Test v.4.30 & my Pentagon 512

And one more thing: I have a capacitor C6 (Pentagon circuit) with a capacity of 0.83 uF (according to the 22n circuit), if you set it less, the keyboard (film extended) freezes, andNow
some signs are slowing down. I tried the int scheme with a trigger (ZX-Review 95),
I didn’t notice any changes. I can’t understand “where the dog is buried”? :(
Yes, regarding the test - I would like to know your opinion on what I read,
constructive suggestions.

With best wishes, Eduard Pilipenko.

And one more thing: I have a capacitor C6 (Pentagon circuit) with a capacity of 0.83 uF (according to the 22n circuit), if you set it less, the keyboard (film extended) freezes, andNow
some signs are slowing down. I tried the int scheme with a trigger (ZX-Review 95),
I didn’t notice any changes. I can’t understand “where the dog is buried”? :(
Yes, regarding the test - I would like to know your opinion on what I read,
constructive suggestions.

With best wishes, Eduard Pilipenko.

22 Nov 99 23:51, Eduard Pilipenko wrote to All:
EP> Here's the thing. Checking my computer with a subject test, I received
EP> the following data:

EP> Not complete decryption of port #7FFD

Quite typical of the Pentagon.

EP> Test of port #7FFD Reading test is impossible

And this is an improvement. Unnecessary.

EP> Takt/INT:71680, V=102.31538%

The test was probably done on a brake spool. 100% however.

EP> Length of INT:60 Cycles - this value changes to 64

Up to 70 on the Pentagon is the norm. Although this test measures crookedly,
take Kovalevsky’s 2-test, nothing better has been done yet.

EP> (i.e. not stable)

Nothing like that. They cannot measure more precisely 4 clock cycles, hence
and the fluctuations within 4 clock cycles.

EP> Low Mem speed: 71682-71673

I wonder what can be measured there???

EP> High Mem speed: 72-71673 - here it is synchronous with Length of INT

Well, not surprising. The tests work simultaneously.
Only your Hi and Low must match.

EP> changes to 71682-71673 with the following message:
EP> This Spectrum has a separate memory field. Frame length
EP> is less than 20ms. You have a general memory field without inhibition.

Glitches. Definitely.

EP> After checking TR-DOS:
EP> The data bus of this Spectrum in IM2 mode in TR-DOS is clogged

A problem for all pentagons.There the Beta-Disk is connected via a buffer
and they forgot to put resistors in it.

EP> And one more thing: I have a capacitor C6 (Pentagon circuit) with a capacity of 0.83 uF
EP> (according to the circuit - 22n), if you set less - the keyboard (extended film)
EP> freezes, and even now some the signs are slow. I tried the int circuit with
EP> trigger (ZX-Review 95), I didn’t notice any changes. I can’t understand “where is
EP> the dog buried”? :(

IMHO already ruki.sys

EP> Yes, regarding the test - I would like
EP> to know your opinion on what you read, constructive suggestions.

Aim the sight and press F8.

Kirill Frolov. [ZX]

EP>> changes to 64 Cycles
FK> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
FK> Here's the trouble. Your INT is too long. It needs to be somewhere around 24-28-32. 28
FK> is ideal.

For the Pentagon and 70 is the norm!

EP>> (i.e. not
EP>> stable)
FK> It happens. One of the reasons is poor TKE of the capacitor in the diff circuit for
FK> INT formation.

The reason is in the meter. She can’t accurately measure.

And where have you seen capacitors like that, so that the capacitance trembles?

Of course, there are “microphones”...

EP> 0.83 uF (according to the diagram - 22n), if you set it less - keyboard
^^^^^^^^
Something is wrong. :) There is no such denomination. Maybe 0.33uF?

EP> (film extended) freezes, and even now
EP> some signs are slowing down. I tried the int circuit with the
EP> trigger (ZX-Review 95), I didn’t notice any changes. I can’t understand “where is
EP> the dog buried”? :( Yes, regarding the test -

Reduce the duration of INT by selecting a capacitor.

EP> I would like to know your opinion on what you have read,
EP> constructive suggestions.

See above.

+2Author of the test: The test reports that I have a Pentagon-512 (the memory is really
512K). :) Although it writes that "... the total memory field _with_braking_...". Pentagons
- they are wait-free, unless, of course, they are turbocharged. I have “Leningrad” aka Scorp.

Reason: I just have a quartz resonator at 14.5 MHz, and not 14.0 MHz
exactly, as in Scorps, i.e. The car is slightly turbocharged, which is what the performance test shows. It should be recorded. ;)

Sincerely, Felix.

