mdv@comtst.UUCP (Mike Verstegen) (11/28/90)
In the December 11, 1990 issue of PC Magazine, high speed dialup modems are evaluated. The Telebit T2500 "was not successful for any of our 24 impaired line tests, despite repeated tries and extensive consultation with the company." Though the tests were primarily for V.32 mode, the same failures were obtained in the PEP mode. The impairments used were "EIA-standard impairments" 1V32 through 6V32. Does anyone know what these are and how well they simulate impairments on real telephone lines? We have used Trailblazer+ modem (close relative of the T2500) for inbound and outbound, domestic and international with no problems from the modem. Are we just lucky, or did Telebit get a bad review? On a related note, I recently did an experiment with our connection to UUNET with respect to through-put rate. Previously we had been using MCI and get throughput (logged by HDB uucp xferstats) of about 800 bytes/sec. I then changed the UUNET number to include the 10288 AT&T selection prefix, and now we're getting 1050 bytes/sec throughput. The 30% increase certainly makes up for the few percent difference in cost. Mike -- Mike Verstegen Domain Systems, Inc Voice +1 407 686-7911 ..!uunet!comtst!mdv 5840 Corporate Way #100 Fax +1 407 478-2542 mdv@domain.com West Palm Beach, FL 33407
jimmy@denwa.info.com (Jim Gottlieb) (11/29/90)
In article <326@comtst.UUCP> mdv@comtst.UUCP (Mike Verstegen) writes: >...inbound and outbound, domestic and international with >no problems from the modem. Are we just lucky, or did Telebit get a >bad review? Telebit ALWAYS gets a bad review. All such tests in Byte and PC Magazine always have bad things to say about the Telebits in impaired-line situations. The catch is that they don't use noisy phone lines. They use equipment to _simulate_ bad connections. I suspect that this does not match reality. What they really should do is ship one of type of modem to India and then try to connect to it. All their fancy test equipment means nothing. In the real world, it's the Telebits that keep the data flowing.
schoff@uu.psi.com (Martin Schoffstall) (11/29/90)
I had responded privately to the original posting, but seeing this response I think I'll post publicly now. PSINet operates V.32 dialups throughout the US, with a cast of LEC's providing the service from PACBell, to NETEL, to Rochester Telephone, etc.. The T2500 in V.32 mode DOES appear to us to be succeptible to line noise. We've had problems. In PEP mode (only good for UUCP these days it would appear) there isn't a problem. What PC Week reported was on V.32 performance, and it was no surprise to us. Marty ----------------- In article <709@denwa.uucp> jimmy@denwa.info.com (Jim Gottlieb) writes: > >Telebit ALWAYS gets a bad review. All such tests in Byte and PC >Magazine always have bad things to say about the Telebits in >impaired-line situations. > >The catch is that they don't use noisy phone lines. >They use equipment to _simulate_ bad connections. I suspect that this >does not match reality. What they really should do is ship one of type >of modem to India and then try to connect to it. > >All their fancy test equipment means nothing. In the real world, it's >the Telebits that keep the data flowing.
cec@cup.portal.com (Cerafin E Castillo) (11/30/90)
The TELEBIT T2500 and T1500 modems utilize the Rockwell Chip set to
perform V.32. While I was with TELEBIT, the quality control problems
experienced with this chipset where quite suprising.
The chipset enable a phone line equalization function which killed
V.32 connections, between Telebits and non-Telebits, within seconds.
This was compensated for in Rev. GE6.01 of the firmware release.
Other such surprises where found, the biggest of which was that the
Rockwell Chipset is only capable of doing 12 kbps as an option and
can never do, in its present form, V.32bis (14.4 kbps).
Telebit has just announced the T1600 modem with its Telebits own DSP
implementation of V.32. This is the basis for TELEBIT's second generation
of modem products which promise V.32bis and beyond, BUT surprisingly,
this modem is not PEP capable, at present. The T1600 has a list price
of $795US.
I would really like to see a magazine review of this new modem versus
the T2500/T1500 modems.
BTW, I assisted the techs conducting these tests, along with a few other
Telebitians. We found that the test where conducted correctly and the
results where legitimate. THESE WHERE V.32 TEST ONLY! PEP was never
tested for inclusion in the review criteria.
