home *** CD-ROM | disk | FTP | other *** search
- Path: sobt.accessorl.net!user
- From: eric@accessorl.net (Eric Shaw)
- Newsgroups: comp.dcom.modems
- Subject: Re: Is 33.6 only available for USR?
- Date: Fri, 02 Feb 1996 00:00:12 -0500
- Organization: Access Orlando
- Message-ID: <eric-0202960000120001@sobt.accessorl.net>
- References: <4dd14b$odp@grid.direct.ca> <4ddchf$c4i@shellx.best.com> <30FA9B32.4862@fishnet.net> <4dh7j3$81i@shellx.best.com> <30FCE339.36B9@fishnet.net> <4dlt8f$1qom@seminole.gate.net> <DLGM5w.H8o@freenet.carleton.ca> <4dqj68$dcak@navajo.gate.net> <eric- <199601310359.WAA10609@iii2.iii.net> <Pine.A32.3.91.960131004959.20602A-100000@navajo.gate.net>
- NNTP-Posting-Host: sobt.accessorl.net
-
- In article <Pine.A32.3.91.960131004959.20602A-100000@navajo.gate.net>,
- doug haire <dhaire@gate.net> wrote:
-
- >> >{On} 08/05/94 5:49 am 26400 bps ours II 08/05/94
- >> >Z 983040 115200 bps 11327 cps 0 errors 1 1024
- d:/download/1x30.tst 17972
- >> >Z 327680 115200 bps 7465 cps 0 errors 0 1024
- d:/download/2x10.tst 17972
- >> ...
- >You overlook his main thrust which was that USR didn't even approach a
- >3/1 compression. His statement that even limiting the connect rate to
- >14400 was blatantly false and I have posted data to show that. Since only
- >Hayes (I think, there may be one other) has the 230k DTE rate
- >availability and damn few comm/term programs have that as a setting.
-
- Between most PC's with standard 16550 cards, you won't get much over 11K/s
- even on all zero's being sent, but this is because of interupt latency and
- time to check the Zmodem acks. You will still get at least close to it,
- and this data DOES show a transfer over 11K/s, at 11327 cps. With
- Ymodem-G, you'll get even closer to the 11520cps theoretical max. 11327 I
- would say *IS* limited by the 115200 DTE, although 7465 cps transfers may
- or may not be, depending on whether the file in question was evenly
- compressible throughout.
-
- >> But locking the modems at 14.4k clearly has no direct bearing on this,
-
- Locking the modem at 14.4K shows that not only can the modem acheive
- better than 4:1 compression, but if the DTE isn't limiting it *as much* it
- can achieve better than 7:1, possibly much more.
-
- >inefficient, period. But his, and Hayes, claims of superior compression
- >are false: V.42bis is a standard and all modems compress using the same
- >algorithm.
-
- The algorithms *may* or may not be identical, but if they are, then the
- processor doing the compression in a modem that can only acheive a little
- over 6K/s is obviously slower than in one that is limited by the 115200
- DTE on the same file, unless there are other ridiculously high latencies
- in the slower modem.
-