What's the problem?

When downloading mail, there are repetitions every 10240 bytes of information. Why should we sin? On Melon or modem?
Sincerely, SP/PoS.

It was 27 Nov 98 at 00:38:34, when
Konstantin Samoylov nevertheless risked writing All
about What is the problem?
I simply could not notice this

KS> When downloading mail, there are repetitions every 10240 byte

Repetitions? Or is the file received in blocks of 10240 bytes?
If in blocks, then I don’t see anything wrong with it...
It all depends on the boss’s mailer.
I’ll give an example: my boss had FrontDoor, the length of the 1st block was
1024
When I switched to T-Mail, it became 10240 bytes, which only increased the
exchange speed data.

KS> information. Why should
KS> sin? On Melon or a modem?

Imho Melon and a modem have nothing to do with it...
╟─┤asta la vista !!!

[SG] [Clawfinger] [Megadeth] [SSAU] [ZX] [Beer]

√ E-mail: w...@chat.ru <-> POWer/WDG

Looks like Melon. I had such a bug when I downloaded mail in Zmodem Fast mode. And I thought that this was a problem with my node.

Good Bye, Konstantin Samoylov! With best wishes, Vova Bakum.

... "Long memory is worse than syphilis..." (c) 1980 by BG.

Once on 27 Nov 98 at 00:38:34 comrade Konstantin Samoylov noticed comrade All
about What is the problem? which forced me to be the third

KS> When downloading mail, there are repetitions every 10240 bytes of information. Ha
KS> what

OP> Repetitions?
My T-Mail says that there is a repetition from such and such a position.

OP> Or is the file received in blocks of 10240 bytes?
There are no blocks of that length. There are 64,128,256,512,1024,2048,4096,8912
For ZMODEM, the block length is 1024 bytes, this is not 10240, maybe 10
times larger. :)

OP> If it's in blocks, then there's nothing wrong with that...
OP> It all depends on the soaper standing at the boss.
T-Mail.

OP> Let me give you an example: my boss had FrontDoor, the length of the 1st block
OP> was 1024
OP> When transition to T-Mail became 10240 bytes, which only increased
OP> the data exchange speed.
It did not increase anything, after the repetition occurs, CPS drops and all mail
is taken to 170 CPS. Is this really fast?!
See. above.

KS>> information. Why

One day on the 28th of Nov 98 at exactly 21:03:10 Oleg Pritula
wrote to Konstantin Samoylov on the topic 'What's the problem?':

OP> When switching to T-Mail, the bytes became 10240, which only
OP> increased the speed of data exchange.

... And if an error occurs, then The block will pull first...
And even a modem without protocols. I have already encountered this :-(

With best wishes, Dmitry.

Once on 29 Nov 98 at 00:24:14 there was a conflict
Denis Dmitriev and Konstantin Samoylov were arguing on the topic What is the problem?
I decided to continue this dialogue:

KS>> When downloading mail, there are repetitions every 10240 bytes of information. Ha
KS>> what

VB> Looks like Melon. I had such a bug,
How did you get rid of it then?

VB> when I almost downloaded in
VB> Zmodem Fast mode. And I thought that this was a problem with my node.

With respect, SP/PoS.

Once, 29 Nov 98, someone Denis Dmitriev wrote to Konstantin Samoylov
on the topic 'What's the problem?'...

KS>> When downloading mail, there are repetitions every 10240 bytes of information.
DD> On Melon.

What scheme is the modem based on and what version of Melon? Use 1.80p

Sincerely, Evgeniy aka EVP.

VB> Looks like Melon. I had such a bug when I downloaded
VB> mail in Zmodem Fast mode. And I thought that there was a problem with my
VB> node.

Are you sure that this is a bug? In the help it is written in Russian that the fast
mode forces the sending computer to transmit the block continuously, so it transmits, and why
be offended by this?

With best wishes, Dmitry.

Here someone said that there are no lengths of received blocks of 10240, I hasten
to disappoint, there are...
And the increase in speed occurs due to the continuity of the flow of received
data, compared to the intervals with the calculation of the checksum that occur with
blocks of 1024 b.