===============================================================================
Cerafin E. Castillo || //\\ ||\\ ||
Network Consultant || //__\\ || \\ || Los Altos
Los Altos Networks || // ---\\|| \\|| Networks
340 Second St. #6 ||___// \ | \ |
Los Altos, CA 94022
(415) 941-8031 UUCP: {apple,sun,uunet}!portal!cup.portal.com!cec
INTERNET: cec@cup.portal.com
"...No hay mal que por bien no venga..."
===============================================================================dts@pwllheli.sw.stratus.com (Daniel Senie) (11/30/90)
I've done a significant amount of work with Telebit's V.32 between two T2500s on transatlantic lines between France and Boston. Most of the work was done using satellite and undersea copper, since it was at a time when Telecom France had screwed up their connection to the fiber optic line. We never had any noise problems running with MNP4 over V.32. We also have done lots of PEP work through a 40 year old step-by-step phone office with LOTS of noise, and get 1400cps or better on those lines. The text in the article leads me to believe that the PC Mag. reviewers did not even look at the documentation, and merely tried to plug and go. Byte did a review a while back on high speed modems and did the testing work properly. They used a similar telephone line simulator, but they also chose a reference modem and used that where appropriate. The ability of a modem to talk to a like modem is only part of the game. Byte does seem to at least try to do a thorough job with their investigations. -- Daniel Senie UUCP: uunet!lectroid!dts Stratus Computer, Inc. ARPA: dts@lectroid.sw.stratus.com 55 Fairbanks Blvd. CSRV: 74176,1347 Marlboro, MA 01752 TEL.: 508 - 460 - 2686
alan@adept.UUCP (Alan Ruffer) (11/30/90)
In article <326@comtst.UUCP> mdv@comtst.UUCP (Mike Verstegen) writes: >In the December 11, 1990 issue of PC Magazine, high speed dialup modems >are evaluated. The Telebit T2500 "was not successful for any of our 24 >impaired line tests, despite repeated tries and extensive consultation >with the company." Though the tests were primarily for V.32 mode, the >same failures were obtained in the PEP mode. > >The impairments used were "EIA-standard impairments" 1V32 through 6V32. >Does anyone know what these are and how well they simulate impairments >on real telephone lines? We have used Trailblazer+ modem (close relative >of the T2500) for inbound and outbound, domestic and international with >no problems from the modem. Are we just lucky, or did Telebit get a >bad review? > I have not seen the magazine tests you refer to in your posting, and do not know about these simulated impairments. All I can base my opinions on is 1 year of real-world experience with using a T2500 for news/mail/BBS line dial-in. The phone lines here tend to have noise and are otherwise dirty. The Telebit has been a magnificent performer during this period for us. The majority of the connections are either 2400 or 9600 PEP with T1000's on the other end. There are few V.32 modems around here to give any opinion as to how well it does in that mode. Perhaps these magazine tests (which I have been doubtful of for years now) are not really good indicators of real world conditions. Also PEP connections are superior under less than optimum line conditions when compared with v.32. Conclusion: Bad review. You are lucky to have a Telebit. Alan
rick@uunet.UU.NET (Rick Adams) (12/01/90)
UUNET (as a Telebit distributor) received a copy of the following letter from Telebit to PC Magazine today: November 27, 1990 Mr. Bill Machrone Editor-in-Chief & Publishing Director PC magazine One Park Avenue, 4th Floor New York, NY 10016 Dear Mr. Machrone: This letter is in response to your review of the Telebit T2500 modem which appeared in the December 11, 1190 issue of PC Magazine. Telebit Corporation disputes the results of this review and questions the credibility of the testing. Author Mike Byrd states, "Because it did not pass the impaired line tests, we cannot recommend this (Telebit) modem." Telebit Corporation believes this conclusion to be flawed and based on faulty test conditions and procedures. In fact, it was clearly demonstrated that an aborted PC file transfer caused the lack of data transfer. The modem itself did not in any way contribute to or relate to the authors difficulties; moreover: 1. Tests conducted by Telebit technicians and provided to PC Magazine verify that incomplete file transfers were caused by a bug in HyperAccess/5 software. Telebit Technicians removed the T2500 modem from the test setup, establisihing a direct connection; the problem continued to occur. There could not have been any element of the difficulty attributable to the Telebit modem. 