levy@ttrdc.UUCP (Daniel R. Levy) (05/31/87)
In article <16906@amdcad.AMD.COM>, phil@amdcad.AMD.COM (Phil Ngai) writes: < Speaking of terminals, I note that PC clones with floppy only drives, < monitor, and DOS can be gotten in the $600 range, which is less than < we pay for DEC VT-220s. < < Why would you want a PC instead of a terminal? Well, in addition to < the fact that it can be cheaper (particularly if you want graphics) < the IBM monitor makes really nice characters and the IBM keyboard < feels VERY nice. I don't know how well the clones do in this area. Some (all?) PC/clones cannot keep up with speeds > 4800 baud in the VT100 terminal emulator (they lose bursts of characters, even with Xon-Xoff flow control). You want a PC, you buy a PC. You want a terminal, well, I hope you get an AT&T terminal (plug plug :-) ). -- |------------dan levy------------| Path: ..!{akgua,homxb,ihnp4,ltuxa,mvuxa, | an engihacker @ | vax135}!ttrdc!ttrda!levy | at&t data systems division | Disclaimer: try datclaimer. |--------skokie, illinois--------|
robert@gitpyr.gatech.EDU (Robert Viduya) (06/01/87)
>levy@ttrdc.UUCP (Daniel R. Levy) (levy@ttrdc.UUCP, <1731@ttrdc.UUCP>): > Some (all?) PC/clones cannot keep up with speeds > 4800 baud in the VT100 > terminal emulator (they lose bursts of characters, even with Xon-Xoff flow > control). This is just plain wrong. I've written a comm program (vt100) that gets about 9000 baud max throughput on a standard 4Mhz 8088 PC (running at 9600 baud occasionally requires flow control, 7200 baud doesn't need any). This program is written in 90% C and 10% assembly. If it was entirely in assembly it could probably run at over 15000 baud. Since most clones are faster than a PC (8Mhz or a V20 or 8086 or various combinations) they ought to be able to handle true 9600 baud with no flow control. And I when I say true 9600 baud, I mean characters being pumped over the line at the maximum 960 cps sustained (assuming 8 bits/1 start/1 stop), which you might call "bursts". > You want a PC, you buy a PC. You want a terminal, well, I hope > you get an AT&T terminal (plug plug :-) ). Based on my experience with the AT&T 4425, I wouldn't recommend an AT&T terminal if speed was a major criterion. Now I don't have any experience with any of the other terminals that AT&T sells (save for the 5620, which I'm not sure really counts), but the 4425 is the slowest VT100 emulating terminal I've ever used. I've not timed it, but I don't think it'll handle greater than 4800 baud (certainly not 7200 baud). Don't get me wrong, it's not a bad terminal in most other respects. robert -- Robert Viduya robert@pyr.gatech.edu Office of Computing Services Georgia Institute of Technology (404) 894-4660 Atlanta, Georgia 30332
john@viper.Lynx.MN.ORG (John Stanley) (06/01/87)
In article <1731@ttrdc.UUCP> levy@ttrdc.UUCP (Daniel R. Levy) writes: >In article <16906@amdcad.AMD.COM>, phil@amdcad.AMD.COM (Phil Ngai) writes: >< Speaking of terminals, I note that PC clones with floppy only drives, >< monitor, and DOS can be gotten in the $600 range, which is less than >< we pay for DEC VT-220s. >< >< Why would you want a PC instead of a terminal? Well, in addition to >< the fact that it can be cheaper (particularly if you want graphics) >< the IBM monitor makes really nice characters and the IBM keyboard >< feels VERY nice. I don't know how well the clones do in this area. > >Some (all?) PC/clones cannot keep up with speeds > 4800 baud in the VT100 >terminal emulator (they lose bursts of characters, even with Xon-Xoff