DS> ... And if an error occurs, the block will first pull...
DS> Yes also a modem without protocols. I have already encountered this :-(

Yes, in this case it is not good...
What can I say, only purchasing a modem with an error correction protocol can help here.
At least your nerves are more expensive...

KS> byte of information.

EP> What is the scheme of the modem and what version of Melon? Use 1.80p

Strange, I observed the same phenomenon during sessions with DT - its modem was
connected using an interrupt. Moreover, on both sides only
v42bis was used (MNP5 is basically always turned off for me - anyone who understands understands why).

Sincerely, Alexey

DD>> On Melon.
AI> You’d better tell me when the new one will come out, have you already given up on this matter?
I won’t say when;)
but it will come out. for one simple reason - well, I don’t like izdos!!! ;))))

With the best, Denis Dmitriev.

... I’ll buy a moonshine still. intermediary phone number 02.

One day, 03 Dec 98 at 01:58:55, comrade Oleg Pritula noticed comrade
Dmitry Sharikov about “What is the problem?>, where I intervened:

OP> Here someone said that there are no lengths of received blocks of 10240, I hasten
OP> to disappoint, there are...

It doesn’t happen!
10240b is the size of the internal buffer in Melon, = 5*2048
brakes due to repetitions - at the moments of sending received data from the reception buffer
to the file buffer (at the same stage writing to disk occurs), in this case part of the
information received at this time is lost, and you have to request a repeat of the position.
All this is due to unoptimized procedures of the Zmodem protocol in this
version.

Thursday December 03 1998 08:33, Alexey Ivanov wrote to Eudgine Palenock:

AI> only v42bis was used (MNP5 is basically always
AI> turned off - whoever understands understands why).

I apparently don’t understand well and I’m interested. Why?

AI> Sincerely, Alexey

With respect, Alex Noman.

AI> is constantly turned off - those who understand understand why).

AN> Apparently I don’t understand it well and I’m interested. Why?

Too lazy to read the description of the protocols in the modem book?

Because MNP5 always compresses data with hardware. If you transfer an already archived file (and most of them are like that), the brakes begin, because momed
tries to compress already packed data.

V42bis is an intelligent protocol for error correction and data compression.
If the data is compressed, it does not pack it, if not (bare text),
momed it packets in hardware and, accordingly, the transfer speed increases.

Sincerely, Alexey

DD> I won’t say when;)
DD> but it will happen. for one simple reason - well, I don’t like izdos!!! ;))))
Me too so-so. I’ll really wait, I liked Melon, but it’s a little buggy
:( . I hope there will be a new version without this, otherwise with a modem on 14400 I’ll set the port
to 33600 and... More than half of the letters were gone... And on 57600 it refused to work at all, and I should have set it to that much.

All the best, Tigrr.
[Team NeOS] [Team GOA] [Team TRANCE] [Team PRODIGY]
<< Brainwave Association >>

KS> How did you get rid of him then?
No way. I wait until the entire file is completely swallowed. There seems to be a complex system in Melon... Although, most likely, the problems are with the ZModem protocol in Melon.

P.S. You can leave the ZFast-type installations in Melon alone - it’s useless. The mailer makes its own settings. ;)

2Denis Dmitriev: Say something worthwhile... ;)

VB>> in Zmodem Fast mode. And I thought that this was a problem with my node
VB>>.
DS> Are you sure that this is a bug? In the help it is written in Russian,
DS> that the fast
DS> mode forces the sending computer to transmit the block continuously, so
DS> it transmits, and why
DS> be offended by this?
But no! The sending computer transmits to me in Fast mode, but I cannot
enable _any_ settings (Fast, Cheat, CRCW, etc.) when
downloading mail. So it turns out that Melon sets his own settings for ZModem. It accepts 1Kb as it should, and when Fast floods
it can sometimes cycle (in short, slow down).
There can be a global problem if the connection is established
without protocols (or a modem like that;) - the brakes are wild.
But in general - IT WORKS! ;)
P.S. For reference, I am using version 1.8special

Good Bye, Dmitry Sharikov!

