, AmiTCP v4.5 DialUP is a
>> >nice piece of software, easy to install/configure (GUI driven) and
>>
AmiTCP v4.5? when did that come out?
--
Regards,
Zap
"I didn't do it... No one saw me do it... You can't prove anything..."
Bart Simpson.
Chris Wiles
> Matthew Hunter wrote in comp.sys.amiga.networking about "Re: NSDi News?":
> I definitely second Matthew's concerns... One of my biggest gripes is that
> AmiTCP doesn't come with lpr support natively and various other "niceties"
> that TCP stacks for other platforms support (as well as other TCP stacks
> for the Amiga, from my understanding).
>
> If all of the newer versions of AmiTCP are only going to benefit PPP
> and SLIP users with their setups, then I would say that it is time for
> some of us to start shopping around for a new stack.
99.9% of users who use AmiTCP 4.5 Dialup are pleased with it - hey, I don't
need Miami.
Many of these users have seen Miami and find the main GUI overly difficult
to understand compared with AmiTCP 4.5 Dialup.
We use AmiTCP 4.5 Dialup within NetConnect. IMHO, and many of the testers
opinions, it is all you need if you are an Internet new user.
later,
Chris.
--
-------------------------------------------------------------------------
ACTIVE SOFTWARE | Tel - +44 (0)1325 352260
P.O. Box 151, | Fax - +44 (0)1325 482343
Darlington, | E-Mail - act...@netcomuk.co.uk <Active Software>
Holger Kruse
In article <3304452E.MD-...@netcomuk.co.uk>, Chris Wiles
<act...@netcomuk.co.uk> wrote:
> Many of these users have seen Miami and find the main GUI overly difficult
> to understand compared with AmiTCP 4.5 Dialup.
You are really comparing apples and oranges here...
To *configure* Miami all you need is the GUI of MiamiInit, and that
GUI is really very simple. I don't think I have received a single
complaint so far about MiamiInit being difficult to use -- quite the
contrary really :-)
Out of the main Miami GUI (which you claim is "overly difficult") all
you need for daily use are the separate "Online" and "Offline" buttons,
and the close gadget -- not that difficult either IMHO.
All those other buttons are only needed for advanced customization
and tuning, so you don't *have* to use them if you don't know or
understand what they do. Assuming you *want* to do some fine-tuning or
other changes, let's see what you have to for both stacks:
- with Miami you have a GUI for *all* settings, you usually see all
available choices in cycle gadgets or pop-up menus, you can get MUI
"bubble help" for most gadgets, and you can get extensive
context-sensitive help simply by pressing the "Help" key.
- with AmiTCP you need to get your manual, find out which text files
contain the options you need, find out which options to change and
which keywords are valid for them, then edit those text files with
a text editor (without getting feedback whether you got the option
right), and in many cases shut down and restart AmiTCP.
I have difficulty understanding how even the main Miami GUI is "overly
difficult" compared to *that*.
Don't get me wrong - I have nothing personal against AmiTCP at all.
But I would appreciate it if you would not post such obviously
misleading half-truths about a competitor's product...
--
Holger Kruse kr...@nordicglobal.com
http://www.nordicglobal.com
Matthew Hunter
In article <3304452E.MD-...@netcomuk.co.uk>, Chris Wiles wrote:
>> Matthew Hunter wrote in comp.sys.amiga.networking about "Re: NSDi News?":
I didn't write anything that you quoted. Please be careful to snip
unnecessary attributions.
>> If all of the newer versions of AmiTCP are only going to benefit PPP
>> and SLIP users with their setups, then I would say that it is time for
>> some of us to start shopping around for a new stack.
>99.9% of users who use AmiTCP 4.5 Dialup are pleased with it - hey, I don't
>need Miami.
That's fine. I really don't care all that much if 100% of your users of 4.5
DialUp are pleased. I'm -not- a dialup user for 9 months of the year, and I can
get by just fine with 4.3 during those 3 that I am. Most of the time, I'm using
an ethernet connection.
>Many of these users have seen Miami and find the main GUI overly difficult
>to understand compared with AmiTCP 4.5 Dialup.
I don't care what they think of the GUI for the dialup configuration.
>We use AmiTCP 4.5 Dialup within NetConnect. IMHO, and many of the testers
>opinions, it is all you need if you are an Internet new user.
I am not an Internet new user.
Let me point out that your "responses" to my concerns, and the concerns of
the person who originally responded to my concerns, are little but thinly-veiled
advertisements for your product, and do *NOT* in any way address the concerns
that were raised.
If you have no information with which to allay the concerns that were raised,
please refrain from advertising your product, and simply tell us that you don't
have any information on the matter.
In case you forgot:
I use ethernet. I am not a dialup user for most of the year, and I have no
concerns whatsoever about setting up AmiTCP for a dialup connection, using my
current version, when I need to do so.
I purchased, for quite a steep price, AmiTCP 4.3.
What improvements, if any, are intended for non-dialup users? What
improvements, if any, are planned in the core TCP/IP protocol code? Will AmiTCP
implement T/TCP? What advantages, if any, will AmiTCP offer over Miami Deluxe
by Holger Kruse, which supports (as I understand it) a much more current
implementation of the TCP/IP protocol?
I would be much gratified to have these questions answered directly, whether
they are "I don't know" or something more substantial.
--
Matthew Hunter (mhun...@andrew.cmu.edu) (Amiga 4000/060)
J. Bronson
J. Bronson Replies To,
> CUT CUT
>Don't get me wrong - I have nothing personal against AmiTCP at all.
>But I would appreciate it if you would not post such obviously
>misleading half-truths about a competitor's product...
>--
>Holger Kruse kr...@nordicglobal.com
> http://www.nordicglobal.com
You tell them Holger!
How can entering your username, Phone number and password be
hard to do or understand.
I spent 3 weeks trying to getting Amitcp up and running. On top
of that they do not answer their mail!! Even after I paid them a
hundred dollars U.S. Amitcp 4.3 never did work right and if it had
not been for Holger telling me how to fix Amitcp I never would
have got it going!
Holger has ALWAYS answered my questions with in a couple of
hours. Twice within three minutes, now thats service, that want I am
willing to pay for.
<tsb>
jbro...@maxinet.com
Jerry Bronson
Shasta Ca. U.S.A
14-Feb-97 15:52:12
Amiga 4000/40 10 MEGS CD-ROM DESKJET USR-SPORTSTER
<tsb>
If we don't survive, we don't do anything else.
Allen Wittenauer
Chris Wiles wrote in comp.sys.amiga.networking about "Re: NSDi News?":
> 99.9% of users who use AmiTCP 4.5 Dialup are pleased with it
> We use AmiTCP 4.5 Dialup within NetConnect. IMHO, and many of the
> testers opinions, it is all you need if you are an Internet new user.
In my home, I have a 10BaseT network with several machines hooked to it,
including a NeXT Cube and a BeBox. At work, I support well over 300
UNIX machines of various flavors, including a large portion of the
network connections these machines make. I *hardly* consider myself an
Internet new user.
I'm an _advanced_ TCP/IP user. If you are gearing the newer versions
of AmiTCP towards Dialup users and leaving behind those of us who
bought the package to begin with (of which, I would be willing to
bet a large chunk have dedicated network connections), then you are going
to lose a fairly large base of your customers who have needs that are not
met by your product.
I find it very suspicious that you are pulling up Miami, a topic of which
I don't think has ever appeared in this thread up until your mention of it.
I can only suspect that NSDi (and/or yourselves) have truly rested on your
laurels and now realize that other people are entering a market that you
either have lost or are about to lose... and are (if you pardon an
American phrase) "scared shitless."
Chris Wiles
Hello,
> > Many of these users have seen Miami and find the main GUI overly difficult
> > to understand compared with AmiTCP 4.5 Dialup.