flow >control). You want a PC, you buy a PC. You want a terminal, well, I hope >you get an AT&T terminal (plug plug :-) ). Heck, why not just pick up an Atari ST? You can get a very-very sharp clear monochrome monitor, 1meg Ram, 1 720-800k 3.5" disk drive and a nice keyboard (better in-my-opinion by a long shot than most of the PC clones) for around $550. You can easily pick up good PD terminal software that doesn't have dropped-char problems even as fast as 19.2Kbaud for the cost of a few phone calls. I use one for my primary terminal in 50-line vt100 mode and it works like a gem.. (pun intended :) I don't want to start any sys.flame.wars I'm just pointing out a good alternative to the one-track mindset that says it's gotta be a personal clone... Why buy a 256-640k 8086 based machine for $600 when you can get a nicer keyboard with a good monitor and a higher density disk drive on a one megabyte 68000 system (that has enough speed to handle high speeds -and- lots of features without dropping chars) for less money...? --- John Stanley (john@viper.UUCP) Software Consultant - DynaSoft Systems UUCP: ...{amdahl,ihnp4,rutgers}!{meccts,dayton}!viper!john
mikel@codas.UUCP (06/02/87)
>> Why would you want a PC instead of a terminal? Well, in addition to >> the fact that it can be cheaper (particularly if you want graphics) >> the IBM monitor makes really nice characters and the IBM keyboard >> feels VERY nice. I don't know how well the clones do in this area. > > Some (all?) PC/clones cannot keep up with speeds > 4800 baud in the VT100 > terminal emulator (they lose bursts of characters, even with Xon-Xoff flow > control). You want a PC, you buy a PC. You want a terminal, well, I hope > you get an AT&T terminal (plug plug :-) ). There is a good reason why I use my Pc (Pc7300 upgraded to a 3B1) instead of the AT&T 4425 sitting beside it. The 4425's max baud rate is 19200, however our data switch can't handle more than 9600. The 3B1 is on our StarLan. I can open up a window whatever size I want, and using my "rlogin" program (forget "cu" [ugh] over StarLan!) talk to all of our other computers at well over 19200 baud. You should see the window refresh at this speed on the Unix Pc! Not to mention that I can have windows up to over half a dozen machines at the same time. Of course I can run windows on the 4425 too, but it's not as nice as the Pc, and neither come close to a 5620 DMD (which is not *just* a terminal - btw). -- Mikel Manitius @ AT&T-IS mikel@codas.att.com.uucp Copyright 1987. Redistribution via Stargate PROHIBITED
caf@omen.UUCP (06/02/87)
In article <1731@ttrdc.UUCP> levy@ttrdc.UUCP (Daniel R. Levy) writes:
:Some (all?) PC/clones cannot keep up with speeds > 4800 baud in the VT100
:terminal emulator (they lose bursts of characters, even with Xon-Xoff flow
:control). You want a PC, you buy a PC. You want a terminal, well, I hope
:you get an AT&T terminal (plug plug :-) ).
:--
That also depends on what software you're running. For example, DOS 3.2
causes some problems at 19200 bps that don't happen with DOS 2.1. Some
memory resident programs and device drivers cause problems with high speed
communications.
But the biggest problem may be in the terminal emulation software. Many
programs run into trouble above 1200 or so.
On the other hand, ZCOMM (shareware) or Pro-YAM can handle 9600 or 19200
bps without dropping characters. They can even handle EDT or the DEC Store
demo at those speeds. They can handle bursts of >20k at these speeds, and
assert flow control correctly for sustained output.
