ISP Information:
A type of memory developed by NEC. VCM memory adds some low latency SRAM registers onto standard SDRAM, much like EDO compared to Fast Page Mode memory. VC memory never took off, as DDR memory and RDRAM superseded SDRAM.
ISP Glossary:
VCM - "E. Kunze" wrote in messagenews:iff3mvkit9sovmekd01sok4qhd08ihlkjt@4ax.com... On Thu, 11 Sep 2003 00:09:01 GMT, "Hooda Gest" wrote: Thank you all for your kind and elaborate answers, which will take me awhile to work through. I'll try to preliminarily answer your posts here to try to keep the thread a bit linear. Let's say, I used until now modem_A. As Aaron pointed out, this modems records don't show the use of an error correcting protocol. ATI4 in Hyperterm shows, that &M4 is set. I can alter the setting, too. The question now is, wether this modem can't engage error correction or wether it just doesn't report the use.This bothers me. It suggests either the modem (_A) is not able to negotiateerror correction or is incapable of reporting in the post-call diagnosticsthat it had done so.We can maybe confirm this when we locate the modem log for you.The reason you cannot find it now is you are looking for the wrong thing. InWin95, the modem log is called (strangely enough) "modemlog.txt" and is inthe Windows folder. It also has a limitation: it will grow no bigger than~256k and then cease recording data. So, probably your modem log is therebut full of rather old data. Easy to fix, though. Locate the file, deleteit, and it will be rebuilt on the next call. The modem log will have theinit strings sent to the modem and it wil also have th eactual connectreport of the modem when a connection was made. I would like to see both ofthese. Today I replaced the modem with another one (very similar USR 56k Message Modem), just exchanging the hardware without changing any software settings. It gave the following report: (Modem B): Chars sent 498718 Chars Received 3596553 Chars lost 0 Octets sent 240039 Octets Received 2721608 Blocks sent 17801 Blocks Received 22270 Blocks resent 103 Retrains Requested 0 Retrains Granted 0 Line Reversals 0 Blers 154 Link Timeouts 0 Link Naks 13 Data Compression V42BIS 2048/32 Equalization Long Fallback Enabled Protocol LAPM Speed 45333/26400 V.90 Peak Speed 45333 Last Call 00:21:33 Modulation V.90 Carrier Freq (Hz) None/1920 Symbol Rate 8000/3200 Trellis Code None/64S-4D Nonlinear Encoding None/ON Precoding None/ON Shaping ON/ON Preemphasis (-dB) 1/8 Recv/Xmit Level (-dBm) 19/16 Near Echo Loss (dB) 12 Far Echo Loss (dB) 0 Carrier Offset (Hz) NONE Round Trip Delay (msec) 4 Timing Offset (ppm) 2657 SNR (dB) 0.0 Speed Shifts Up/Down 1/1 Status : aa,5,12N,12.4,-0,0N,0,42.1,18.9 Obviously, this Modem_B reports the use of error correction. My first impression is, however, that the performance is not better than that of Modem_A, but I'll have to do some more measurements to be sure.It's probably as poor or close to. Try the &u6&n29 to limit modem_B to44000bps maximum and see if it isn't a bit smoother. Remember, throughput ismore important than connect speed. Now let's discuss the SNR value. I believe that this Signal to Noise Ratio should be large to get a good connect speed. (Funnily, Modem_B reports SNR 0.0 ).That's a glitch in the firmware. Not sure if there might be a later versionavailable which corrects that. Probably not. I believe you can get asomewhat valid SNR if taken while online. You would have to get it this way:Dial up the ISP number in Hyperterm.Do not attempt to log in.Execute the escape code to put the modem in command mode.(+++)Query the modem with ati11 and you should see something other than 0.0 forSNR.Return online with ato or just hangup with ath. My records show, that t
|