>
> To *configure* Miami all you need is the GUI of MiamiInit, and that
> GUI is really very simple. I don't think I have received a single
> complaint so far about MiamiInit being difficult to use -- quite the
> contrary really :-)
Hold on. Miami *is* very easy to use. That isn't even an argument! What
I said was AmiTCP v4.5 NetConnect version was *also* easy to install ;)
> Out of the main Miami GUI (which you claim is "overly difficult") all
> you need for daily use are the separate "Online" and "Offline" buttons,
> and the close gadget -- not that difficult either IMHO.
No, they are not and I agree. BUT AmiTCP 4.5 NetConnect also offers
simplistic Internet connectivity - which was flamed by a number of
people who haven't seen it. I was just making a point.
> - with AmiTCP you need to get your manual, find out which text files
> contain the options you need, find out which options to change and
> which keywords are valid for them, then edit those text files with
> a text editor (without getting feedback whether you got the option
> right), and in many cases shut down and restart AmiTCP.
You haven't seen AmiTCP 4.5 NetConnect then <grin> :)
> I have difficulty understanding how even the main Miami GUI is "overly
> difficult" compared to *that*.
<sigh> I am comparing it to AmiTCP 4.5 Dialup (NetConnect version), not
the old AmiTCP which, as you say, is certainly not user friendly! :)
> Don't get me wrong - I have nothing personal against AmiTCP at all.
> But I would appreciate it if you would not post such obviously
> misleading half-truths about a competitor's product...
I am not. The main Miami GUI *is* nice and easy for most existing Internet
users (but often over the top for new users). Miami is a superb piece of
Internet software and would have been used by us as our *first choice* if
we could have agreed on a price for distribution within NetConnect.
All the best,
Matthew Hunter
In article <3306ECE0.MD-...@netcomuk.co.uk>, Chris Wiles wrote:
>No, they are not and I agree. BUT AmiTCP 4.5 NetConnect also offers
>simplistic Internet connectivity - which was flamed by a number of
>people who haven't seen it. I was just making a point.
I haven't seen ANYONE flaming AmiTCP 4.5, Chris. I HAVE seen a lot
of people asking about what exactly is in store for the people who
bought AmiTCP 4.3 with the expectation of at least a FEW updates
eventually, and for whom AmiTCP 4.5 doesn't exactly constitute an
upgrade. (Particularly since it isn't publically available, even
to registered users). That doesn't constitute a flame to anyone, I
don't think.
Those questions have still not been answered, not even with "I
don't know" or "I am not allowed to say" or "See, there's this cat
with a deathgrip on my tongue..."
Zap
>> to understand compared with AmiTCP 4.5 Dialup.
>You are really comparing apples and oranges here...
>To *configure* Miami all you need is the GUI of MiamiInit, and that
>GUI is really very simple. I don't think I have received a single
>complaint so far about MiamiInit being difficult to use -- quite the
>contrary really :-)
buy the time you run the m(u)iamiInit program (waiting for MUI to load and
steal all of your memory) you could have finished editing the text files in
AmiTCP:db
>Out of the main Miami GUI (which you claim is "overly difficult") all
>you need for daily use are the separate "Online" and "Offline" buttons,
>and the close gadget -- not that difficult either IMHO.
GUI's arent needed to run a TCP Stack
>All those other buttons are only needed for advanced customization
>and tuning, so you don't *have* to use them if you don't know or
>understand what they do. Assuming you *want* to do some fine-tuning or
>other changes, let's see what you have to for both stacks:
M(u)iami in its default config, is left wide open to attack.
>- with Miami you have a GUI for *all* settings, you usually see all
> available choices in cycle gadgets or pop-up menus, you can get MUI
> "bubble help" for most gadgets, and you can get extensive
> context-sensitive help simply by pressing the "Help" key.
if you need "extensive context-sensitive help" you shouldnt be touching any of
the settings.
>- with AmiTCP you need to get your manual, find out which textfiles
>contain the options you need, find out which options to change and
>which keywords are valid for them, then edit those text files with
>a text editor (without getting feedback whether you got the option
>right), and in many cases shut down and restart AmiTCP.
CTRL-F works 90% of the time
>I have difficulty understanding how even the main Miami GUI is "overly
>difficult" compared to *that*.
>Don't get me wrong - I have nothing personal against AmiTCP at all.
>But I would appreciate it if you would not post such obviously
>misleading half-truths about a competitor's product...
>--
>Holger Kruse kr...@nordicglobal.com
> http://www.nordicglobal.com
Regards,
Zap
--
z...@betty.uchicago.edu
http://coming.soon/~zap
Holger Kruse
In article <2498.698...@betty.uchicago.edu>, Zap <z...@betty.uchicago.edu>
wrote:
> M(u)iami in its default config, is left wide open to attack.
Argl... I wish people would stop posting such lies.
The truth is: Miami 2.0f in its *default* config complies with all
the latest CERT advisories against Internet attacks. In fact Miami
is the *only* TCP/IP stack for AmigaOS at the moment which is
invulnerable to TCP sequence number attacks *and* IP fragment
attacks (which are the most dangerous attacks these days), all
that without any changes to the default config.
I don't even want to mention that huge, gaping security hole in
the fingerd of AmiTCP 4.3. Hopefully this has been fixed in 4.5
though...
Michael Brockman
>>In article <3304452E.MD-...@netcomuk.co.uk>, Chris Wiles
>><act...@netcomuk.co.uk> wrote:
>>
>>> Many of these users have seen Miami and find the main GUI overly difficult
>>> to understand compared with AmiTCP 4.5 Dialup.
>>You are really comparing apples and oranges here...
>>To *configure* Miami all you need is the GUI of MiamiInit, and that
>>GUI is really very simple. I don't think I have received a single
>>complaint so far about MiamiInit being difficult to use -- quite the
>>contrary really :-)
>buy the time you run the m(u)iamiInit program (waiting for MUI to load and
>steal all of your memory) you could have finished editing the text files in
>AmiTCP:db
You forgot about the time spent getting the information from your ISP about
DSN IP's, Your gateway IP, etc...The time spend running the Config_AmiTCP
to create your startnet script. Not to mention the time spent configuring and
testing your SANA 2 configuration. With Miami all you need is the ISP's phone
number, your userid, and password. An average user doesn't understand all the
insides of the TCP/IP setup. And there is no need for them to. I personally
run AmiTCP, but I am very impressed with Miami ease-of-use and setup time. You
just seam to have a problem with MUI, not with Miami. By the way, Miami and
MUI use 250K less memory than AmiTCP (On my system 750K for Miami vrs 1 Meg
for AmiTCP).
>>Out of the main Miami GUI (which you claim is "overly difficult") all
>>you need for daily use are the separate "Online" and "Offline" buttons,
>>and the close gadget -- not that difficult either IMHO.
>GUI's arent needed to run a TCP Stack
True, but to Administer a TCP stack, A gui really helps. Must be why Solaris
(Sun's OS) has one, IRIX (SGI's OS) has one, etc...etc... I could keep going
on with more examples. Most users prefer looking/controlling the stack through
a GUI.
>>All those other buttons are only needed for advanced customization
>>and tuning, so you don't *have* to use them if you don't know or
>>understand what they do. Assuming you *want* to do some fine-tuning or
>>other changes, let's see what you have to for both stacks:
>M(u)iami in its default config, is left wide open to attack.
Please expand on this statement. This is a rumour stll not supported by any
facts. Give us proof of what you are saying.
>>- with Miami you have a GUI for *all* settings, you usually see all
>> available choices in cycle gadgets or pop-up menus, you can get MUI
>> "bubble help" for most gadgets, and you can get extensive
>> context-sensitive help simply by pressing the "Help" key.
>if you need "extensive context-sensitive help" you shouldnt be touching any
>of the settings.
