home *** CD-ROM | disk | FTP | other *** search
- Path: giskard.demon.co.uk!dale
- From: dale@giskard.demon.co.uk (Dale Shuttleworth)
- Newsgroups: comp.sys.mac.comm,comp.dcom.modems
- Subject: Re: faster than 28.8
- Followup-To: comp.sys.mac.comm,comp.dcom.modems
- Date: Sun, 4 Feb 1996 19:49:53 GMT
- Organization: Dale's home Linux box.
- Message-ID: <DM9oF5.1BM@giskard.demon.co.uk>
- References: <Pine.LNX.3.91.960203221724.10222B-100000-100000@eola.ao.net>
- X-NNTP-Posting-Host: giskard.demon.co.uk
- X-Newsreader: TIN [UNIX 1.3 950824BETA PL0]
-
- Butt-head (eric@eola.ao.net) wrote:
-
- [...]
-
- : 3. Hayes files are made so the searches through the "dictionary" for v.42
- : go really fast so they can show off the Optima's 230400 DTE performance
- : without doubling the CPU in it. These easily (CPU time wise)
- : compressible files would also work well on other modems.
- :
- : The third possibility seems most likely to me, while the second is most
- : unlikely, as the amount of data in the files and limited ROM size would
- : make a cheating modem actually more impressive than a working one.
-
- Well, here is the data from my tests. The connection is between my
- desktop Courier and the USR TC rack (basically the Courier design) at
- my ISP. The connection speed was 28.8kbps, V42 and V42bis were used and
- the connection MTU was 576 bytes. No blers or retrains occured during
- the test. It should be noted that there are 80 modems on the host
- (demon-du) connected via SCSI terminal servers.
-
- I have used the pppstats program to give an indication of throughput.
- This program includes packet headers and such like in the "in" and
- "out" byte counts. A new line is displayed approximately every 5
- seconds.
-
- From my Courier to the ISP:
-
- cat /dev/zero | telnet demon-du discard
-
- in pack comp uncomp err | out pack comp uncomp ip
- 549 54 54 0 0 | 50867 98 98 0 0
- 564 56 56 0 0 | 50348 97 97 0 0
- 585 56 56 0 0 | 50867 98 98 0 0
- 573 56 56 0 0 | 50346 97 97 0 0
- 571 56 56 0 0 | 50869 98 98 0 0
- 554 55 55 0 0 | 50346 97 97 0 0
- 561 55 55 0 0 | 50349 97 97 0 0
- 564 56 56 0 0 | 50868 98 98 0 0
- 549 52 52 0 0 | 50348 97 97 0 0
- 566 56 56 0 0 | 50346 97 97 0 0
- 575 56 56 0 0 | 50868 98 98 0 0
-
- = about 10kbytes/sec.
-
- We can also do:
-
- telnet localhost chargen | telnet demon-du discard
-
- in pack comp uncomp err | out pack comp uncomp ip
- 625 59 59 0 0 | 57148 108 108 0 0
- 630 61 61 0 0 | 59008 112 112 0 0
- 625 60 60 0 0 | 57076 107 107 0 0
- 651 62 62 0 0 | 57212 108 108 0 0
- 658 62 62 0 0 | 57743 109 109 0 0
- 647 63 63 0 0 | 58321 111 111 0 0
- 651 63 63 0 0 | 58384 110 110 0 0
- 633 62 62 0 0 | 54480 103 103 0 0
- 623 61 61 0 0 | 58592 110 110 0 0
- 645 62 62 0 0 | 58400 110 110 0 0
-
- = about 11.5kbytes/sec.
-
- Hmmm, either you are right and the Courier doesn't like all zeros, or the
- filesystem adds a fair bit of overhead. Neither speed is particularly
- poor however.
-
- From the ISP to my Courier:
-
- telnet demon-du chargen > /dev/null
-
- in pack comp uncomp err | out pack comp uncomp ip
- 33316 63 63 0 0 | 634 63 63 0 0
- 31847 60 60 0 0 | 605 60 60 0 0
- 33319 63 63 0 0 | 633 63 63 0 0
- 33172 62 62 0 0 | 623 62 62 0 0
- 33024 63 63 0 0 | 634 63 63 0 0
- 32862 62 62 0 0 | 623 62 62 0 0
- 33157 62 62 0 0 | 622 62 62 0 0
- 31550 60 60 0 0 | 604 60 60 0 0
- 33180 62 62 0 0 | 622 62 62 0 0
- 33091 63 63 0 0 | 634 63 63 0 0
- 33193 62 62 0 0 | 622 62 62 0 0
-
- = about 6.6kbytes/sec.
-
- Now, this is down to the kind of speed you claim, unfortunately we don't
- know if it is the system at the far end which can't continuously feed 80
- modems, or the poor performance of the modems themselves.
-
- My ISP also has a number of Ascend Max units with 6x 8-modem cards
- installed (Rockwell based), those quite happily push data down the line
- at 11kbytes/sec - I didn't connect to one so I can't show you the data.
-
- My Courier certainly seems to be able to send the data out at adequate
- speed. It can also receive it at that speed quite happily (although
- this has not been demonstrated with a USR modem).
-
- Dale.
-
- --
- ******************************************************************************
- * Dale Shuttleworth *
- * Email: dale@giskard.demon.co.uk *
- ******************************************************************************
-