Holy shit!

04 Dec 99 20:41, Alex Noman -> Alexey Ivanov:

FK> withstood more than 21600, set 33600 ;). then
FK> was picking up his jaw from the floor for a long time when he saw how at
FK> at a speed of 31200 a file of 200
FK> kilos in size with CPS=4000 flew towards me... 8-( )

AI>> Nothing surprising. There was an unpacked file. Its
AI>> modem
AI>> was compressed by hardware to v42bis.

AN> What does it have to do with v42? The point is that the connection was 31200, port
AN> (one must think) 57600 or 115200, and it was with this CPU that the file
AN> flew to the port. And whether the modem pressed him or not - one horseradish...

Read the top paragraphs more carefully and you will understand. With a connection speed of
31200, its cps is too high - the unpacked text/non-soap file was
compressed by the modem in hardware. Because of this, such a high speed appeared.
In reality, all logs show cps not transmissions to the port, but transmissions by the modem to the line.
Statistics are taken from the modem registers. The computer-modem connection is not shown in the logs and does not participate.

Sincerely, Alexey

AI> Read the top paragraphs more carefully and you will understand. With a connection speed
AI> of 31200, its cps is too high

What difference does it make whether the cps is high or not (especially since the cps is the left one)?? :)) These are all details and details that do not play any role. The point is that the
_Spekov_ terminal _doesn't_ glitch and doesn't lose data at _any_ speeds.
unlike analogues. ;). Because of this,
AI> seemed to have such a high speed. In reality, all the logs show cps not
AI> transmissions to the port, but modem transmissions to the line.

Oh, well? ;))

AI> Statistics are taken from the modem registers. The computer-modem connection is not shown in
AI> logs and is not involved.

I'll know. :)) That is The modem keeps statistics, and Evgeniy Milun's program accesses it all the time. So? :)

2EM: I knew, of course, about your extraordinary approaches to writing programs, but not to the same extent... :))

AI> Sincerely, Alexey

ps/ Tell me more about these registers. :)
pps/ And what kind of “computer-modem” connection is this? :) Explain what you mean, plz.

With respect, Alex Noman.

AI> Read the top paragraphs more carefully and you will understand. When the connection speed
AI> is 31200, its cps is too high - the unpacked text/non-soap file
AI> was compressed by the modem in hardware. Because of this, such a high speed appeared.
AI> In reality, all the logs show cps not transmissions to the port, but transmissions by the modem to the
AI> line.
Nothing of the kind. Just how and when the modem transmits something to the line, you actually cannot find out by any means. The only thing
you can do is an indirect method - analyze after what time the modem is again
ready to receive data from the modem.

09 Dec 99 14:58, Alex Noman -> Alexey Ivanov:

AI>> Statistics are taken from the modem registers. The connection
AI>> computer modem is not shown in the logs and is not involved.

AN> I'll know. :)) That is The modem keeps statistics, and the program
AN> Evgeniy Milyna accesses it all the time. So? :)

But I don’t know how Evgeniy’s program takes statistics.

AN> ps/ Tell me more about these registers. :)
AN> pps/

Did you bother reading the description of the modem registers?

AN> And what kind of “computer-modem” combination is this? :)

Sorry, I thought everything was clear and it would be so. This means a computer-modem, not a modem-telephone line.

Sincerely, Alexey

09 Dec 99 21:41, Evgeny Milun -> Alexey Ivanov:

EM> Just how and when the modem
EM> transmits something
EM> to the line, you actually cannot find out
EM> by any means. The only thing you can do is an indirect method -
EM> to analyze after what time the modem is again ready to receive
EM> data from the modem.

So is this shown to you during a communication session or after its completion?
I believe that the most correct statistics are issued by the modem after completion
of a communication session that lasted at least minute.

10 Dec 99 11:49, Alexey Ivanov -> Evgeny Milun:

AI> after completing a communication session that lasted at least a minute.
AI> For example, USR via ATIn

AI> Maybe I didn’t understand what we were talking about, but the point was that with uncompressed
AI> in the CPC files between the computer and the modem exceeds the real CPS

The original message was about how fast the interface works on
the real spectrum...
For example, when connecting a modem via SMUC on a non-tuned Scorpio, it is not higher than
2000 cps pulls...

With respect, Evgeniy aka EVP.

AI> with uncompressed files, the CPC between the computer and the modem
AI> exceeds the real CPS (modem-telephone line) by almost 4
AI> times.

Yes, all this is understandable. The point of the subject is that there WAS a connection 31200
on a real Speck! Those. The modem's com port was locked to 56K (!)
and no bytes were lost. That's it. :)

Sincerely, Felix.

[I ZX]

AI>>> Statistics are taken from the modem registers. The connection
AI>>> computer-modem is not shown in the logs and is not involved.
AN>> I'll know. :)) That is The modem keeps statistics, and the program
AN>> by Evgeniy Milyna refers to it all the time. So? :)

Yeah, then these are S-registers. This is exactly what I was talking about - their presence/
absence or variety is the envy of the developer companies, and is not included in the
HAYS-compatible modem standard.

AI> Maybe I didn’t understand what we were talking about, but the point was that with uncompressed
AI> files there is a CPC between the computer and the modem exceeds the real CPS
AI> (modem-telephone line) by almost 4 times.
Absolutely right. It's really you who just haven't figured it out. Alex
this is exactly what I saw - that the DCT-DTE (modem-to-computer) exchange rate was exactly this - ~3500 DSP (or how many were there?)

AI> Sorry, I thought everything was clear and it would be so. We mean a computer-modem, and not a modem-telephone line.