Why not? After all it is my machine we are talking about. One of the ways we
learn is by breaking things. First you argue that Miami is too complex...Then
you berate Miami for giving all the information you need to configure your
system. You can't have it both ways... So which one is it?
>>Don't get me wrong - I have nothing personal against AmiTCP at all.
>>But I would appreciate it if you would not post such obviously
>>misleading half-truths about a competitor's product...
I agree with Holger on this one...It is one thing to not like a product. It is
crossing the line when you put down that product by being misleading and
presenting opinion as fact.
>>--
>>Holger Kruse kr...@nordicglobal.com
>> http://www.nordicglobal.com
>Regards,
>Zap
>--
>z...@betty.uchicago.edu
>http://coming.soon/~zap
--
<sb>
<sb>
/>
/> Michael M Brockman, mi...@skylink.net
/> http://www.skylink.net/~mikeb
|\_______{o}-------------------------------------------_
[\\\\\\\\\\\{MIKEY BAD}>============================- >
|/~~~~~~~{o}-------------------------------------------~
\> Have you ever been cruising on the Information
\> Superhighway, just to pause and take a whiz?
\>
<sb>
<sb>PGP Key Available Upon Request
<sb>
Matthew Hunter
In article <nntp.1997...@innovision-group.com>, Keith Blakemore-Noble wrote:
>On or around Sun, 16 Feb 97, Matthew Hunter is alledged to have muttered
>something along the lines of -
>MH> Those questions have still not been answered, not even with "I
>MH> don't know" or "I am not allowed to say" or "See, there's this cat
>MH> with a deathgrip on my tongue..."
>Hear hear! As someone who just bought the commercial 4.3, I too would be
>interested to know if it is actually still being developed for Ethernet use
>etc...
I received email from the person I was chastising. He didn't have
much to say, but DID indicate that some major improvements in the
stack itself were planned over the next several months, including
T/TCP.
Chris Wiles
Hello,
> >buy the time you run the m(u)iamiInit program (waiting for MUI to load and
> >steal all of your memory) you could have finished editing the text files in
> >AmiTCP:db
>
> You forgot about the time spent getting the information from your ISP about
> DSN IP's, Your gateway IP, etc...The time spend running the Config_AmiTCP
> to create your startnet script. Not to mention the time spent configuring and
> testing your SANA 2 configuration. With Miami all you need is the ISP's phone
> number, your userid, and password. An average user doesn't understand all the
> insides of the TCP/IP setup. And there is no need for them to. I personally
> run AmiTCP, but I am very impressed with Miami ease-of-use and setup time. You
> just seam to have a problem with MUI, not with Miami. By the way, Miami and
> MUI use 250K less memory than AmiTCP (On my system 750K for Miami vrs 1 Meg
> for AmiTCP).
Pardon? AmiTCP uses very little memory indeed. How can anything that use MUI
use less memory?? It can only use less memory when you are using other MUI
programs at the same time!
> >>Don't get me wrong - I have nothing personal against AmiTCP at all.
> >>But I would appreciate it if you would not post such obviously
> >>misleading half-truths about a competitor's product...
>
> I agree with Holger on this one...It is one thing to not like a product. It is
> crossing the line when you put down that product by being misleading and
> presenting opinion as fact.
..yes but I wasn't talking about the old AmiTCP 4.3 - I was talking about
AmiTCP 4.5 Dialup *with* a GUI and with a completely different system
of handling the preference files etc. And, who said I don't like Miami?
Chris Wiles
Hello,
> I find it very suspicious that you are pulling up Miami, a topic of which
> I don't think has ever appeared in this thread up until your mention of it.
> I can only suspect that NSDi (and/or yourselves) have truly rested on your
> laurels and now realize that other people are entering a market that you
> either have lost or are about to lose... and are (if you pardon an
> American phrase) "scared shitless."
"Scared shitless" of what? NetConnect is a complete suite of commercial
software for around $65!
This includes a *full* TCP stack with GUI, VNG, Microdot-II, AmFTP, AmIRC,
AmTelnet, AmTerm, AmFinger etc.
How can I be "scared shitless" with those specs for $65?
<snicker>
Keith Blakemore-Noble
On or around Tue, 18 Feb 97, Chris Wiles is alledged to have muttered
something along the lines of -
CW> > I find it very suspicious that you are pulling up Miami, a topic of which
CW> > I don't think has ever appeared in this thread up until your mention of it.
CW> > I can only suspect that NSDi (and/or yourselves) have truly rested on your
CW> > laurels and now realize that other people are entering a market that you
CW> > either have lost or are about to lose... and are (if you pardon an
CW> > American phrase) "scared shitless."
CW>
CW> "Scared shitless" of what? NetConnect is a complete suite of commercial
CW> software for around $65!
CW>
CW> This includes a *full* TCP stack with GUI, VNG, Microdot-II, AmFTP, AmIRC,
CW> AmTelnet, AmTerm, AmFinger etc.
CW>
CW> How can I be "scared shitless" with those specs for $65?
However, to answer the question which you seem to be avoiding -
Is there any development of AmiTCP (other than making dial-up easier)?
Will you (or NSDi or whoever is doing it these days) actually be doing any
REAL development work on imporiving and updating AmiTCP for ALL users, or
are you just GUIfying the config? I mean, if you ARE turning AmiTCP into
just a dial-up stack (which can handle Ethernet if you want but we can't be
bothered to support that anymore) then please tell us all now so we can act
accordingly.
Will TCP/T be added?
NFS server (which NSDi have been muttering about for as long as I can
remember)?
Anything else?
WHEN?
TTFN,
Keith.
CW> <snicker>
Says it all really :((
--
Http://www.personal.u-net.com/~amiga Team AMIGA
"Climb Every Woman"
- Whitney Houston.
Holger Kruse
In article <4250.698...@betty.uchicago.edu>, Zap <z...@betty.uchicago.edu>
wrote:
> m(u)iami is wide open to a number of DOS and other attacks in its default
> config and is harder to fix than AmiTCP (that isnt open to any attacks in
> default config) and didnt I read that m(u)iami now supports a very crude
> packet filter looking for matching packet sizes from the same IP's?
Complete nonsense.
> and all
> this would do is free up some cpu time.. the incoming packets are still going
> to hit the users SanaII device.. need I say more?
No, but you should probably say less, since you don't seem to have any
idea what you are talking about...
Now here are, once again, the *facts*:
The commercial/registered versions of AmiTCP (I am referring to
4.3 throughout) and Miami both allow you to filter out packets and
create log entries according to certain criteria, using basically
the same kind of mechanism. Not much difference there, except that
Miami supports TCP and UDP, AmiTCP only TCP. The demo versions of
both stacks don't have this function.
*In addition* reg. Miami has an *optional* ping flood detection
heuristic. Obviously there is NO way to completely protect a machine
from ping floods (other than at the ISP side), so such a filter cannot
be perfect. It is not completely useless either though: the filter not
only reduces the CPU load, but also *informs* the user of ping flood
attacks (so he can e.g. reconnect to a different ISP port), and reduces
the load on the modem return channel. If you want to use that filter
in Miami - fine. If not, then just keep it disabled.
Once the packets reach the protocol stack (i.e. after filtering),
Miami and AmiTCP process them in the same way, with two exceptions,
regarding the way both stacks handle the two most common Internet
attacks recently:
- Miami contains the 4.4BSD bug fix to detect and eliminate invalid
IP fragments -- AmiTCP does not. This means Miami is invulnerable
to those popular Windows-95-ping attacks that recently crashed
thousands of commercial routers all over the Internet.
- Miami uses a cryptographically strong hash function to calculate
TCP sequence numbers, making TCP sequence number attacks (combined
with IP spoofing attacks) impossible. AmiTCP is theoretically
vulnerable to this kind of attack (although such an attack is
hardly possible in dial-up environments), if you are running
rshd/rlogind or similar programs. This kind of attack and its
correct workaround, as implemented in Miami is described in
RFC 1948.