There was nothing to do, and 28 Nov 98 at 23:27:03
Vova Bakum wrote a letter for Konstantin Samoylov
I saw this and decided to answer.

KS>> When downloading mail, there are repetitions every 10240 bytes
VB> Looks like Melon. I had such a bug when I downloaded mail in
VB> Zmodem Fast mode. And I thought that this was a problem with my node.

How did you deal with this?

For this, let me bow out. Sincerely, Andrey Khoroshilov.

··· The mind is stronger than the chest. Hippies are people too.
··· What were you going to see there? Glass fabric? And thought from electrons?

OP> Yes, in this case it is not good...
OP> What can I say, only purchasing a modem
OP> with an error correction protocol can help here.
OP> At least the nerves are more expensive...

I didn’t buy a modem without protocols, but for a gift horse....
;-)

With best wishes, Dmitry.

Wednesday December 02 1998 12:28, Dmitry Sharikov ─── Vova Bakum:

VB>> in Zmodem Fast mode. And I thought that this was a problem with my node.

DS> Are you sure that this is a bug? In the help it is written in Russian that the fast
DS> mode forces the sending computer to transmit the block continuously, so it transmits, and
DS> why be offended by this?
Well, it can’t do it completely continuously, only within one window. Before
this, the receiver must inform the transmitter the size of its receiving buffer.
Then, then, continuous transmission of blocks within this window will take place.
Absolute continuity can be achieved only in one case -
if the receiver sets the flag “I can redirect the data flow directly to disk"
(and this is fundamentally impossible on the Spectrum), and the transmitter is capable of operating
in full duplex - reading data from the input stream simultaneously with sending
(this is possible on the Spectrum, but only with the Kondratieff connection scheme
and using interrupts (which Melon does not do maybe
principally)).

Good luck! Evgeny.

■ The night is shorter than the day... The day will kill me... The world of illusions burns down in it. (C) Aria.

AI> Are you too lazy to read the description of the protocols in the modem book?
No, that’s exactly why he’s asking. :)))

AI> Because MNP5 always compresses data with hardware. If you are transmitting already
[.....]
AI> If the data is compressed, it will not package it, if not (bare text),
AI> the momed will package it in hardware and, accordingly, the transmission speed will increase.
Yes, but if for one reason or another you have a bad connection (or something else), then if v42b will not turn on, you will have "NONE" and instead of a reliable but correct transmission, you will receive... a cookie - there will be no guarantee that the information
is not distorted. And if the connection is normal, then... then even with MNP5 enabled, it
is unlikely to ever turn on...

VB>> in
VB>> Zmodem Fast mode. And I thought that this was a problem with my node
VB>>.
AK> How did you deal with this?
I, briefly, already answered this question in the echo.
In general, nothing. ;)
I just waited until Melon “wasted” the entire file.
Note: I used version MELON 1.8s. It works better than
1.79 ;)
I emphasize: If you receive mail in FAST mode (your boss chose ZModem Fast, _but_not_you_) without protocols, and even with poor
connection quality, repetitions are inevitable. (Why? I won’t explain
;)
It so happened that I received mail not in Fast (again, my boss
wanted it that way) - without any problems at all, especially at low speed
(2400 both with V42bis and without protocols;).

Good Bye, Andrey Khoroshilov!

DD>> izdos!!! ;))))
IT> I'm also so-so. I'll be waiting a lot, I liked Melon
IT>, but it's a little buggy
A little? 8-() Everything (seemingly;) is in order... Hm... ;)
IT> :( . I hope there will be a new version without this, otherwise with a modem
IT> at 14400 I set port
IT> at 33600 and...
I understand, but without interruptions, well, _no_ it will not last at such
speed (in a turbo no more 19200)...
IT> More than half of the letters were gone...
Otzh. ;)
IT> And on
IT> 57600
IT> refused to work at all,
In Melon, forget about this loading speed altogether - it doesn’t work.
With interruptions - easily (without a turbo;)
IT> but you should set it to less.
But this one I didn't understand the phrase, sorry. :(

IT> * Origin: Beer is better than water... (2:5052/22.18)
That's for sure! :)

Good Bye, Igor Turashev!