[I ZX]

FK>> It happens. One of the reasons is the poor TKE of the capacitor in
FK>> the diff circuit for
FK>> INT formation.

KF> The reason is in the meter. She can't measure accurately.

I know. +/- 4 ticks. NOP measures.

KF> And where have you seen such capacitors so that the capacitance
KF> trembles?

Do you know TKE 'H90'? :) I have one like this. In summer, INT
generally “leaves” to 22 cycles (decreases). :)

EP>>> The data bus of this Spectrum in IM2 mode in TR-DOS'e
EP>>> is clogged
FK>> This is bad. Loaders with IM2 (music loaders)
FK>> will fail
FK>>. :(

KF> Maybe they won’t. A 257-byte plate needs to be made for trdos
KF> too.

Well, our programmers, you know, write for the hardware that they have. In
in any case, the hardware needs to be straightened out.

FK>> Run
FK>> a signal on the TR-DOS input-output device decoder /M1
FK>> of the processor. Most likely, the feces are read from the #FF port (6&7
FK>> bits)
FK>> TR-DOS.

KF> That's right. You just need to start /RD instead of /M1.

What's the difference :) On VG93 /RD works 100%. But when reading #FF,
only /WR and /IORQ are used, if done there on ID7.

KF> take them Kovalevsky’s 2-test, nothing better has been done yet.

PLZ, throw it in the echo. It will be useful to many.

Sincerely, Alex from FFC COMPUTERS.

>E-MAIL: asde...@softhome.net ICQ: #13064553

My on-board systems detected that on Friday November 26, 1999 01:18, Felix
Knajzev wrote to Eduard Pilipenko:

FK> +2 Test author: The test reports that I have a Pentagon-512 (memory
FK> is really 512K). :) Although he writes that "...the general memory field
FK> _with_inhibition_...". Pentagons are wait-free, unless, of course, they are turbocharged. I have "Leningrad" aka Scorp.

FK> Reason: I just have a quartz resonator at 14.5 MHz, and not
FK> 14.0 MHz exactly, as in Scorps, i.e. The car is slightly turbocharged, as
FK> shows in the performance test. It should be recorded. ;)
Have you tested your car? ;-)

Best regards, Musa Khaybullin.

EP>> TR-DOS is clogged

FK> This is bad. Loaders with IM2 (music loaders) will fail. :(
FK> Connect the /M1
FK> signal to the TR-DOS input/output device decoder. Most likely, the feces are read from port #FF (6&7 bits)
FK> TR-DOS.
That's right! In the Pentagon there is nothing else to detect this! This is the only way!

Yours sincerely, Musa Khaybullin.

EP>> The data bus of this Spectrum in IM2 mode in TR-DOS is clogged

KF> A problem for all pentagons. There is a Beta-Disk connected through a buffer,
KF> and they forgot to put resistors in it.
There is no buffer in the PENTAGON!!! Tama just needs to be set up to
decipher the M1 controller.

EP>> And one more thing: I have a capacitor C6 (Pentagon circuit) with a capacity of 0.83
EP>> uF (according to the circuit - 22n), if you set less, the keyboard (film
EP>> extended) freezes, and even now some signs are slowing down.
EP>> I tried the int circuit with a trigger (ZX-Revue 95), I didn’t notice any changes
EP>>. I can’t understand “where the dog is buried”? :(

KF> IMHO already ruki.sys
There should be a 1KΩ resistor and a capacitance of 10 to 15 nf in the circuit. With these
parameters you can achieve an INT duration of 32 cycles. And for the film keyboard to work normally, it is advisable to buffer it, i.e. install buffers with an open collector in
the keyboard. They need to be installed exactly in
KEYBOARD, otherwise there will be no sense ;(

EP>> Yes, about the test - I would like to know your opinion on what you read,
EP>> constructive suggestions.
Write with soap, I’ll tell you some smart improvements ;)

Yours sincerely, MusaKhaybullin.

wall...@mail.ru

FK>> Connect the processor signal /M1
FK>> to the TR-DOS input/output device decoder. Most likely the feces are read from port #FF (6&7 bits)
FK>> TR-DOS.

AL> PLZ, throw it in the echo. It will be useful to many.

It's gone!!!

There is a built-in help inside, you just need to press the letter H.

Kirill Frolov. [ZX]

KF>> connected, but they forgot to put resistors in it.
MK> There is no buffer in the PENTAGON!!! Tam just needs to set
MK> to decrypt the M1 controller.