Chuck Forsberg WA7KGX Author of Pro-YAM communications Tools for PCDOS and Unix
...!tektronix!reed!omen!caf Omen Technology Inc "The High Reliability Software"
17505-V Northwest Sauvie Island Road Portland OR 97231 Voice: 503-621-3406
TeleGodzilla BBS: 621-3746 2400/1200 CIS:70007,2304 Genie:CAF Source:TCE022
omen Any ACU 1200 1-503-621-3746 se:--se: link ord: Giznoid in:--in: uucp
omen!/usr/spool/uucppublic/FILES lists all uucp-able files, updated hourlydc@sdd.UUCP (06/03/87)
An AT&T terminal !!!! UGH!! UGH!!! I have an AT&T 610 terminal that I wish I didn't!!! -dc
elg@killer.UUCP (06/05/87)
in article <1071@viper.Lynx.MN.ORG>, john@viper.Lynx.MN.ORG (John Stanley) says: > > In article <1731@ttrdc.UUCP> levy@ttrdc.UUCP (Daniel R. Levy) writes: > >In article <16906@amdcad.AMD.COM>, phil@amdcad.AMD.COM (Phil Ngai) writes: > >< Speaking of terminals, I note that PC clones with floppy only drives, > >< monitor, and DOS can be gotten in the $600 range, which is less than > >< we pay for DEC VT-220s. > Heck, why not just pick up an Atari ST? You can get a very-very sharp > clear monochrome monitor, 1meg Ram, 1 720-800k 3.5" disk drive and a nice > keyboard (better in-my-opinion by a long shot than most of the PC clones) > for around $550. NICE KEYBOARD? Welllll.... Right now I am using a Commodore 128 with a VT100 emulator. The C-128 keyboard fixes all the atrocities of previous Commodore keyboards -- it has a full numeric pad, keys for almost all ASCII characters (the only ones missing are the little curly braces), and decent if mushy feel. The only problem: it's built into the system console. And believe me, a full keyboard is BIG. and CLUMSY. Where with a IBM clone I can move the keybord around to relieve fatigue, or lean back with it on my lap, or otherwise enjoy myself, with the C-128, it just SITS there (and I can't even move the thing, because I have all sorts of quirky hardwar plugged into the expansion ports). And putting it on a typewriter stand would be out of the question because of its size and all the cables exiting from various directions, so it sits up there higher than my lap -- right where typing teachers tell you you're NOT supposed to have your keyboard. My next computer is NOT going to have an all-in-one keyboard... life is too short for that kind of frustration. Now, I guess if you're using the Religious Crusade method of typing ("seek, and thou shalt find"), it's not too bad, but to a touch-typist.... BTW, this is NOT an endorsement of the C-128 -- its maximum baud rate is 2400 baud (and that AT&T dude thought that IBM equipment was slow!). -- Eric Green elg%usl.CSNET CS student, University of SW Louisiana {cbosgd,ihnp4}!killer!elg Apprentice Haquer, Bayou Telecommunications Snail Mail P.O. Box 92191 BBS phone #: 318-984-3854 300/1200 baud Lafayette, LA 70509 I disclaim my existence, and yours, too.
ken@argus.UUCP (Kenneth Ng) (06/09/87)
In article <1731@ttrdc.UUCP>, levy@ttrdc.UUCP (Daniel R. Levy) writes: > Some (all?) PC/clones cannot keep up with speeds > 4800 baud in the VT100 > terminal emulator (they lose bursts of characters, even with Xon-Xoff flow > control). You want a PC, you buy a PC. You want a terminal, well, I hope > you get an AT&T terminal (plug plug :-) ). Don't be too sure about that. The AT&T 4425 has a hard time keeping up with speeds beyond 2400 baud for extended periods of time unless the X-ON, X-OFF protocol is enabled. BUT, that makes the terminal HANG on one of our half duplex systems. The only way out is a terminal reset, which drops the line, and signs me off. Conversely I often use a Phillips IBM PC (believe it or not *SLOWER* than the original IBM) with Procomm at 9600 baud with very few problems. I guess the key is design implementation. < Fodder for INEWS> < < < -- Kenneth Ng: Post office: NJIT - CCCC, Newark New Jersey 07102 uucp !ihnp4!allegra!bellcore!argus!ken *** NOT ken@bellcore.uucp *** bitnet(prefered) ken@orion.bitnet
katzung@laidbak.UUCP (Brian Katzung) (06/13/87)
- From: levy@ttrdc.UUCP (Daniel R. Levy)
- Some (all?) PC/clones cannot keep up with speeds > 4800 baud in the VT100
- terminal emulator (they lose bursts of characters, even with Xon-Xoff flow
- control). You want a PC, you buy a PC. You want a terminal, well, I hope
- you get an AT&T terminal (plug plug :-) ).
- --
- |------------dan levy------------| Path: ..!{akgua,homxb,ihnp4,ltuxa,mvuxa,
- | an engihacker @ | vax135}!ttrdc!ttrda!levy
- | at&t data systems division | Disclaimer: try datclaimer.
- |--------skokie, illinois--------|
It sounds as if you're suggesting that any AT&T terminal will keep up at
speeds greater than 4800 baud! I have news for you: not all of them do.
The one I have in mind has no host flow control.
If you want something that really works, I hope you test it before you buy
it (or can get a refund), or forever hold your PC!