AmiTCP has that famous fingerd security hole (which allows anyone
on the net to execute shell commands on the victim's machine), but
an easy fix for it has been posted here several times. Miami has
no such problem.
Miami complies with ALL the latest CERT advisories regarding Internet
security (up until February 97) and has NO known security holes. Period.
Now if you really know about any security problem in Miami that nobody
has found yet then please tell us. If not, sh*t the h*ll up, because
I am really getting tired and annoyed at those constant lies about
security issues, and I will have to consider any further such articles
to be libelous in nature.
Michael Brockman
>>Argl... I wish people would stop posting such lies.
>No lies.. just the truth
For the second time....Prove it. Not "I am all intelligent and know all"
attitude..PROVE IT! If it is so true, why not share it with the rest of us?
>>The truth is: Miami 2.0f in its *default* config complies with all
>>the latest CERT advisories against Internet attacks. In fact Miami
>>is the *only* TCP/IP stack for AmigaOS at the moment which is
>>invulnerable to TCP sequence number attacks *and* IP fragment
>>attacks (which are the most dangerous attacks these days), all
>>that without any changes to the default config.
>Bullshit.. and if you rely on CERT releases to protect your stack you will
>find yourself about 1 month behind the latest attacks
Again, we are waiting great knowledgable one.....So far all we hear is air..
>>Holger Kruse kr...@nordicglobal.com
>> http://www.nordicglobal.com
>Regards,
>Zap
>--
>z...@betty.uchicago.edu
>http://coming.soon/~zap
--
<sb>
<sb>
/>
/> Michael M Brockman, mi...@skylink.net
/>
Zap
>>buy the time you run the m(u)iamiInit program (waiting for MUI to load and
>>steal all of your memory) you could have finished editing the text files in
>>AmiTCP:db
>You forgot about the time spent getting the information from your ISP about
>DSN IP's, Your gateway IP, etc...The time spend running the Config_AmiTCP
IF your provider doesnt supply this info when you get an account with them
they aint worth using. (or they dont know it them selfs)
>to create your startnet script. Not to mention the time spent configuring and
>testing your SANA 2 configuration. With Miami all you need is the ISP's phone
SanaII config? last time I checked it was all done in the Config_AmiTCP setup
>number, your userid, and password. An average user doesn't understand all the
>insides of the TCP/IP setup. And there is no need for them to. I personally
>run AmiTCP, but I am very impressed with Miami ease-of-use and setup time.
Yes it is easy to setup
>You just seam to have a problem with MUI, not with Miami. By the way, Miami
>and MUI use 250K less memory than AmiTCP (On my system 750K for Miami vrs 1
>Meg for AmiTCP).
Yes I do have a prob with mui.. and yes I do have a prob with m(u)iami
>>GUI's arent needed to run a TCP Stack
>True, but to Administer a TCP stack, A gui really helps. Must be why Solaris
>(Sun's OS) has one, IRIX (SGI's OS) has one, etc...etc... I could keep going
I see you left out NT and Win95.. tho Im sure you know as well as I do that
running a real ISP from a NT/win box is beyond a joke. Sure Solaris, IRIX, etc
have GUI's.. but I bet the real ISP's dont use them for a waste of time that
they are.
>on with more examples. Most users prefer looking/controlling the stack
>through a GUI.
and why does the home user need to keep "looking/controlling the stack"?
Hands on is where it counts, if you know what you want changed its faster to
do it by hand then wait for a gui to load-open-etc
>>M(u)iami in its default config, is left wide open to attack.
>Please expand on this statement. This is a rumour stll not supported by any
>facts. Give us proof of what you are saying.
m(u)iami is wide open to a number of DOS and other attacks in its default
config and is harder to fix than AmiTCP (that isnt open to any attacks in
default config) and didnt I read that m(u)iami now supports a very crude
packet filter looking for matching packet sizes from the same IP's? and all
this would do is free up some cpu time.. the incoming packets are still going
to hit the users SanaII device.. need I say more?
>>if you need "extensive context-sensitive help" you shouldnt be touching any
>>of the settings.
>Why not? After all it is my machine we are talking about. One of the ways we
I wasnt refuring to your machine in my post.
>learn is by breaking things. First you argue that Miami is too complex...Then
>you berate Miami for giving all the information you need to configure your
>system. You can't have it both ways... So which one is it?
wasnt me
>>>Don't get me wrong - I have nothing personal against AmiTCP at all.
>>>But I would appreciate it if you would not post such obviously
>>>misleading half-truths about a competitor's product...
>I agree with Holger on this one...It is one thing to not like a product. It
>is crossing the line when you put down that product by being misleading and
>presenting opinion as fact.
Nothing I said is/was misleading
>>>--
>>>Holger Kruse kr...@nordicglobal.com
>>> http://www.nordicglobal.com
>>Regards,
>>Zap
>>--
>>z...@betty.uchicago.edu
>>http://coming.soon/~zap
>--
><sb>
><sb>
> />
> /> Michael M Brockman, mi...@skylink.net
> /> http://www.skylink.net/~mikeb
> |\_______{o}-------------------------------------------_
> [\\\\\\\\\\\{MIKEY BAD}>============================- >
> |/~~~~~~~{o}-------------------------------------------~
> \> Have you ever been cruising on the Information
> \> Superhighway, just to pause and take a whiz?
> \>
><sb>
><sb>PGP Key Available Upon Request
><sb>
Regards,
Zap
--
z...@betty.uchicago.edu
http://coming.soon/~zap
"I didn't do it... No one saw me do it... You can't prove anything..."
Bart Simpson.
Zap
>Argl... I wish people would stop posting such lies.
No lies.. just the truth
>The truth is: Miami 2.0f in its *default* config complies with all
>the latest CERT advisories against Internet attacks. In fact Miami
>is the *only* TCP/IP stack for AmigaOS at the moment which is
>invulnerable to TCP sequence number attacks *and* IP fragment
>attacks (which are the most dangerous attacks these days), all
>that without any changes to the default config.
Bullshit.. and if you rely on CERT releases to protect your stack you will
find yourself about 1 month behind the latest attacks
>I don't even want to mention that huge, gaping security hole in
>the fingerd of AmiTCP 4.3. Hopefully this has been fixed in 4.5
>though...
>--
>Holger Kruse kr...@nordicglobal.com
> http://www.nordicglobal.com
Regards,
Zap
--
z...@betty.uchicago.edu
http://coming.soon/~zap
"I didn't do it... No one saw me do it... You can't prove anything..."
Bart Simpson.
Chris Wiles
Hello,
> In article <33098B4C.MD-...@netcomuk.co.uk>, Chris Wiles wrote:
> >"Scared shitless" of what? NetConnect is a complete suite of commercial
> >software for around $65!
> >This includes a *full* TCP stack with GUI, VNG, Microdot-II, AmFTP, AmIRC,
> >AmTelnet, AmTerm, AmFinger etc.
> >How can I be "scared shitless" with those specs for $65?
>
> Because those programs are all shareware and available elsewhere?
1. They are *all* commercially licensed - NOT shareware.
> And because some aren't available at all?
2. What aren't available? They are all available, except for Microdot-II
which is on NetConnect and will be released within the next 2 weeks.
Chris Wiles
On Tue, 18 Feb 1997 15:10:42 Keith Blakemore-Noble said about Re: NSDi News?:
> Will you (or NSDi or whoever is doing it these days) actually be doing any
> REAL development work on imporiving and updating AmiTCP for ALL users, or
> are you just GUIfying the config? I mean, if you ARE turning AmiTCP into
> just a dial-up stack (which can handle Ethernet if you want but we can't be
> bothered to support that anymore) then please tell us all now so we can act
> accordingly.
