🚀 go-pugleaf

>
>It should be noted that PSI has officially stated that they are moving
>away from *new* PEP installations, and focusing on V.32.
>In my miserable opinion, I think this is a mistake.
>
"New PEP installations" means "new POPs", additional PEPs are and
will be added to current PEP rotaries in places like NY, Reston,
SantaClara etc...

If the ONLY consideration is marketing into an installed base of
UUCP'ers than I would concur with you. If you consider issues such
as 

- price/performance
- professional high quality rack mount modem issues (vs standalone)
- PEP latency
- SLIP/PPP market
- mass market availability of V.32
- rapidly developing new markets for "electronic mail" and other things

Then you can come to different conclusions - your mileage may vary
as they say....

PEP has multiple technical losses, multiple marketing losses and one
marketing win.  [Now Telebit MODEMS via UUCP spoofing get another
technical win under certain circumstances.]

Then there is V.32bis....

Marty

time@ice.com (Tim Endres) writes:
>The best throughput I think I have *ever* seen from a T2500 (without
>compression, since it was batched compressed news) is about 1330 cps.
>I ordinarily see something on the order of 1100 cps daily.
>I don't think you will ever see over 1400. Anyone seen better?

I find that there is a big difference between talking to various types
of machines.  When I talk to a 3b2/400 (aka. the pig in an ugly box),
all I get is ~1000 bps.  Talking from a Sparc SLC to a Sparc SLC I
regularly get >1400 bps.

SparcSLC!news M (6/19-1:30:45) [cua0] -> 40392 / 26.200 secs, 1541 bytes/sec
SysV368!news S (6/18-23:56:40) [ttya] -> 50684 / 34.000 secs, 1490 bytes/sec
SparcSLC!news M (6/19-1:45:00) [cua0] -> 72727 / 48.960 secs, 1485 bytes/sec
SparcSLC!news M (6/19-1:17:17) [cua0] -> 74067 / 50.060 secs, 1479 bytes/sec
SparcSLC!news M (6/19-1:16:19) [cua0] -> 86215 / 58.510 secs, 1473 bytes/sec
SparcSLC!news M (6/19-1:27:42) [cua0] -> 94789 / 64.480 secs, 1470 bytes/sec
SparcSLC!news M (6/19-1:28:55) [cua0] -> 94499 / 64.250 secs, 1470 bytes/sec
SparcSLC!news M (6/19-1:13:56) [cua0] -> 84943 / 57.800 secs, 1469 bytes/sec
SparcSLC!news M (6/19-1:12:49) [cua0] -> 99385 / 67.780 secs, 1466 bytes/sec
SparcSLC!news M (6/19-1:32:02) [cua0] -> 98805 / 67.370 secs, 1466 bytes/sec
SparcSLC!news M (6/19-1:15:12) [cua0] -> 99235 / 67.710 secs, 1465 bytes/sec
SparcSLC!news M (6/19-1:44:04) [cua0] -> 96669 / 65.950 secs, 1465 bytes/sec
SparcSLC!news M (6/19-1:47:30) [cua0] -> 99241 / 67.700 secs, 1465 bytes/sec
SparcSLC!news M (6/19-1:30:10) [cua0] -> 98615 / 67.320 secs, 1464 bytes/sec
SparcSLC!news M (6/19-1:42:50) [cua0] -> 97945 / 66.890 secs, 1464 bytes/sec
SparcSLC!news M (6/19-1:46:14) [cua0] -> 95865 / 65.440 secs, 1464 bytes/sec
SysV368!news S (6/18-23:57:26) [ttya] -> 54420 / 37.300 secs, 1458 bytes/sec

-wolfgang
-- 
Wolfgang Rupprecht    wolfgang@wsrcc.com (or) uunet!wsrcc!wolfgang
Snail Mail Address:   Box 6524, Alexandria, VA 22306-0524

In article <1991Jun19.175922.8524@uu.psi.com> schoff@uu.psi.com (Martin Schoffstall) writes:
> If the ONLY consideration is marketing into an installed base of
> UUCP'ers than I would concur with you.

I wouldn't say it's your only consideration, but I can't see why it shouldn't
be *a* consideration. That's who your original customers were, after all.

> - price/performance
> - professional high quality rack mount modem issues (vs standalone)
> - PEP latency
> - SLIP/PPP market
> - mass market availability of V.32
> - rapidly developing new markets for "electronic mail" and other things

On the other hand, you have:

  - Installed Base
  - UUCP spoofing
  - "poor" exchanges (for example, Sugarland Telephone)
  - PBXes

What would be your cost for transferring *one* TB on Houston?
-- 
Peter da Silva; Ferranti International Controls Corporation; +1 713 274 5180;
Sugar Land, TX  77487-5012;         `-_-' "Have you hugged your wolf, today?"