Once, 05 Dec 98, someone Vova Bakum wrote to Dmitry Sharikov
on the topic 'What's the problem?'...

VB> for ZModem. It accepts 1Kb as it should, and when Fast fills
VB> - sometimes it can cycle (in short, it slows down).

Fast works fine for me... Although IMHO there is one glitch...
I observed it when they brought me a 2400 modem for testing...
IMHO Fast's port speed should be two to three times higher than the
connection speed, otherwise while the file is being written by the modem hangs...

VB> if the connection is established
VB> without protocols (or such a modem;) - the brakes are wild.

How can there be a connection without protocols? What modem do you have?

VB> P.S. For reference, I'm using version 1.8special

Doesn't it add the keyboard driver after Upload?
Do you have 512 kB?

07 Dec 98 09:43, Evgeny Milun -> Alexey Ivanov:

EM> Yes, but if for one reason or another you have a bad
EM> connection (or something else), then if v42b does not turn on, you will have
EM> "NONE" and instead of a reliable one, but correct transmission, you will receive...
EM> you sneer - there will be no guarantee that the information is not distorted. And if
EM> the connection is normal, then... then even with MNP5 enabled, it is unlikely
EM> it will ever turn on...

It will never be possible to prohibit a connection attempt without v42b and NONE (AT&M5 - for
USR modems).

In the case of NONE, the modem itself will hang up quit.

Sincerely, Alexey

DS>> he conveys it, and why be offended by it?
VB> But no! The sending computer transmits to me in Fast mode, but I
VB> cannot enable _any_ settings (Fast, Cheat, CRCW
VB>, etc.),

why can’t you turn on ?

VB> when I download mail. So it turns out that Melon sets
VB> his settings for ZModem. 1Kb at a time accepts as it should,
VB> and when Fast floods it can sometimes cycle (in short
VB> slow down).
VB> There may be a global problem here if the connection is established

DS>> Are you sure this is a bug? In the help it is written in Russian
DS>> that the fast mode forces the transmitting computer
DS>> to transmit the block continuously, so it transmits, and
DS>> why be offended by this?
EM> Well, it cannot do it completely continuously, only within
EM> one window. Before

I had this in mind... ;)

With best wishes, Dmitry.

╟─┤asta la vista !!!

[SG] [Clawfinger] [Megadeth] [SSAU] [ZX] [Beer]

√ E-mail: w...@chat.ru <-> POWer/WDG

EM> Well, he can’t do it completely continuously, only within one window. Before
EM> this, the receiver must inform the transmitter the size of its receiving buffer.
EM> So, then, there will be a continuous transmission of blocks within this window.

... If the transmitter does not care what was told to him there. More than once I have come across protocols that _always_ work fast. And in soapers, in general, it is established by hand
for which link to poke fast, and for which - not fast. :(

EM> and using interrupts (which Melon cannot do
^^^^^ sux :(

EM>> one window. Before
DS> I had this in mind... ;)
Well, as far as I understand, not quite. You said that there are no glitches
(repeat after 10240 bytes), and all are the consequences of continuous transmission. I
say that continuous transmission (if it is done without glitches) has no
consequences.
Or are you already talking about something else, and I fapped my ears? :((

VB>> when I download mail. So it turns out that Melon sets
VB>> his settings for ZModem. It accepts 1Kb as it should,
VB>> and when Fast floods it can sometimes cycle (in short
VB>> slow down).
VB>> There may be a global problem here if the connection

EM> Well, as far as I understand, not quite. You said that
EM> there are no glitches (repeat after 10240 bytes), and all are
EM> consequences of continuous transmission. I

I said that Melon transmits continuously (naturally within one window),
as he was asked, i.e. the checkbox is checked

EM> Or are you already talking about something else, and I fapped my ears? :((

no, it seems, it’s all about the same thing...

With best wishes, Dmitry.