Brian Katzung
(A contractor at AT&T Information Systems, Naperville, Illinois)
The comments expressed here are my own, and do not necessarily reflect the
views of my employer or my client.davev@spked.UUCP (06/14/87)
In article <1075@laidbak.UUCP>, katzung@laidbak.UUCP (Brian Katzung) writes:
] - From: levy@ttrdc.UUCP (Daniel R. Levy)
] - Some (all?) PC/clones cannot keep up with speeds > 4800 baud in the VT100
] - terminal emulator (they lose bursts of characters, even with Xon-Xoff flow
] - control). You want a PC, you buy a PC. You want a terminal, well, I hope
] - you get an AT&T terminal (plug plug :-) ).
] - --
] - | dan Levy | vax135}!ttrdc!ttrda!levy
]
] It sounds as if you're suggesting that any AT&T terminal will keep up at
] speeds greater than 4800 baud! I have news for you: not all of them do.
] The one I have in mind has no host flow control.
] Brian Katzung
Some terminals will keep up with 4800, or 9600, and a Wyse 75 keeps up pretty
darn well at 19200. BTW, the AT&T terminals I've seen don't keep up
well, but I don't have exact figures.
--
The contents of this message are totally unauthorized, and represent no person
or entity within any agency, nor any statement of policy.
Standard Form 1 Disclaimer (Rev. 4-87)
{{seismo|ihnp4!}lll-crg|sdcsvax|{decvax!}ucbvax}!ucdavis!spked!davevsmvorkoetter@watmum.UUCP (06/14/87)
In article <94@spked.UUCP> davev@spked.UUCP (Systems Integrator van De Kerk) writes: ] In article <1075@laidbak.UUCP>, katzung@laidbak.UUCP (Brian Katzung) writes: ] ] ] From: levy@ttrdc.UUCP (Daniel R. Levy) ] ] ] Some (all?) PC/clones cannot keep up with speeds > 4800 baud in the VT100 ] ] ] terminal emulator (they lose bursts of characters, even with Xon-Xoff flow ] ] ] control). You want a PC, you buy a PC. You want a terminal, well, I hope ] ] ] you get an AT&T terminal (plug plug :-) ). ] ] It sounds as if you're suggesting that any AT&T terminal will keep up at ] ] speeds greater than 4800 baud! I have news for you: not all of them do. ] ] The one I have in mind has no host flow control. ] ] Brian Katzung ] Some terminals will keep up with 4800, or 9600, and a Wyse 75 keeps up pretty ] darn well at 19200. BTW, the AT&T terminals I've seen don't keep up ] well, but I don't have exact figures. I am using an AT Clone, running a VT102 emulation package (Terminal Emulator and File Transfer 2.40a), and it is having no trouble at 19200 baud, using only XON/XOFF (no hardware flow control). I am connected to a Sytek box by 50 feet of 4 conductor telephone wire from Radio Shack. BTW, "vi" really flies at 19.2KB.
sl@van-bc.UUCP (Stuart Lynne) (06/15/87)
In article <1017@watmum.UUCP> smvorkoetter@watmum.UUCP (Stefan M. Vorkoetter) writes: >] ] ] Some (all?) PC/clones cannot keep up with speeds > 4800 baud in the VT100 >] ] ] terminal emulator (they lose bursts of characters, even with Xon-Xoff flow >I am using an AT Clone, running a VT102 emulation package (Terminal Emulator >and File Transfer 2.40a), and it is having no trouble at 19200 baud, using >only XON/XOFF (no hardware flow control). I am connected to a Sytek box by I have a desk accessory for my Mac which keeps up to 19.2 kbps without using any flow control. Reasonable vt100 compatibility too. Enough to run vi. Problem is most small unix boxen don't quite cut the mustard. They don't really send out a full 19.2 kbps data stream, even with cat foo. To test my DA I built a special test program and ran it on a second Mac. Then measured actual flow to ensure we where getting full 19.2 kbps data stream. -- Stuart Lynne ihnp4!alberta!ubc-vision!van-bc!sl Vancouver,BC,604-937-7532
mpl@sfsup.UUCP (06/16/87)
In article <1075@laidbak.UUCP>, katzung@laidbak.UUCP writes: > - From: levy@ttrdc.UUCP (Daniel R. Levy) > - Some (all?) PC/clones cannot keep up with speeds > 4800 baud in the VT100 > - terminal emulator (they lose bursts of characters, even with Xon-Xoff flow > - control). You want a PC, you buy a PC. You want a terminal, well, I hope > - you get an AT&T terminal (plug plug :-) ). [text deleted] > It sounds as if you're suggesting that any AT&T terminal will keep up at > speeds greater than 4800 baud! I have news for you: not all of them do. > The one I have in mind has no host flow control. [more text deleted] Well, as far as I know, all the AT&T *real* terminals for sale right now (ie 610) keep up with 9600 just fine (I've seen them do 19.2K without burping in the lab, but my mahcine doesn't go that fast, so I don't know if the "production models really go that fast). Just stay away from the "fancy features" terminals like the 510 (built-in phone, touch screen) if you really need performance. Mike Lindner ....!ihnp4!attunix!mpl
