Path: utzoo!attcan!utgpu!news-server.csri.toronto.edu!rutgers!usc!pollux.usc.edu!kjh From: kjh@pollux.usc.edu (Kenneth J. Hendrickson) Newsgroups: comp.dcom.modems Subject: Re: review of Practical Peripherals 9600SA Message-ID: <28577@usc> Date: 2 Dec 90 19:48:35 GMT References: <6058.27588AA8@zswamp.fidonet.org> Sender: news@usc Organization: EE-Systems, USC, Los Angeles Lines: 49 Nntp-Posting-Host: pollux.usc.edu In article <6058.27588AA8@zswamp.fidonet.org> root@zswamp.fidonet.org (Geoffrey Welsh) writes: >Kenneth J. Hendrickson (kjh@pollux.usc.edu ) wrote: > >I found that _disallowing_ Data Compression yielded > >higher data transfer rates! This was true both for compressed files > >(.ZIP) and for text files. This was suprising, especially in the > >case of text files. > On the other hand, I am puzzled by your results with text files. Were you >using a non-streaming protocol (such as Kermit, XMODEM, etc.)? Higher levels >of MNP can diminish interactive throughput; MNP5 should provide a significant >boost to text file transfers using streaming protocols such as ZMODEM. I was using zmodem. MNP5 made throughput worse with text files! I can't explain this. > We all hope, but it's been my experience that most of the low priced 9600 >bps modems have as their single goal 9600 bps performance; in many cases the >2400 bps circuits used do not seem to have been adjusted for use with the >(different) electrical characteristics of the 9600 bps circuits. This has the >sad effect of doing real damage to the 9600's non-9600 bps performance. Can anybody recommend a good V.32 9600 bps modem that also has at least useable performance at V.22bis, and V.22? As reported before, My Practical Peripherals PM9600SA rev 1.05 is TOTALLY UNUSABLE at 2400 bps and 1200 bps. I have corresponded with some others who also have this modem by email, and at least one person is able to use it at 2400 bps. Perhaps I just got a bad sample. However, I feel that the "black eye" that I have given Practical Peripherals over their totally non-usable 2400 and 1200 bps performance is well earned on their part. If their engineers knew that the 2400 and 1200 bps performance would be lousy (and I have to believe that they were cognizant of that fact) then it probably would have been better that the modem didn't even support V.22 and V.22bis. There is another problem with the Practical Peripherals 9600SA. Occasionally, when the modem takes a hit of line noise, the carrier will be lost. When this happens, a couple of relay clicks are heard from the modem, and data communications is lost, but the OH and CD lights stay on! The modem doesn't respond to lowering the DTR line (I have done "at &d2", nor does it respond to the "+++" sequence to return to command mode. There is no way to get ahold of the modem any more other than cycling the power. This is OK for interactive use, but it is totally intolerable for non-interactive use like getting a news feed or mail via uucp or some other means. My recommendation: I can't recommend the Practical Peripherals 9600SA. -- favourite oxymorons: student athlete, military justice, mercy killing Ken Hendrickson N8DGN/6 kjh@usc.edu ...!uunet!usc!pollux!kjh