NetNews Offline 2/NetNews Offline Volume 2.iso/news/comp/dcom/modems-part1/2306

Text File | 1996-08-05 | 6.8 KB | 139 lines

Newsgroups: comp.dcom.modems
Path: FreeNet.Carleton.CA!an171
From: [e-mail] (Anthony Hill)
Subject: Re: Is 33.6 only available for USR?
Message-ID: <[e-mail]>
Sender: [e-mail] (Anthony Hill)
Reply-To: [e-mail] (Anthony Hill)
Organization: The National Capital FreeNet
References: <4dd14b$[e-mail]> <4ddchf$[e-mail]> <[e-mail]> <4dh7j3$[e-mail]> <[e-mail]> <4dlt8f$[e-mail]> <[e-mail]> <4dqj68$[e-mail]> <DLHzr <4dsqot$[e-mail]>
Date: Mon, 22 Jan 1996 03:07:08 GMT


doug haire ([e-mail]) writes:
> [e-mail]>:
> Distribution:
>
> Anthony Hill ([e-mail]) wrote:
> :
> : doug haire ([e-mail]) writes:
> : > Anthony Hill ([e-mail]) wrote:
> : > :     I believe that the original point of the message was that IF your
> : > : SNR is not good enough to support v.34 connections now, the new features
> : > : of the annex to v.34 will make NO difference in your connect speed
> : > : (assuming everything else is the same).  What this means is that if you're
> : > : getting a 26.4 connection with teh 3429 symbol rate with 28.8 v.34, you
> : > : will NOT see your connect speed increase to 28.8 by simply adding the 33.6
> : > : firmware UNLESS other things were changed (eg. when USR released their
> : > : 33.6 firmware for their Couriers).
> : >
> : > I believe there was a change included with the V.34+ upgrade (at least in
> : > the Courier) that improved speeds with the lower symbol rates, allowing a
> : > 28.8k connect (stable) at a lower symbol rate. I could be wrong about
> :
> :     Yes, that is correct, and that's why I specified that the
> : connection was using the 3429 symbol rate already but still only getting a
> : 26.4 connection due to poor SNR.  FWIW, the annex to v.34 allows all the
> : symbol rates to connect at one 2400bps higher then with the original v.34
> : spec, with the exception of the 2400 symbol rate (still stuck at 21.6
> : connections) and the 3429 symbol rate which can go up two steps (from 28.8
> : to 33.6).
>
> I believe you were talking to someone who had a V.34 and was not
> achieving 28.8k connects, correct? It would seem then that going to V.34+
> might actually help him achieve them. If he's using a Sporty the upgrade
> cost is $25 (I think). If he has a Courier, it's free.

    It may help them but ONLY if the reason why they aren't getting
28.8 connections is because of limited bandwidth.  Of course, that's
assuming USR hasn't improved their v.34 code in general with their newest
firmware, which is possible.

> : > this but I don't think so. I do know, subjectively speaking, that my
> : > unstable 28.8k connects became more stable when I went to the V.34+ and
> : > that I started getting more 28.8k connects than I had. Admittedly, this
> : > is subjective since I did not run control tests.
> :
> :     USR did some very effective refining of their v.34 code with their
> : 33.6 code for their Couriers.  Sportsters saw many of the same refinements
> : with the 04/18/95 code.
>
> As I was told by someone at USR (no, not the folks you get on the support
> line) that the Sporty got the same code that the Courier did, essentially.

    They're based on similar code, but they are not identicle.

> : > :     This has absolutely NOTHING to do with the annex to v.34, it's
> : > : merely coincidence that USR decided to add both new features in the same
> : > : firmware for their Couriers.  Also, unless something has changed since I
> : > : last checked, these new v.42 features (244 byte LAP-M packets and SREJ) have
> : > : NOT been added to the Sportster line.
> : >
> : > It was included in the V.34+ ROM code. As I understand it, it should also
> : > be in the Sporty V.34+ code but perhaps someone who has a V.34+ Sporty
> :
> :     There's not particular reason why these enhancements should be
> : added with any one code.  They do not require any of the new features of
> : the annex to v.34, in fact, they don't require v.34 at all.  Zyxel's 14.4
> : modems (and 28.8s I presume) included both of the features that USR added
> : with their 33.6 code.
>
> The Zyxel 14.4k did not have the expanded V.42 (I own a U-1496E+) and I

    Perhaps these features were only added with newer firmwares,
because I know of many people who get 256 byte LAP-M packet/SREJ
connections with their Zyxels.  Either way, the point is that these
features are in no way dependant on v.34, or any other actual modem
protocol.

> got tired of waiting for a Zyxel 28.8k so I can't comment on those. I
> beleive you will find that the new Sporty's have the 244...

    Mine doesn't, but that may not be indacative of others.

> : > : > You will find that it outperforms the "standard" implementation in the
> : > : > Rockwell chipset modems every time. That is, I get high 3300's to low
> : > : > 3400's using zmodem (w/Mobyturbo) on a 28.8k connection. I get high
> : > :
> : > :     FWIW the PPI has a maximum LAP-M packet size of 512 bytes, more
> : > : then twice the size of the USR Couriers 244 bytes, so the PPI will have a
> : > : higher maximum theoretical throughput then the USR Courier.  Of course,
> : > : PPI does not support SREJ (unless it's been added recentely), so if there
> : > : are any noticible amounts of errors on the line, the Courier may end up
> : > : outperforming the PPI.
> : >
> : > Would you run some tests with the PPI and see what kind of rates you get?
> : > I'd certainly be interested in seeing them.
> :
> :     Several magazines did this a while back, the result was that PPI
> : and Hayes blew the competition out of the water on throughput tests (done
> : over perfect connections).
>
> I don't read "Several magazines" so I don't know what you're referring to
> here. I don't trust most magazine tests since they tend to come up with
> some pretty bizarre results ... such as Hayes modems actually performing
> well.

    The results can be quite misleading, but there ARE some things
which Hayes modems will do well at.  Unforutnatly those conditions don't
exist in the real world :>.

> Now, if you would run some simple tests, I would like to see the results.

    Unforutnatly I no longer have a PPI modem to run tests against.

> Oh, by the way, I just got a 3940 cps rate on a download of a 334884 byte
> compressed file. Seems I got one of those elusive 33600 connects (through
> an LD connection no less!).

    It should be obvious to you that the PPI will NOT get those sorts
of rates, however, when connected at the same speed, the PPI does have a
higher theoretical maximum throughput then the USR Courier.  Of course,
theoretical throughput isn't too important as compared to real world
performance.

Anthony

--
Anthony Hill | [e-mail]