ken@argus.UUCP (Kenneth Ng) (06/17/87)
In article <1516@sfsup.UUCP>, mpl@sfsup.UUCP writes: > Well, as far as I know, all the AT&T *real* terminals for sale right now (ie > 610) keep up with 9600 just fine (I've seen them do 19.2K without burping in > the lab, but my mahcine doesn't go that fast, so I don't know if the > "production models really go that fast). Just stay away from the "fancy > features" terminals like the 510 (built-in phone, touch screen) if you really > need performance. So I guess the AT&T 4425 is not a real terminal. Is that why we got all these terminals as a donation from AT&T? (:-> > Mike Lindner > ....!ihnp4!attunix!mpl ... This signature was put in in a way to bypass the ... bogus artificial line limit on the .signature file. ... Also, by its length it adds fodder to help avoid having ... my followups being bounced due to the restriction on ... followup articles. Kenneth Ng: Post office: NJIT - CCCC, Newark New Jersey 07102 uucp !ihnp4!allegra!bellcore!argus!ken *** NOT ken@bellcore.uucp *** bitnet(prefered) ken@orion.bitnet
kaufman@Shasta.UUCP (06/20/87)
In article <1017@watmum.UUCP> smvorkoetter@watmum.UUCP (Stefan M. Vorkoetter) writes: >I am using an AT Clone, running a VT102 emulation package (Terminal Emulator >and File Transfer 2.40a), and it is having no trouble at 19200 baud, using >only XON/XOFF (no hardware flow control)... My teeth itch when I read this. Let's get it straight. The ability to read individual characters at a speed of 19200 bits per second is NOT THE SAME as the ability to asimilate characters at the rate of 1920 Characters per second. Got that? If you need/use XON/XOFF protocol, you are NOT keeping up at the speed in question. We are talking about the ability to process characters WITHOUT flow control at a given speed. BTW: just because your VAX, etc., is set to send characters at a (bit) speed of 19200 bits per second, is no reason to assume that is is sending 1920 Characters per second... you have to do some timing studies to determine what is really happening.
smvorkoetter@watmum.UUCP (Stefan M. Vorkoetter) (06/21/87)
In article <1790@Shasta.STANFORD.EDU> kaufman@Shasta.stanford.edu (Marc Kaufman) writes: >In article <1017@watmum.UUCP> smvorkoetter@watmum.UUCP (Stefan M. Vorkoetter) writes: > >>I am using an AT Clone, running a VT102 emulation package (Terminal Emulator >>and File Transfer 2.40a), and it is having no trouble at 19200 baud, using >>only XON/XOFF (no hardware flow control)... > >My teeth itch when I read this. Let's get it straight. The ability to >read individual characters at a speed of 19200 bits per second is NOT THE >SAME as the ability to asimilate characters at the rate of 1920 Characters >per second. Got that? If you need/use XON/XOFF protocol, you are NOT >keeping up at the speed in question. We are talking about the ability to >process characters WITHOUT flow control at a given speed. BTW: just >because your VAX, etc., is set to send characters at a (bit) speed of >19200 bits per second, is no reason to assume that is is sending 1920 >Characters per second... you have to do some timing studies to determine >what is really happening. The point is, the program can and does assimilate characters at 19200 bps. The only time it sends XON/XOFF is when its 4000 character buffer is full. I have used it with EMACS with flow control ENABLED, and it works no problem, because it never has to send an XOFF because a full screen update is less than 4000 characters. There is a reason to assume the characters are coming in at 19200 bps because I am using a Sytek box, which sends me the characters in 19200 bps bursts. I have seen other programs drop characters in the same situation. If the AT/Terminal Emulator combination could not keep up, plain XON/XOFF protocol wouldn't help a whole lot since several characters could still come