RS> Presumably you mean you see those on the ZedZap status
RS> screen thats up during the actual transfer of the file.

NP> That's what I am talking about.  CRC error at 25198
NP>                                  CRC error at 47910....etc

NP> They chances of their appearance depends on Who I am calling.
NP> From Teddy (same exchange) 4 crc errors in 90k
NP> From Brisbane             no crc errors in 2.8 MEG

RS> Thats interesting. Corse its possible that its just Teddys
RS> system thats fucking up too, not yours. You would think that
RS> it would be yours since you see the problem with others tho.

NP> It appears to be WHO I call

NP> On a second checking, it seems to relate to the time I am calling, which
NP> gave the appoearance of who I was calling.  Definitely a line/exchange prob

Trouble with that theory is that a line/exchange problem should
be transparently fixed between the modems at the LAPM level with
no CRC errors visible at the file protocol level at all.

Its MUCH more likely that the time of calling is related to what
else the system you are calling is doing at the time you call,
and its THAT which is producing those file protocol CRC errors.

RS> Which also supports the proposition that its them. Particularly when
RS> Paul now says they appear to have gone from his callers now. Are you
RS> still getting CRC errors with Paul now that he has upgraded Bink ?

NP> Not for two calls, then on the third it cropped up badly.
NP> CRC error and the "Nonsense end of file received. Ignored"

Also most unlikely to be line/exchange because any line errors
should be transparently fixed between the modems at the LAPM level.

NP> Down here, it appears to be more of a line
NP> problem than an operating system problem

RS> It shouldnt be if you achieving LAPM or MNP4 sessions. That error
RS> correction should be complete transparent between the modems and you
RS> shouldnt see any CRC errors at the file transfer protocol level at all.

NP> CONNECT 14400 /LAPM/ v42 bis

RS> Yes, but we dont know that thats due to LINE ERRORS. The should be being
RS> transparently fixed between the modems by the LAPM error correction and
RS> not visible as CRC errors on the ZedZap status screen at al.

NP> They appear.

Yes, they appear, no, you do not know they are due to LINE ERRORS.

NP> Teddy cps rate is always down 1500's.

And thats not a good measure of line errors either, if its ALWAYS
that speed, regardless of when you call, it can be just because
her system doesnt put the byte stream out quite so quickly.

NP> Others are 1615 - 1670 cps

Much more likely to be her system. Presumably still an Amiga ?

NP> Anyway ... All 4 com ports have 16550's

RS> Yeah, cant be that. There is a slight possibility that you have
RS> an interrupt latency problem due to a cache particularly, but that
RS> shouldnt normally give varying results depending on who you call.

NP> Soon to have a dedicated 3meg HD cache on the controller.

RS> Looks more likely it the system you are calling with the problem.
RS> And pretty convincing evidence if you no longer get CRC errors with Paul.

NP> Unfortunate they are still there occassionally.

Yeah, you do appear to have a problem. But it cant be a line/exchange
problem coz line errors are transparently fixed between the modems at the
LAPM level and so arent visible as CRC errors at the file protocol level.
@EOT:

---