Yes. I just never saw the diagram. There is no need for M1 there. ID7 must be torn off and thrown away.

MK> In order for the film keyboard to work normally, it is advisable
MK> to sample it, i.e. install buffers in the keyboard with an open
MK> collector. They need to be placed exactly in the KEYBOARD, otherwise there will be no
MK> of any use ;(

Yes? And what, does the resistance of the wires have such an effect?
Yes, it turns out that all scorpions and kayas do not work and zones and dumb fools?

Kirill Frolov. [ZX]

AI> M1 online.

See my letter to Musa. M1 is possible, but not necessary.

Only this was probably in ZX.SPECTRUM. :-/

All the problems are purely in the decoding logic, you can’t do this.
It is not necessary to include RD or WR on the data inputs of the decoder,
because there is IORQ on the CS of the decoder, and it will be
in different phases with RD or WR.

Kirill Frolov. [ZX]

MK> Is your car tested? ;-)

Test, of course! ;) Have fun: Pentagon and... a general memory field with inhibition. :)
Imho, in this case (like mine) you should
write: “Computer: ZX-Spectrum 512K.”

FK>> somewhere
FK>> 24-28-32.
FK>> 28 is ideal.

MK> In our test, the ideal number of cycles is 32!

What does “in our test” have to do with it? :) There are a bunch of INT measurements available. On my
computer it always (almost depending on the temperature in the room :) shows
the average (optimal) 28 clock cycles. Kovalevsky explained quite clearly why this particular number was chosen. 32 is also normal...

MK>> open collector. They need to be placed exactly in the
MK>> KEYBOARD, otherwise there will be little use ;(

KF> Yes? And what, does the resistance of the wires influence this?
Not the resistance of the wires, but the resistance of the film keyboard! On
the film keyboard it is impossible(!) to achieve full contact, i.e. R=0 Om, there
it turns out from 100 to 300 Ohm maximum. If you put a buffer in the keyboard, then the buffer field will have a normal contact closure and the characters will not disappear ;)
KF> Yes, it turns out that all the scorpions and zones
KF> are not working and they are dumb?
How many scorpions have I come across - they all had a reed keyboard! And I have never seen Kaev in my life ;)

Long life and prosperity to you, Kirill!

With best wishes, Musa Khaybullin.

Saturday December 04 1999 05:14, Kirill Frolov wrote to Musa Khaybullin:

KF> Once again - _HA_ _PENTAGONE_! There must be a long int so that
KF> all the demos work normally.
Where did you get this from? After all, not all Pentagons initially have a long int. And
the long one needs to be fixed, definitely.

Yuri

KF>> so that all demos work normally.
YV> Where did you get this from? After all, not all pentagons initially have a long
YV> int. And the long one needs to be fixed, definitely.
EXACTLY! Otherwise, recapture often happens ;)

With respect, Musa Khaybullin.

MK> Not the resistance of the wires, but the resistance of the film
MK> keyboard! On a film keyboard it is impossible(!) to achieve full contact,
MK> i.e. R=0 Om, there it turns out from 100 to 300 Ohms maximum.

Now explain - what difference does it make where to put the buffer, in the keyboard or on the computer?

MK> If you put the buffer in the keyboard, then the buffer field will be normal
MK> the contacts will be closed and the characters will not disappear ;)

In scorps and kayas you know where it is.

And you can’t live without a buffer.

MK>> reed clave! And I have never seen Kaev in my life ;)

KF> I have a film one. In the ZX-48 I had to install a buffer, in the
KF> computer.

KF> P.S.: And if you install a buffer in the keyboard, then you will have a computer address bus going through the
KF> keyboard wire (1.5 meters).
^^^^^^^^^^ 3 Meters ;)
I actually spoke about amplifying buffer elements with an open
collector. At the Pentagon, which you and I are thinking about, there is already an AP5 through which the addresses pass. And you’re telling me about your Leningrad 48, in which
there are only DIODES AT ALL!

MK> When you put a buffer in the keyboard, and your keyboard is located
MK> from the computer at a distance of 3 meters from the connector, then THIS is of great importance! For example, the signal voltage drops ;)

Yep. Wet rope connection. How is the address bus of a computer with 3
meter
wires doing?

MK> I actually talked about amplifying buffer elements with an open
MK> collector. The Pentagon, which we are talking about, is already standing

The resistance of 3 meters of the thinnest MGTF is not that great.
There is no need for fairy tales about the skin effect. And if something falls there, then only
from interference in the wires themselves - reduce the resistance in the receiving
multiplexers.

Kirill Frolov. [ZX]