in after the XOFF was sent. Some programs cannot keep up at this rate because one character is still being processed when the next comes in. Eventually (5 or 6 characters down the line) things get far enough behind so that an overrun error results. BTW: On buffering: When I press CTRL-S (or No Scroll), the program sends XOFF to the host. Any characters in the buffer will be read in and displayed. This can cause aggravation. HOWEVER, if I press CTRL-NumLock (or FN-Pause on a PCjr) then the program itself pauses. Characters keep coming in from the host and piling up in the buffer. If the buffer becomes full, an XOFF is sent to the host automatically. Thus you get full scrolling control, and still no loss of characters. Even though the program is paused, the serial drivers keep going (they are interrupt driven). I can even to a Print Screen in the middle of incoming characters; everything stops, the screen gets printed, and no characters are ever lost. To the above flamer (with the itchy teeth): Don't assume someone who says something that makes no sense to you doesn't know what they are talking about. I am quite comfortable in the picky details of serial communications. After all, I wrote the program in question. Got that? ~
mpl@sfsup.UUCP (06/24/87)
In article <898@argus.UUCP>, ken@argus.UUCP writes: > In article <1731@ttrdc.UUCP>, levy@ttrdc.UUCP (Daniel R. Levy) writes: > > control). You want a PC, you buy a PC. You want a terminal, well, I hope > > you get an AT&T terminal (plug plug :-) ). > > Don't be too sure about that. The AT&T 4425 has a hard time keeping up > with speeds beyond 2400 baud for extended periods of time unless the Mine does fine at 4800, unless a large proportion of the characters are character inserts/deletes. Line inserts/deletes/scrolling don't seem to slow it down much (it manipulates pointers to lines instead of the lines themselves). Only time I've seen it go slow is with the printer port on or when playing around with viewports (windows).
mikel@flmis06.ATT.COM (Mikel Manitius) (06/25/87)
>> Well, as far as I know, all the AT&T *real* terminals for sale right now (ie >> 610) keep up with 9600 just fine (I've seen them do 19.2K without burping in >> the lab, but my mahcine doesn't go that fast, so I don't know if the >> "production models really go that fast). Just stay away from the "fancy >> features" terminals like the 510 (built-in phone, touch screen) if you really >> need performance. > > So I guess the AT&T 4425 is not a real terminal. Is that why we got > all these terminals as a donation from AT&T? (:-> Hmm. I used to have a 4425 connected to the serial port on my 3B1, and I would have it run at 19.2kb. It would handle the output very well, and it was noticably faster than 9.6kb. However it still wasn't as nice as the throughput on the 3B1's monitor. Now I use it [3B1 monitor] over StarLan to login to all of our other machines, and it is really nice and fast. Of course I use my "rlogin" program (which I posted an article about a few months ago), rather than cu. Cu does 1 byte I/O, which makes it very slow, especially over StarLan. -- Mikel Manitius @ AT&T Network Operations mikel@codas.att.com.uucp | attmail!mikel
ken@argus.UUCP (Kenneth Ng) (06/30/87)
In article <13@flmis06.ATT.COM>, mikel@flmis06.ATT.COM (Mikel Manitius) writes: [edited discussion on using terminals at 9600 baud *without* flow control] > Hmm. I used to have a 4425 connected to the serial port on my 3B1, > and I would have it run at 19.2kb. It would handle the output very > well, and it was noticably faster than 9.6kb. > Mikel Manitius @ AT&T Network Operations Try turning the flow control off, and then doing a cat of a large file to the screen and tell me what happens after about 3 screens go by. The 4425 needs flow control enabled to keep up with 9600 baud for more than a screen of data. ... This signature was put in in a way to bypass the ... bogus artificial line limit on the .signature file. ... Also, by its length it adds fodder to help avoid having ... my followups being bounced due to the restriction on ... followup articles. Kenneth Ng: Post office: NJIT - CCCC, Newark New Jersey 07102 uucp !ihnp4!allegra!bellcore!argus!ken *** NOT ken@bellcore.uucp *** bitnet(prefered) ken@orion.bitnet