Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!think.com!hsdndev!husc6!popvax!conrad From: conrad@popvax.uucp (M20400@c.nobili) Newsgroups: comp.sys.mac.comm Subject: Re: Summary of 9600 V.32 bis modem info Summary: Some corrections; read comp.dcom.modems for more complete info... Message-ID: <6913@husc6.harvard.edu> Date: 31 May 91 03:16:03 GMT References: <1991May30.000326.22453@nntp-server.caltech.edu> Sender: news@husc6.harvard.edu Followup-To: comp.dcom.modems Distribution: na Organization: Harvard University, Office for Information Technology Lines: 107 <1991May30.000326.22453@nntp-server.caltech.edu> suresh@cheme.caltech.edu (Suresha Guptha B.A.): >I am including both my original posting and the responses that I got below. >*******My original posting****** > I am considering buying a 9600 V32.bis modem. Fax capabilities >of the modem are not important to me. Could some one suggest a good >modem that I should purchase ? I would appreciate if you can share the >experiences you had or the reviews you have read about the available >9600 baud modems. If I get enough replies I will summarize it to the >net. >*******Responses that I received********* > I use a US Robotics Courier V.32 modem. I can get the free upgrade to >V.32bis but I haven't. ... No, you can't. No such thing. Check it out with USR. >************** >I have a Hayes 9600 UltraSmartModem v42bis (read 38.4k throughput) modem. I'd >definitely go for v42 as it allows 4;1 compression vrs 2:1 allowed by >MNP5. I paid around $800.00 for it and have been thorougly pleased >with it's performace. I use White Knight as the modem program. I can >get _real_ throughput of about 22K with other v42 modems during >transfers of files that were already compressed (.sit.hqx). ... Hmmm. Your "_real_ throughput of about 22K" on "files that were already com- pressed" seems questionable. I believe your modem supports a maximum link rate of 9,600 bps. So with a pair of such modems, connected at 9,600 bps and using V.42bis compression your could _hope_ for a maximum throughput of 38,400 bps if you believed the manufacturers' claims of 4:1 compression for V.42bis.... You would probably have to "cook" a file to get anything close to this (i.e., make a big file of just spaces, or the letter x or something). Consensus seems to be that ratios of 2.5:1 to 2.7:1 are more usual. Your claimed throughput works out to a ratio of 2.3:1, which is pretty close to what one should expect for "normal" compressible text. On stuff that was _already_ compressed (as tightly as V.42bis can compress things) you will see "real throughput" of _no greater than_ 9,600 bps! No (further) compression possible = 1:1 compression ratio => throughput of 1 X link rate = 9,600 bps.... Perhaps your files are in fact somewhat compressible. After all, it looks like they have been put through BinHex. (The whole point of which is to _avoid_ a bunch of sequences of eight bits when encoding things, and _must_ not be very "efficient" at representing information....) It just strikes me that your ratio seems higher than could be explained by just the expansion of the StuffIt file by BinHex. Perhaps StuffIt is also not as efficient as V.42bis? Or are you not measuring the right thing? At any rate, I have seen lots of misleading info about compression ratios and expected throughputs, and didn't want any people to think that their V.42bis modems can further compress efficiently compressed files.... >documentation is excellent. Learned more than I ever needed to know >about modems. Try a USRobotics manual! Truly impressive.... >************** >My impression is that the current V.32bis modems aren't very reliable yet. >I'm still buying V.32/V.42bis only. ... Hmmm. _I_ am overwhelmed by the pair of USRobotics Courier V.32bis modems I just bought. I have found the new modems to be rock solid and the expected 1.5 times as fast as the V.32-only modems on clean links (note that this ratio would be even larger on noisy links, as the V.32bis modems don't have to fall back nearly as far _and can fall forward again when the link improves_).... _Way_ superior to the USRobotics Courier V.32 and Telebit T2500 modems that I had been using. Perhaps this neanderthal would like to buy a used USR Courier V.32 from me? ;-) >************** >Try the new US Robotics V32.bis. It is state of the art and runs as low as >$575 (I think?). Indeed. Quite awesome. Where did you see that price?! Lots of people would be interested. Best I've seen is $599 from a mail-order house that does not provide after-sale support.... >************** >-Suresh I have put comp.dcom.modems in the Followup-To: field of this message. You should really read that group if you are interested in modems! There are some wonderfully knowledgeable people there (I don't include myself in that category...). You will get very thorough answers to all kinds of modem-related questions there in a timely fashion. Questions like this one should probably be posted there for best results. If you are not asking about the hardware, but rather the software or cable or other interface for the Mac, then this _is_ the place to ask.... Really, you will be pleasantly surprised at the quality of the information and the signal-to-noise ratio there (just use K in rn if you see an article about a Hayes lawsuit...). +---- C o n r a d C . N o b i l i ----+ | | | Harvard University | Internet: conrad@harvarda.harvard.edu | | Office for Info. Tech. | conrad@popvax.harvard.edu | | Information Services | BITNET: CONRAD AT HARVARDA | | Technical & User Services | CONRAD AT HARVSPHB | | 1730 Cambridge Street | voice: (617) 495-8554 | +---- Cambridge, MA 02138 | fax: (617) 495-0715 ----+