Argh!! I will say this *once* - AmiTCP 4.5 contains changes that we, Active
Software, wanted for release within NetConnect:
1. GUI to control and manage AmiTCP (we created the GUI)
2. Preferences are now stored in env:/envarc:netconfig/
3. User/provider.conf added so the programs (Microdot-2 etc) can read the
user and ISP settings into its preferences.
4. A few other operational changes
5. Fixed fingerd security loophole etc
Therefore, and if you haven't already noticed, AmiTCP 4.5 Dialup is a full
TCP stack with some structural changes needed for our suite of commercially
licensed Internet software.
> Will TCP/T be added?
..which, incidentally, causes more problems than it is worth and is *not*
supported by Linux based servers (appox. 75% of the servers worldwide).
> NFS server (which NSDi have been muttering about for as long as I can
> remember)?
> Anything else?
> CW> <snicker>
>
> Says it all really :((
You are a fool quite frankly. I am not going to sit here and tell you
what is being developed as it is a commercial secret. There *will* be
improvements and changes to AmiTCP in the next few months (which will
most likely result in the release of v5), but I am not in a position
to inform people publically what these improvements will be.
We are producing a commercial suite of Internet software. I am here to
tell me people about what *we* are doing with NetConnect and am not in a
position to talk about what NSDi are planning.
Sorry.
Holger Kruse
In article <1126.6989...@betty.uchicago.edu>, Zap <z...@betty.uchicago.edu>
wrote:
> >The truth is: Miami 2.0f in its *default* config complies with all
> >the latest CERT advisories against Internet attacks. In fact Miami
> >is the *only* TCP/IP stack for AmigaOS at the moment which is
> >invulnerable to TCP sequence number attacks *and* IP fragment
> >attacks (which are the most dangerous attacks these days), all
> >that without any changes to the default config.
>
> Bullshit..
No, just the truth, as you should know, if you have any kind of
background knowledge about TCP/IP at all. If you don't believe me
then why don't you check for yourself ? It is easy enough to
do -- all you need is netstat, tcpdump and a C compiler.
> and if you rely on CERT releases to protect your stack you will
> find yourself about 1 month behind the latest attacks
If nothing else then this statement finally shows that you simply
have no clue... Do you actually know who/what CERT *is*, and what
a CERT advisory is ? Have a look at www.cert.org when you get the
chance.
CERT advisories often have a turn-around time of a few days, sometimes
even only a few hours, between the time a reproducable problem is
reported and the time major vendors start making patches available.
CERT advisories are based on reproducable, clearly defined security
problems, and the correct workaround/solution for them, not some
silly rumors about "DOS-based attacks". CERT is the ONLY widely
recognized, official authority on Internet security.
ALL major commercial TCP/IP stacks plus most Freeware stacks such
as FreeBSD and NetBSD use CERT advisories as a primary source of
information about Internet attacks and recommended fixes. Just last
week FreeBSD 2.1.6 was recalled from ftp.wustl.edu because of a
CERT advisory.
Btw, you still have not told us exactly what that alleged
"security hole" in Miami is. By responding to verifiable facts
with words like "bullshit" and by withholding "information" you
claim you have, you just lose what little credibility you have
left here.
Ian Craig Armitage
On 18-Feb-97 20:38:47, Zap Wrote :
>>You forgot about the time spent getting the information from your ISP about
>>DSN IP's, Your gateway IP, etc...The time spend running the Config_AmiTCP
>IF your provider doesnt supply this info when you get an account with them
>they aint worth using. (or they dont know it them selfs)
most providers supply the information but it means shit to a newbie, you
may as well be giving them gibberish!
>>You just seam to have a problem with MUI, not with Miami. By the way, Miami
>>and MUI use 250K less memory than AmiTCP (On my system 750K for Miami vrs 1
>>Meg for AmiTCP).
>Yes I do have a prob with mui.. and yes I do have a prob with m(u)iami
if you dont like mui, contact stefan stuntz. why tell people you hate it
as if it's gonna make them stop using it?
>>>GUI's arent needed to run a TCP Stack
>>True, but to Administer a TCP stack, A gui really helps. Must be why Solaris
>>(Sun's OS) has one, IRIX (SGI's OS) has one, etc...etc... I could keep going
>I see you left out NT and Win95.. tho Im sure you know as well as I do that
>running a real ISP from a NT/win box is beyond a joke. Sure Solaris, IRIX,
>etc have GUI's.. but I bet the real ISP's dont use them for a waste of time
>that they are.
Well, this is plain rediculous. GUI's are a waste of time?! well, tell
me, do you run your amiga from shell or workbench?!
heres a simple question even you should be able to answer....
give a complete newbie amitcp4.3 and miami.. which one will he get working
first?
>>on with more examples. Most users prefer looking/controlling the stack
>>through a GUI.
>and why does the home user need to keep "looking/controlling the stack"?
>Hands on is where it counts, if you know what you want changed its faster to
>do it by hand then wait for a gui to load-open-etc
plain bollocks! do you run mui from floppy or something?!
after miami has loaded, the mui library has already been read so at the
very least, all you do is de-iconify it.
>nothing i said is/was misleading
just dumb ;)
/---------------------------------------------------------------------\
| Craig Armitage | Developing the AmNetControl Internet Organiser! |
| -------------- | -------------------------------------------------- |
| 10mb A1200 | Latest version available from Aminet now! |
|---------------------------------------------------------------------|
| Email : ste...@netcomuk.co.uk |Aminet File Name : AmNet12b.lha|
| (put Craig or AmNet in the subject) | |
\---------------------------------------------------------------------/
©1997 crap sigs inc.
I think, therefore I'm overqualified !!!
Ian Craig Armitage
On 18-Feb-97 05:58:20, Chris Wiles Wrote :
>Hello,
>> I find it very suspicious that you are pulling up Miami, a topic of which
>> I don't think has ever appeared in this thread up until your mention of it.
>> I can only suspect that NSDi (and/or yourselves) have truly rested on your
>> laurels and now realize that other people are entering a market that you
>> either have lost or are about to lose... and are (if you pardon an American
>> phrase) "scared shitless."
>"Scared shitless" of what? NetConnect is a complete suite of commercial
>software for around $65!
>This includes a *full* TCP stack with GUI, VNG, Microdot-II, AmFTP, AmIRC,
>AmTelnet, AmTerm, AmFinger etc.
>How can I be "scared shitless" with those specs for $65?
><snicker>
Can you give us a definate day as to when this is released?
/---------------------------------------------------------------------\
| Craig Armitage | Developing the AmNetControl Internet Organiser! |
| -------------- | -------------------------------------------------- |
| 10mb A1200 | Latest version available from Aminet now! |
|---------------------------------------------------------------------|
| Email : ste...@netcomuk.co.uk |Aminet File Name : AmNet12b.lha|
| (put Craig or AmNet in the subject) | |
\---------------------------------------------------------------------/
©1997 crap sigs inc.
You know you've been spending too much time with computers and the 'net when
you haven't called your Mum in weeks because she hasn't got a modem.
Holger Kruse
In article <1902.698...@mail.usyd.edu.au>, Angus Mann
<am...@mail.usyd.edu.au> wrote:
> Which is *your* personal experience. In my experience, Holger has taken more
> than several days (usually over a week) to answer a message (if he does), and
> even then the reply hasn't always been helpful or polite. Actually, from
> memory
> it's never been either of those, and it's been the same case with other
> people
> I know.
I always respond to all direct questions or suggestions I receive by
email, usually the same day I get them, unless I am out of town or
otherwise unable to respond.
If you don't get any response from me please just resend the mail. It
is possible that my reply to you has bounced. I usually don't notice
that because the mailing lists I run typically cause >500 bounces every
day, so a single bounced (personal) mail is difficult to notice...
As for mails being helpful/polite: I guess that depends on the nature
of your mail more than on anything else...
