I think (and perhaps someone can refresh my memory) that 3com actually responded to this ascend phenomenon, and said that its actually more or less a bug in the ascend code, and not a problem in the 3com code, in that the ascend picks piss poor connections to negotiate v90 with. But I would be interested in knowing the stability of the connection, the number of retrains, and the ongoing average xmit/rcv rates of such a connection, raw thruput tests would be cool too. I don't know if anyone has done this. I would imagine, actually hope, that 3com's tests labs, have done many tests against their own stuff and against competitors. I would think they would not make a policy of releasing such data, unless of course 3com was best in every case, which I doubt it was, but certainly coudln't have done too poor. Brian On Sun, 17 Oct 1999, Greg Coffey wrote:
If it comes down to significantly degrading other connections to our system in order to enhance some on the fringe to get them v90, no, I don't want that. I would like to hear it from someone at 3Com that this is the current thinking on the matter.
If you are going to pursue this rather dubious argument, why not just say what you really want? You want 3com chassis to _connect_ at some outrageously high rate, say, 53k, and then silently fall back to a slower speed since it's unlikely the customer will notice the fallback.
Hey 3com, how about that? Make all v.90 connections, regardless of quality, always "CONNECT 53000" and then fall back to 40k or V.34 later? Please make sure it's an option that can be disabled, I'd don't need any more "mysterious disconnects" or "neglible throughput" trouble calls than I already get.
Thanks,
Greg Coffey, Visionary Communications V 307-234-5443 F 307-234-5446 ===================================================================== 100 N. Center St., Casper, WY 82601 WWW.VCN.COM
- To unsubscribe to usr-tc, send an email to "majordomo@xmission.com" with "unsubscribe usr-tc" in the body of the message. For information on digests or retrieving files and old messages send "help" to the same address. Do not use quotes in your message.
----------------------------------------------------- Brian Feeny (BF304) signal@shreve.net 318-222-2638 x 109 http://www.shreve.net/~signal Network Administrator ShreveNet Inc. (ASN 11881) - To unsubscribe to usr-tc, send an email to "majordomo@xmission.com" with "unsubscribe usr-tc" in the body of the message. For information on digests or retrieving files and old messages send "help" to the same address. Do not use quotes in your message.