It just seemed inappropriate to me to mention the “computer-modem” connection. She just doesn't roll there. We were talking about the COM port after all. Or was I the only one talking about him? :)

AI> Sincerely, Alexey

With respect, Alex Noman.

I think so.

AI> but the point was that with
AI> uncompressed files, the CPC between the computer and the modem exceeds the real
AI> CPS (modem-telephone line) by almost 4 times.

Even if everything is so, that’s not what we’re talking about. G)

Sunday December 12, 1999 00:51, Felix Knajzev wrote to Alexey Ivanov:

FK> Yes, all this is clear. The point of the subject is that there WAS a connection 31200
FK> on a real Speck! Those. The modem's com port was locked to 56K (!)
FK> and no bytes were lost. That's it. :)

And at 115200 there are no losses, I hope?

FK> Sincerely, Felix.

With respect, Alex Noman.

AN> That is. Do you think that all programs always do ATIx after the end of the
AN> communication session? :)
By the way, I checked it yesterday. _After_ - no longer gives anything. It is necessary exactly
_during_the_session_...

AN> Do they send pluses during the transmission? :)
By the way, yes. :). :)
Wow!!! Can you imagine if we do something like this? :))

14 Dec 99 19:36, Evgeny Milun -> Alexey Ivanov:

EM> Absolutely right. It’s really you who just didn’t
EM> figure it out. Alex saw exactly this - that the speed
EM> exchange of DCT-DTE (modem-computer) was exactly this - ~3500
EM> DSP (or how many was there?)

4000. Everything is fine. I reacted precisely to the fact that a man was surprised by such a cps...

Sincerely, Alexey

EM> Calculating time through interruptions and dividing by this the length of the incoming
EM> block.

Have you tried to take time from space?

Kirill Frolov. [ZX]

EM> Absolutely right. It's really you who just haven't figured it out.

Stop! The DSP there does not exceed the real DSP. THIS IS THE REAL DSP!

EM> Alex saw exactly this - that the exchange rate of DCT-DTE
EM> (modem-computer) was exactly this - ~3500 DSP (or how many were there
EM>?)

Computer-computer was about the same!
And between the modems Fewer bits were transmitted.

Kirill Frolov. [ZX]

AN>> if they hadn’t, they wouldn’t have done that. :)
EM> Wow!!! Can you imagine if we do something like this? :))

I tried it. It didn't work out. ;) For some reason, the connection is not always restored later. :)
Probably not enough experience. ;)

EM> Good luck! Evgeny. (Aka 2:5057/18.73)

With respect, Alex Noman.

15 Dec 99 09:51, Alex Noman -> Alexey Ivanov:

AN> It just seemed inappropriate to me to mention the link
AN> “computer-modem”. She just doesn't roll there. After all
AN> we were talking about the COM port. Or was I the only one talking about him? :)

Same eggs, only two COM ports if the modem is external.
Those who need it understand what we’re talking about. Or are you such an academician that you need to describe all things
in an encyclopedic dictionary, and not on your fingers?

Sincerely, Alexey

AN> And at 115200 there are no losses, I hope?

See. origin. ;)

In general, this is a question for the author of S-Terminal and owners of 33600 modems.

AI>>> But I don’t know how Evgeniy’s program takes statistics.
EM>> Calculating time through interruptions and dividing by this the length of the incoming
EM>> block.

AI>>> Maybe I didn’t understand what we were talking about, but the point was that with
AI>>> uncompressed files, the CPC between the computer and the modem exceeds the real one
AI>>> CPS (modem-telephone line) by almost 4 times.
EM>> Absolutely right. It’s you who really just didn’t understand.

KF>> file and the number of bytes already transferred.
EM> Yes, this is better, but the calculation will take longer.

And since you wrote it completely wrong. Pauses between blocks do not count.

And don’t you care how long the calculation takes? At 2400, 90%
of the time is still spinning in place. And the HALT of the ztnik is incorrect...

Kirill Frolov. [ZX]

EM> No, due to lack of space. :)

And if kmos is present?

Kirill Frolov. [ZX]

FK> See origin. ;)
FK> In general, this is a question for the author of S-Terminal and the owners of modems on
FK> 33600.

Precisely for the author. ;) Because there are no hardware limitations here.

AI> Same eggs, only two COM ports if the modem is external.

Hmm... no comments...

AI> Those who need it, understand what we’re talking about.

You, apparently, don’t really need it.

AI> Or are you such an academician that you need to describe all things in an encyclopedic
AI> dictionary, and not on your fingers?

Because I’m not Academician, I didn’t understand the meaning of this phrase at all. Sorry. And it’s definitely not me who needs to describe everything with my fingers. It just turns out that I’m talking about Thomas, and you’re talking about Yerema. Let's change it up, Lesh.

EM>>>> Calculating time through interruptions and dividing by this the length of the incoming
EM>>>> block.
KF>>> And did you try to take time from space?
EM>> No, due to the lack of space. :)

EM> Well, I don’t know my own computer? :)

Have you tried to connect the plugin with kmos? Helps....

But seriously, not only does it work on your computer.

Kirill Frolov. [ZX]

FK>> at 33600.
AN> Exactly to the author. ;) Because there are no hardware limitations here.

And at what frequency in your spectrum does the ztnik work?

Kirill Frolov. [ZX]

KF> Did you try to connect the plugin with kmos? Helps....
And to what place in the scorpion does it connect? :)

KF> But seriously, not only does it work on your computer.
Let's do it. Do you remember somewhere at the end of summer there was a subject here called “Baytor-tearing spectacle”? So, I decided to finish my scorpion. :)

FK>>> In general, this is a question for the author of S-Terminal and the owners of modems
FK>>> at 33600.
AN>> Specifically for the author. ;) Because there are no hardware limitations here.