Matthew Hunter
>software for around $65!
>This includes a *full* TCP stack with GUI, VNG, Microdot-II, AmFTP, AmIRC,
>AmTelnet, AmTerm, AmFinger etc.
>How can I be "scared shitless" with those specs for $65?
Because those programs are all shareware and available elsewhere?
And because some aren't available at all?
--
Chris Wiles
On 19 Feb 97 13:31:40 -0500 Ian Craig Armitage said about Re: NSDi News?:
> Can you give us a definate day as to when this is released?
Not definite day! 2 weeks from now. It will go to press on Friday.
Chris.
RHS Linux User
Angus Mann <am...@mail.usyd.edu.au> wrote on 19 Feb 97 02:37:10 +1100:
> >Holger has ALWAYS answered my questions with in a couple of
> >hours. Twice within three minutes, now thats service, that want I am
> >willing to pay for.
> Which is *your* personal experience. In my experience, Holger has taken more
> than several days (usually over a week) to answer a message (if he does), and
> even then the reply hasn't always been helpful or polite. Actually, from memory
Let's not turn this into a battle about the personalities of the
authors. I've had nothing but polite, helpful, and timely answers from
Holger. I haven't had any response at all from the AmiTCP authors, but
this doesn't bother me all that much, since the product works just fine
for me.
Jeff Gill
JeGill
But not registered.
Zap
>> config and is harder to fix than AmiTCP (that isnt open to any attacks in
>> default config) and didnt I read that m(u)iami now supports a very crude
>> packet filter looking for matching packet sizes from the same IP's?
>Complete nonsense.
I think not
>> and all
>> this would do is free up some cpu time.. the incoming packets are still
>> going to hit the users SanaII device.. need I say more?
>No, but you should probably say less, since you don't seem to have any
>idea what you are talking about...
dont I?
>Now here are, once again, the *facts*:
>The commercial/registered versions of AmiTCP (I am referring to
>4.3 throughout) and Miami both allow you to filter out packets and
>create log entries according to certain criteria, using basically
>the same kind of mechanism. Not much difference there, except that
>Miami supports TCP and UDP, AmiTCP only TCP. The demo versions of
AmiTCP does support UDP "filtering" along with all the normal protocols apart
from ICMP internal (stack) error returns.
>both stacks don't have this function.
>*In addition* reg. Miami has an *optional* ping flood detection
>heuristic. Obviously there is NO way to completely protect a machine
>from ping floods (other than at the ISP side), so such a filter cannot
there are ways to make it safer from the users end (modem)
>be perfect. It is not completely useless either though: the filter not
>only reduces the CPU load, but also *informs* the user of ping flood
>attacks (so he can e.g. reconnect to a different ISP port), and reduces
attack succesful.
>the load on the modem return channel. If you want to use that filter
>in Miami - fine. If not, then just keep it disabled.
ohh.. I wouldnt use m(u)iami if you paid me
>Once the packets reach the protocol stack (i.e. after filtering),
>Miami and AmiTCP process them in the same way, with two exceptions,
>regarding the way both stacks handle the two most common Internet
>attacks recently:
did you *read* the CERT releases?
>- Miami contains the 4.4BSD bug fix to detect and eliminate invalid
> IP fragments -- AmiTCP does not. This means Miami is invulnerable
> to those popular Windows-95-ping attacks that recently crashed
> thousands of commercial routers all over the Internet.
>- Miami uses a cryptographically strong hash function to calculate
> TCP sequence numbers, making TCP sequence number attacks (combined
> with IP spoofing attacks) impossible. AmiTCP is theoretically
Umm.. I see *u* know nothing about TCP sequence attacks.. or maybe I didnt
make myself clear, as I recall I said DOS (can we all say that?) that stands
for Denial Of Service. It wouldnt matter how you did your packet sequencing a
normal source IP spoofed TCP sequence DOS attack will still crash the stack
and if you are unlucky crash your machine.
> vulnerable to this kind of attack (although such an attack is
> hardly possible in dial-up environments), if you are running
oh yes it is possable and is happening right now.
> rshd/rlogind or similar programs. This kind of attack and its
> correct workaround, as implemented in Miami is described in
> RFC 1948.
>AmiTCP has that famous fingerd security hole (which allows anyone
>on the net to execute shell commands on the victim's machine), but
>an easy fix for it has been posted here several times. Miami has
>no such problem.
did you have to look that one up in the archives?
>Miami complies with ALL the latest CERT advisories regarding Internet
>security (up until February 97) and has NO known security holes. Period.
yes it does
>Now if you really know about any security problem in Miami that nobody
>has found yet then please tell us. If not, sh*t the h*ll up, because
>I am really getting tired and annoyed at those constant lies about
>security issues, and I will have to consider any further such articles
>to be libelous in nature.
"libelous in nature" hahahaaahaaa it cant be libelous if its just stating
FACT.
>Holger Kruse kr...@nordicglobal.com
> http://www.nordicglobal.com
Regards,
RHS Linux User
Chris Wiles <act...@netcomuk.co.uk> wrote on 19 Feb 97 11:13:32 +0000:
> > In article <33098B4C.MD-...@netcomuk.co.uk>, Chris Wiles wrote:
> > >"Scared shitless" of what? NetConnect is a complete suite of commercial
> > >software for around $65!
> > >This includes a *full* TCP stack with GUI, VNG, Microdot-II, AmFTP, AmIRC,
> > >AmTelnet, AmTerm, AmFinger etc.
> > >How can I be "scared shitless" with those specs for $65?
> > Because those programs are all shareware and available elsewhere?
"pre-registered shareware", then.
> > And because some aren't available at all?
> 2. What aren't available? They are all available, except for Microdot-II
> which is on NetConnect and will be released within the next 2 weeks.
I was referring to MicroDot II specifically... and if it is
released within two weeks, I will be quite happy. :)
Unfortunately, having already registered the programs in that
package that I use, along with havign already bought AmiTCP, I still have
little use for this whole NetConnect thing. :)
Jeff Grimmett
In a message dated Wed 19 Feb 97 20:05, Angus Mann <am...@mail.usyd.edu.au>
wrote:
AM> There's usually more than just that to configuring Miami, however.
AM> Nobody I speak to regularly has seen Miami automatically configure
their
AM> entire setup for them, instead they've had to enter some (often
several)
AM> details themselves.
Well, allow me then to be the first to render that statement false,
although I suppose you can weasel out and claim that you don't speak to me
regularly :-)
I have had the pleasure of configuring Miami on around a dozen different
ISPs at this point, from Red Hat Linux to NT setups, and there were only
two occasions where I had to enter more than the phone number:
-- On the Red Hat system, I had to enter the DNS addresses -- which isn't
*required* for operation.
-- On a Wildcat! system, I had to coax MiamiInit past the fidonet EMSI
prompt, after which everything went like clockwork.
I have ZERO past experience setting up TCP/IP stacks. Not bad, says I.
>Holger has ALWAYS answered my questions with in a couple of
>hours. Twice within three minutes, now thats service, that want I am
>willing to pay for.
AM> Which is *your* personal experience. In my experience, Holger has
AM> taken more than several days (usually over a week) to answer a message
AM> (if he does), and even then the reply hasn't always been helpful or
AM> polite. Actually, from memory it's never been either of those, and
it's
AM> been the same case with other people I know.
I suppose it depends if you act like an asshole in the process ... if this
message and those leading up to it are examples, I would have a hard time
being civil to you myself. Sorry, but if you give someone a ration of
shit, expect a return shipment, possibly by express mail.
For my own part, I have always gotten prompt and courteous responses from
Holger, and I'm not even a registered user (yet -- spare cash is not a
common commodity around here, but soon...).