2. Telebit's technical support personnel offered several solutions to the editors to correct the alleged problem and which would have resulted in a successful file transfer over impaired lines. Telebit is at a loss to explain why these solutions were not implemented. These testing inaccuracies are a disservice to both Telebit Corporation and the readers of PC Magazine. Telebit Corporation has earned an outstanding reputation as a leading manufacturer of high-performance dial-up communications products and we strongly believe that a full retraction of the erroneous test results is in order. Should PC Magazine wish to retest the T2500 or, indeed, any Telebit products, Telebit will be more than pleased to provide evaluation units along with guidance on correct testing configurations and procedures. Sincerely, Mark Huntizinger Director, Product Marketing
joachim@jrix.radig.de (Joachim Riedel) (12/01/90)
I also use a T2500 in PEP-Mode for UUCP-g in connection to a T 2000. I poll daily 1.5 MB to 3. MB and it works and works and works and works. I also connect sometimes to a USR in V.32 Mode and it works and works but only when the phone line is in good condition. When it is a noisy line then there is someone crying on the other end, but not on mine. Therefore: We do not know why it works (assuming the articles in PC-Magazine or in other magaines are correct) but we do know that it works. Dont' worry, we're happy. Joachim ---------------------------------------------------------------------- Joachim Riedel joachim@jrix.radig.de Geschwister-Scholl-Strasse 48 D-6050 Offenbach am Main Tel. +49 69 85 62 25 ----------------------------------------------------------------------
root@zswamp.fidonet.org (Geoffrey Welsh) (12/01/90)
Cerafin E Castillo (cec@cup.portal.com ) wrote: >Telebit has just announced the T1600 modem with its Telebits own DSP >implementation of V.32. This is the basis for TELEBIT's second >generation >of modem products which promise V.32bis and beyond, BUT surprisingly, >this modem is not PEP capable, at present. The T1600 has a list >price of $795US. I would hope that there's a T2600 or some similar product combining DSP V.32 with PEP on the way. -- UUCP: watmath!xenitec!zswamp!root | 602-66 Mooregate Crescent Internet: root@zswamp.fidonet.org | Kitchener, Ontario FidoNet: SYSOP, 1:221/171 | N2M 5E6 CANADA Data: (519) 742-8939 | (519) 741-9553 MC Hammer, n. Device used to ensure firm seating of MicroChannel boards Try our new Bud 'C' compiler... it specializes in 'case' statements!
mje@mje99.UUCP (Mark J Elkins) (12/04/90)
In article <1990Dec1.104606.9580@jrix.radig.de> joachim@jrix.radig.de (Joachim Riedel) writes: >I also use a T2500 in PEP-Mode for UUCP-g in connection to a T 2000. I >poll daily 1.5 MB to 3. MB and it works and works and works and works. So I'm in Southern Africa - and my TB2500 works well too - calling all over the world. >I also connect sometimes to a USR in V.32 Mode and it works and works but >only when the phone line is in good condition. I agree here too - now maybe the throughput on the TB is low (you can never tell easily ?) - but it holds the line - which results in cheaper phone bills. Also - all the V32's that I know of that have tried to get through to me - have done so OK - thought they suffer on bad connections. >Dont' worry, we're happy. ** DO ** worry.... Bad press --> lower sales --> less R&D --> poor future. So is PC Mag going to print a re-test? an appology? - or - some bright folk in the US gona hand-hold them into getting the 'S' Regs correct? If Telebit improve on the TB, then I'd like to see... 1 - add 38K4 and V32bis 2 - A 2 digit display of the current line quality, regardless of connection type [PEP/V??] with a range 0=NC, 99=back-to-back TB's, (kinda like my digital volt meter across my car battery) -- . . ___. .__ Olivetti Systems & Networks, Unix Support - Africa /| /| / /__ UUCP: {uunet,olgb1,olnl1}!olsa99!mje (Mark Elkins) / |/ |ARK \_/ /__ LKINS mje@olsa99.UUCP (Postmaster) Tel: +27 11 339 9093