TIP: Click on subject to list as thread! ANSI
echo: locsysop
to: Bill Grimsley
from: Rod Speed
date: 1996-04-20 07:58:08
subject: When only the best wil do

BG> and don't forget that a 5dB drop is as near
BG> as dammit to HALVING the signal level.

RS> Irrelevant, V34 is designed to be able
RS> to handle those sorts of differences.

BG> Sure, by lowering the symbol rate and frequency,

Nope, if you have a decent bandwidth, and a good enough S/N ratio,
the absolute signal level isnt anywhere near as important as you
might think with V34. Thats another area where modern modem
protocol designs are VERY different to the older protocols.

Its all about S/N and bandwidth now, NOT the absolute signal level.

And in fact even with the fax V29 and V17 you DELIBERATELY reduce the Tx level.

Bob totally mangled the story on Tx levels.

BG> which also lowers the bit rate at the same time.  So HALF the signal
BG> strength is NOT irrelevant at all, not when my lines are shit to begin with

Fraid so, you havent grasped that its all about
bandwidth and S/N now, not the absolute signal level.

In fact there is DELIBERATELY a loss of 6db in the digital network.

RS> In fact, once USR fixed the code in the Sportster, it even
RS> handles MUCH more than that due to YOUR rather poor line as well.

BG> Still doesn't explain why I always get 33k6 connects with
BG> Dave, and between 26k4 and 28k8 connects with Paul though,

Sure it does, Daves line is IMMACULATE. Pauls is fine, a normal line.
With your rather poor line the overall result to Dave is rather better
BECAUSE his line is so immaculate, doesnt make your pretty gross line
any worse. Pauls has the normal characteristics, when added to YOUR
rather poor line the result isnt that terrific. The problem is YOUR
line tho, not Pauls. That can be seen VERY clearly from the calls
Russ made from FNQ to both you and Paul with a USR at the other end.

BG> and that's with the NEW code.

Yes, your line is poor enough that 33K is a problem
when there is a normal line at the other end.

BG> With my dreadful lines, that's more than
BG> enough to create major problems to Paul.

RS> Yes, but thats due to YOUR dreadful line, nothing
RS> to do with Paul which is well within the normal range.

BG> It's JUST within Telstra's accepted limit of
BG> 7dB between sub and exchange, by only 1-2dB.

You are still using the WRONG measure of a line, its NOT about the
absolute signal level, tho that is certainly the measure that Telstra
uses. Its all about bandwidth and S/N now, and those to even interact
very considerably too, allowing Russ to get glorious sessions from
FNQ to Daves immaculate line, even with audible S/N problems, because
that is compensated for by an EXTREMELY flat and wide bandwidth curve.

You really want to try graphing the ATY11 data, is a HELL of an eye
opener and there is a VERY clear correlation between thruput and the
shape of the curve and the S/N data if you understand the principles.

BG> And with my poor lines, that's enough to make a difference.

Nope, its got fuck all to do with it. If it was that
simple, you wouldnt deliberately reduce the transmit
levels for fax. By a HELL of a lot more than 1-2 db too.

RS> Its just not as good as the bulldogs, but
RS> then neither is mine either. Or anyone elses.

BG> Oh crap,

Fraid not, we have MEASURED Pauls line with a Courier on it when
Russ did the FNQ calls. We KNOW what his line is like with real
rigorous science, and we KNOW your line is MUCH worse than his.

BG> anybody who lives as close to an exchange as David
BG> does is going to have equally immaculate lines.

Aint necessarily so. There are other factors like the exchange technology.

BG> I look forward to reading it.  Let's just hope that it gets up here...

RS> Yeah, I have been holding off posting it till Dave Hatch
RS> returned from the dead. It has today, so I will post it
RS> in a day or so when things have settled down a bit more.

BG> Given the current (apparently normal) flow of mail,
BG> it seems that I wasn't missing too much at all.

Yeah, I havent seen any sign of it, tho I havent even
read all the backlog in the Aust echoes alone yet.

Unfortunately I am reading backwards from the latest mail
packets, so its likely I will miss noticing the usual sign of
lost mail, replys to stuff you have never seen the original of.

BG> Mainly stuff from TML and those few boards which use the same local feed
BG> that Paul does (I gather it was Dave Hatch).

Yeah, Dave actually has TWO systems, 808 and 809, the mass mail flows
thru 808. 809 is where Paul gets his, and loons like Grigor Scott.
Tony also had a massive system crash, dead hard drive and all too.

Odd that Easter affected the fundys so badly. You do
realise that Dave Hatch is some form of fundy dont you ?

BG> OS mail was likely fucked as well, but as I only
BG> have HS_Modems tagged, I wouldn't notice that anyway.

I get quite a few and there has been a big surge of that coming in too.
Some have always been very very lumpy, massive block at weekly intervals
isnt that uncommon in some.

Are you saying that your HS_Modems has US mail in it ?
The one we get here never does, tho it did at one time.
@EOT:

---
* Origin: afswlw rjfilepwq (3:711/934.2)
SEEN-BY: 711/934
@PATH: 711/934

SOURCE: echomail via fidonet.ozzmosis.com

Email questions or comments to sysop@ipingthereforeiam.com
All parts of this website painstakingly hand-crafted in the U.S.A.!
IPTIA BBS/MUD/Terminal/Game Server List, © 2025 IPTIA Consulting™.