-- Via DLG Pro v1.15
/* Jeff Grimmett (Jeff_G...@elric.maximumaccess.com)
** DLG Development (http://www.ald.net/dlg)
** The home of DLG Pro (Fido: 1:202/720.0 -- (619) 549 7742)
****************************************************************/
Paul Copsey
On 19 Feb 1997 11:38:20, Chris Wiles <act...@netcomuk.co.uk> wrote:
> On Tue, 18 Feb 1997 15:10:42 Keith Blakemore-Noble said about Re: NSDi News?:
>
> > Will you (or NSDi or whoever is doing it these days) actually be doing any
> > REAL development work on imporiving and updating AmiTCP for ALL users, or
> > are you just GUIfying the config? I mean, if you ARE turning AmiTCP into
> Argh!! I will say this *once* - AmiTCP 4.5 contains changes that we, Active
> Software, wanted for release within NetConnect:
[snip]
> 4. A few other operational changes
>
> 5. Fixed fingerd security loophole etc
4 & 5 are very important though, especially to people using AmiTCP in
a commercial situation
> Therefore, and if you haven't already noticed, AmiTCP 4.5 Dialup is a full
> TCP stack with some structural changes needed for our suite of commercially
> licensed Internet software.
Why haven't people who've already paid for earlier versions been
offered it then?
> > Will TCP/T be added?
>
> .which, incidentally, causes more problems than it is worth and is *not*
> supported by Linux based servers (appox. 75% of the servers worldwide).
75%? You are doing Linux a big favour by suggesting that.
> You are a fool quite frankly. I am not going to sit here and tell you
> what is being developed as it is a commercial secret. There *will* be
No, Keith is not a fool, he is simply aiming criticism at someone who
should be able to give answers. NSDi are well known for being hard to
get hold of, and since you do have AmiTCP 4.5 presumably, it seems
only fair that since you are acting as an agent for NSDi, if you don't
have the answer as to why only you have 4.5, which does fix at least
one very important problem (one that has been around for a very long
time as well), then you shoud be able to raise someone who does.
Paul
Allen Wittenauer
Chris Wiles wrote in comp.sys.amiga.networking about "Re: NSDi News?":
> "Scared shitless" of what? NetConnect is a complete suite of commercial
> software for around $65!
>
> This includes a *full* TCP stack with GUI, VNG, Microdot-II, AmFTP,
> AmIRC, AmTelnet, AmTerm, AmFinger etc.
>
> How can I be "scared shitless" with those specs for $65?
>
...because I don't want *any* of that stuff... I _already_ have those
things or their equivalents. I want T/TCP, an lpr.device, NFS server,
IPv6, and a decent routing mechanism, amongst other things.
That's how.
By changing this thing into a setup for Joe User you are almost guaranting
that you are going to lose out on the people who have full blown networks
that run AmiTCP 24/7.
Hmm... what was the address again? http://www.iworks.com/? I'm sure
they'll appreciate our business since apparantly NSDi and yourselves don't.
Keith Blakemore-Noble
On or around Wed, 19 Feb 97, Chris Wiles is alledged to have muttered
something along the lines of -
CW> On Tue, 18 Feb 1997 15:10:42 Keith Blakemore-Noble said about Re: NSDi News?:
CW> > CW> <snicker>
CW> >
CW> > Says it all really :((
CW>
CW> You are a fool quite frankly.
A fool? For buying 4.3, or for wondering why 4.5 is not being offered to
those who bought 4.3, or for wondering if there will ever be any further
development of 4.3 (given that there has been no sign of such development
for quite some time now :(()? Which is it?
CW> There *will* be
CW> improvements and changes to AmiTCP in the next few months (which will
CW> most likely result in the release of v5), but I am not in a position
CW> to inform people publically what these improvements will be.
Can we assume from your replies that TCP/T is NOT being developed, however?
And presumably no NFS server (announcing that would hardly constitue a
commercial secret - especially as this wouldonly be catching up with INet225
in that respect)?
CW> We are producing a commercial suite of Internet software. I am here to
CW> tell me people about what *we* are doing with NetConnect and am not in a
CW> position to talk about what NSDi are planning.
So who DO we contact to find out? Don't say NSDi, as everyone knows they
are very hard to get hold of these days, which is why people are assuming
YOU are a suitable source of info - you seem to be the only people with 4.5,
for a start!
TTFN,
Keith.
--
Http://www.personal.u-net.com/~amiga Team AMIGA
"If I didn't know Stuart better, I'd say their sexual partner shouted abuse
at Philip Schofield!", moaned Andy Poslthwaite-Mannwering when visiting Kim
Basinger's right nostril in 1891
Holger Kruse
In article <330AE62C.MD-...@netcomuk.co.uk>, Chris Wiles> ..which, incidentally, causes more problems than it is worth and is *not*
> supported by Linux based servers (appox. 75% of the servers worldwide).
That's not exactly true, it really depends on how you implement it :-).
You are right in that older versions of Linux (1.x) have problems with
the TCP packet layout used by T/TCP. Linux 2.x does not directly support
T/TCP, but does not cause any problems either. (After all, T/TCP *is*
fully compatible with TCP, except if the TCP implementation of the
peer was badly broken in the first place, as in Linux 1.x)
Anyway, workarounds for such interoperability problems are possible
(in my code, transparent for users): Miami 2.0f now happily
interoperates with all known hosts, including all Linux versions,
even with T/TCP enabled in Miami, so in Miami 2.0f T/TCP does NOT
cause any problems or restrictions for the user -- quite the
contrary.
With FreeBSD 2.x supporting T/TCP, WWW servers now tend to migrate
towards T/TCP, because of the considerable memory and cpu time savings
(no more long TIME_WAIT states and huge connection tables), and the
much higher number of transactions per second, because the average
number of packets per (small) www request drops from 9 to 3 with
T/TCP. Just look at www.yahoo.com, an extremely busy server with
T/TCP support.
T/TCP support is also available for SunOS 4.1.x, and has been announced
by several commercial vendors for future versions of their stacks.
I guess for NSDi the bigger problem (and bigger amount of work) would
be to first migrate the AmiTCP kernel from 4.3BSD/Reno to 4.4BSD or
FreeBSD :-). I doubt that the current older kernel (without the
required radix routing trees or route cloning) could even be adapted
for T/TCP without completely rewriting it...
--
Neil Bothwick
Chris Wiles said,
>> >This includes a *full* TCP stack with GUI, VNG, Microdot-II, AmFTP, AmIRC,
>> >AmTelnet, AmTerm, AmFinger etc. How can I be "scared shitless" with those
>> >specs for $65?
>>
>> Because those programs are all shareware and available elsewhere?
> 1. They are *all* commercially licensed - NOT shareware.
Are they full shareware versions, that can be updated like normal
registered copies? Or are they CD 'specials'.
Neil
--
Neil Bothwick, Warrington, England - CU Amiga CD compiler
Connected via Wirenet - The UK's first Amiga-only internet access provider
mailto:in...@wirenet.u-net.com WWW: http://www.wirenet.u-net.com
--
Hi, I'm not a signature virus. Why don't you just copy me into your signature?
Jurjen Oskam
In a message of 18 Feb 97 (15:10:42) Keith Blakemore-Noble wrote to All: (about
"NSDi News?")
Hi Keith!
KBN> Will TCP/T be added?
What *is* TCP/T really? Yeah, it's for faster web browsing, but how, where,
why, when, who? If you catch my drift.. ;-)
--
Jurjen Oskam -=*Finger jos...@worldonline.nl for PGP public key *=-
Key fingerprint = E3 9F E7 5C 16 3B D2 B2 4F 86 BA 66 96 0C 44 F1
... This program doesn't work with DOS 1.0 or higher.
Holger Kruse
In article <4359.6989...@betty.uchicago.edu>, Zap
<z...@betty.uchicago.edu> wrote:
> >*In addition* reg. Miami has an *optional* ping flood detection
> >heuristic. Obviously there is NO way to completely protect a machine
> >from ping floods (other than at the ISP side), so such a filter cannot
>
> there are ways to make it safer from the users end (modem)
You cannot do more than filter incoming packets, optimize the response
of the stack (e.g. by inhibiting replies), and inform the user. There
is no way to prevent the other side from sending more packets. For
ping flooding the *modem line* is typically the resource being denied,
not anything in your stack.
> >be perfect. It is not completely useless either though: the filter not
> >only reduces the CPU load, but also *informs* the user of ping flood
> >attacks (so he can e.g. reconnect to a different ISP port), and reduces
>
> attack succesful.
Sure. The attack is *by definition* successful the moment the other
side starts flood-pinging you, no matter what you do. No protection
from "denial of service" on the modem line is possible in your stack.
Should be quite obvious :-)
> Umm.. I see *u* know nothing about TCP sequence attacks.. or maybe I didnt
> make myself clear, as I recall I said DOS (can we all say that?) that stands
> for Denial Of Service. It wouldnt matter how you did your packet sequencing a
> normal source IP spoofed TCP sequence DOS attack will still crash the stack
> and if you are unlucky crash your machine.
No, it won't.
You are mixing up two different issues here. The TCP sequence attack I
was referring to is an attack where someone actually *breaks in* to
your machine and gets root access through rlogind/rshd, i.e. *not*
just a DOS attack, but the attack described in RFC 1948. *This* attack
heavily relies on timing issues and properties of the TCP sequence
number generator. Miami has full protection from it, and *this* attack
is hardly possible on dial-up lines because of timing problems for the
attacker.
There is no way IP spoofing by itself can harm a stack in any way
(cause it to crash etc.). This is only possible if IP spoofing is
combined with some other kind of attack (such as TCP SYN flooding),
and in this case a possible crash could only come from memory
exhaustion. However because of the way Miami's memory management works
it is not possible for Miami to crash simply because of a large number
of connections in SYN_RCVD, so Miami will NOT crash as a result of such
an attack.
If you are worried about the DOS effects from that kind of attack
(because of the limited listen backlog): if you read the CERT
recommendation (and/or apply some common sense :-)) you will see that
NO safe protection is possible. You can only balance one problem against
another one. Miami fully complies with those methods of protection
that *are* possible and recommended by CERT (such as allowing
customization of the listen backlog, automatically filtering out
obviously malformed packets, automatically filtering out packets
arriving from obviously incorrect IP address, and allowing the user
to filter out additional packets, on demand).
Besides, with dial-up lines the DOS effects on the modem line are
much more pronounced than in the stack, so if you are really worried
about DOS then the first thing you should do is ask your provider
to install appropriate filters, so your modem line is not that easy
to flood. DOS effects on listen sockets *behind* a modem dial-up line
are basically inexistant in such a setup anyway.
Personally I am much more worried about actual break-in attacks than
about DOS attacks. With DOS you can just reconnect and no harm is done.
With the kind of break-in attack that has been successfully mounted
against AmiTCP's fingerd you may find your harddisk formatted one
day...
> >AmiTCP has that famous fingerd security hole (which allows anyone
> >on the net to execute shell commands on the victim's machine), but
> >an easy fix for it has been posted here several times. Miami has
> >no such problem.
>
> did you have to look that one up in the archives?
??? I am talking about AmiTCP 4.3. Its fingerd is broken in the
*current* version, *right now*, but a few replacement scripts/programs
exist on Aminet.
> >Miami complies with ALL the latest CERT advisories regarding Internet
> >security (up until February 97) and has NO known security holes. Period.
>
> yes it does
Once again: WHICH ONES ? Exactly WHICH attack did you use, and WHICH
version of Miami did you SEE crash or otherwise behave incorrectly ?
Look, I have the sources of Miami, and *know* how the program works.
You are apparently just guessing, and incorrectly generalizing results
from attack patterns you have heard about somewhere. How can you
possibly claim you have found a vulnerability in one *particular*
stack if you have no proof whatsoever for it ?
If you have found an actual, verifiable vulnerability then finally
tell us exactly what it is, and stop confusing people with your
incorrect guesses.
> >Now if you really know about any security problem in Miami that nobody
> >has found yet then please tell us. If not, sh*t the h*ll up, because
> >I am really getting tired and annoyed at those constant lies about
> >security issues, and I will have to consider any further such articles
> >to be libelous in nature.
>
> "libelous in nature" hahahaaahaaa it cant be libelous if its just stating
> FACT.
The point is that you are NOT stating facts, but incorrect guesses which
you disguise as facts. So far you have not provided a single conclusive
point.
Holger Kruse
In article <330c...@usura.worldonline.nl>, Jurjen Oskam
<jos...@usura.worldonline.nl> wrote:
> KBN> Will TCP/T be added?
>
> What *is* TCP/T really? Yeah, it's for faster web browsing, but how, where,
> why, when, who? If you catch my drift.. ;-)
"T/TCP" (not "TCP/T", even though it means "TCP for Transactions") is a
compatible extension to TCP specifically designed for "transaction-
oriented" (i.e. short-lived) TCP connections, like downloads of short
web pages, identd lookups, finger lookups etc.
With standard TCP such a download usually requires the exchange of 9
packets. With T/TCP only 3 packets are necessary, if both sides support
T/TCP, i.e. you can get a dramatic reduction of traffic on your modem
line, in particular if the amount of data being transferred per
connection is relatively small.
Other benefits from T/TCP compared to TCP are that it caches connection
parameters (such as latency, MTU etc.) across connections, further
improving performance, and that T/TCP considerably shortens TIME_WAIT
states of closed connections, reducing the memory load and cpu load
on your system.
T/TCP is currently supported by some web servers, including Yahoo.
Client support exists in VoyagerNG, if used together with Miami >=2.0.
Other platforms that support T/TCP include FreeBSD 2.x, some
versions of SunOS 4.1.x, some versions of BSD/OS, and Miami 2.x :-).
Several other commercial implementations have been announced.
Some older and/or buggy systems (including some versions of Linux)
cannot properly handle T/TCP packets. Miami contains workarounds for
such problems though, allowing the user to keep T/TCP enabled at
all times.
For more information have a look at "TCP/IP Illustrated, Volume 3"
by W. Richard Stevens, Addison Wesley, 1996. Some (not quite up-to-date)
information can also be found on Stevens' home page:
www.noao.edu/~rstevens
The detailed specs of T/TCP and some background information/motivation
is given in RFC 1379 and RFC 1644.
Chris Wiles
Hiya Neil!
On 20 Feb 97 12:50:31 +0000 Neil Bothwick said about Re: NSDi News?:
> Chris Wiles said,
>
> >> >This includes a *full* TCP stack with GUI, VNG, Microdot-II, AmFTP, AmIRC,
> >> >AmTelnet, AmTerm, AmFinger etc. How can I be "scared shitless" with those
> >> >specs for $65?
> >>
> >> Because those programs are all shareware and available elsewhere?
>
> > 1. They are *all* commercially licensed - NOT shareware.
>
> Are they full shareware versions, that can be updated like normal
> registered copies? Or are they CD 'specials'.
Not specials, not shareware. We have commercially licensed the software
for NetConnect. So, if a use bought NC they would get the full VNG,
Microdot-II etc.
.the royalies are *huge* BTW!
Yes, expect updates as normal :)
Chris Wiles
On 19 Feb 97 18:08:32 +0500 Jeff Gill said about Re: NSDi News?:
>
> > Because those programs are all shareware and available elsewhere?
> >And because some aren't available at all?
>
>
> JeGill
> But not registered.
Pardon? What *do* people mean? Can I ask something? Do people not
understand what "commercially licensed" software means??
Yes they are as if you registered the software. You *will* get upgrades
and fixes as normal. There is very little difference.
The only difference is that YOU haven't registered them, we have
done that for you.