AY?

The question arose from the AY theory:
In registers 8,9,A the amplitude of the signals for channels A,B,C is stored.
The fourth bit of these registers is responsible for turning on the envelope.
Question: what happens if you simultaneously turn on the envelope bit and
set the volume of the channel? For example, the value is 1D. A specific example is the 'Binary Love' demo during the effect with a flashlight illuminating the
line. I don’t know how it should sound correctly, there is no live spec.
But my emulsion plays completely differently than Shalaeva. :(

PS: In the emulator the sound has been converted to DMA, now it has a frequency of 44Khz and
does not slow down! :)

Sincerely
┬ ┬┌┐┬┐ ┐┬┬
╨╜╨╙╜╨╙ ╨ ╨

don’t delay - throw it :)))) (by the way, I couldn’t duplicate it on ran...@cybemail.net
please?)

Regards,
Random
--
Lord of Chaos.
e-mail: ran...@cybermail.net

Concern Chaos WWW HQ: http://www.cybermail.net/~random/

Not so long ago, Vladimir Yudin did not consider it difficult to write All
(I even remembered the date and time: 19 Nov 97 at 21:10):

VY> Question: what will happen if you simultaneously turn on the envelope bit and
VY> set the channel volume?

I have no idea.

I wish you happiness and... good luck.
Andrey.

VY> set channel volume? For example, the value is 1D. A specific example -
VY> demo of 'Binary Love' during the effect with a flashlight illuminating the
VY> line.

Flashlight? Line? Original =) Is this at the beginning? Presents?

The sound there is similar to tone dialing of a telephone number =)
Realizable, IMHO, only in FT =)

VY> I don’t know how it should sound correctly, there is no live spec.
VY> But my emulsion plays completely differently than Shalaeva. :(

Hehe...

VY> PS: In the emulator, the sound has been converted to DMA, now it has a frequency of 44Khz
VY> and
VY> does not slow down! :)

Do it like in zx-jam - normal scans for “non-performing” video cards...

cYa, aRTY/dR! · · ··· Email: Ar...@telecom.nov.ru · · ···

VY> The question arose based on the theory of AY:
VY> In registers 8,9,A the amplitude of the signals for channels A, B, C is stored.
VY> The fourth bit of these registers is responsible for turning on the envelope.
VY> Question: what will happen if you simultaneously turn on the envelope bit and
VY> set the volume of the channel? For example, the value is 1D. A specific example -
VY> 'Binary Love' demo during the effect with a flashlight illuminating

It’s quiet outside the window, and I’m sitting and writing a response to a letter written from Vladimir Yudin to
All Wed Nov 19, 1997 at 21:10...

VY> PS: In the emulator, the sound has been converted to DMA, now it has a frequency of 44Khz and
VY> does not slow down! :)

What about the emulator itself? On weak fours? (I’m already silent about 386).

With all sorts of regards and wishes.
Igor.

... If a person had wings, they would greatly interfere with his crawling.

AT> I have no idea.

Thank you very much for your precise and informative answer!
What a pleasure it is to communicate with a _smart_ person.
Best regards
┬ ┬┌┐┬┐ ┐┬┬
╨╜╨╙╜╨╙ ╨ ╨

20 Nov 97 23:00, Artem Haritonov wrote to Vladimir Yudin:
VY>> 1D. A specific example is the 'Binary Love' demo during the
VY>> effect with a flashlight illuminating the line.

AH> Flashlight? Line? Original =) Is this at the beginning? Presents?

AH> There the sound is similar to tone dialing of a telephone number =)
AH> Realizable, IMHO, only in FT =)

That’s not what it sounds like, I’m talking about something else, after this effect comes
the authors’ presentation, and after it the line on half the screen
moves left and right with real-time lighting. Is this what
music should sound like in this place? And why is the code 1Dh sent to the volume registers, for example?
Sincerely
┬ ┬┌┐┬┐ ┐┬┬
╨╜╨╙╜╨╙ ╨ ╨

VY>> frequency 44Khz and does not slow down! :)

IG> What about the emulator itself? On weak fours? (I’m already silent about 386).

Alas, so far with real speed only on the Pentium and then with a fast video card.
After all, the screen (with border) is redrawn 50 times per second.
It is planned to introduce a mode to disable full border emulation
(only changing the palette without synchronization), then in theory the speed should
more than double (screen size 49Kb, border size 68Kb).

Sincerely
┬ ┬┌┐┬┐ ┐┬┬
╨╜╨╙╜╨╙ ╨ ╨

VY> Alas, so far with real speed only on a Pentium and then with a fast
VY> video. After all, the screen (with border) is redrawn 50 times per second.

Or maybe it’s all about the command emulation block? According to my calculations, the emulsion should
work freely on the 100th quad, if you try, then you can do it on 66.

After all, it turns out to be simple arithmetic: in 1 clock cycle the z80 will have time (at 66 MHz) to complete
~18 clock cycles. When executing a command in two steps, everything turns out very beautiful. The four-cycle z80 command has 70 Intel cycles. Lovely!

VY> It is planned to introduce a mode to disable full border emulation
VY> (only changing the palette without synchronization), then in theory the speed should
VY> more than double (screen size 49Kb, border size
VY> 68Kb).

Emulating a curb is a couple of times easier, IMHO. Well, it depends on how you did it.

With best wishes, John.

===> #/FUCK/# *mUlTiMeDiA*, #/MAKE/# more $dEmOs$ ===>

VY> Alas, so far with real speed only on the Pentium and then with a fast video card.
VY> After all, the screen (with border) is redrawn 50 times per second.
VY> It is planned to introduce a mode for disabling full border emulation
VY> (only changing the palette without synchronization), then in theory the speed should
VY> more than double (screen size 49Kb, border size 68Kb).

Study Dos Extender. ;)

Sunday November 23 1997 21:04, Vladimir Yudin wrote to Artem Haritonov:

VY> It doesn’t sound like that, I’m talking about something else, after this effect comes
VY> the authors’ presentation, and after it the line on half the screen
VY> moves left and right with real time lighting.

Damn =) This is called bump mapping =)))

VY> This is how music should sound in this place?

Some kind of acid... Find a real spec, is it really difficult? How do you even make an emulsion without a real machine?

VY> And why is the code 1Dh sent to the volume registers, for example?

This is what Fast Tracker does... What?

cYa, aARTY/dR! · · ··· Email: Ar...@telecom.nov.ru · · ···

And nothing will happen - as the envelope sounded and will sound - it
has no volume

With respect, Roman aka Megus of Brain Wave corp.

[Team Agatha Christie] [Team Metallica] [Team ZX-Spectrum] [Team Iron Maiden]

VY> Alas, so far with real speed only on the Pentium and then with a fast video card.
VY> After all, the screen (with border) is redrawn 50 times per second.
VY> It is planned to introduce a mode for disabling full border emulation
VY> (only changing the palette without synchronization), then in theory the speed should
VY> more than double (screen size 49Kb, border size 68Kb).

VY>> fast video. After all, the screen (with border) is redrawn 50
VY>> times per second.

JS> Maybe it’s all about the command emulation block? According to my calculations,
JS> the emulsion should work freely on the hundredth four, if
JS> try hard, then you can get it on 66.

Emulation is not a problem, it takes little time.
I’m talking about sending video information.
Count for yourself: there are 50 frames per second, a frame is 115Kb, a total of every second
5.5Mb of video memory is sent. Bandwidth
of my ET6000 110Mb, S3Trio64V+ 40Mb, ATI 10MB, ISA Cards 1-4Mb.
Thus, on slow cards, screen output eats up more
half of the computer time, and on slow ones it clearly doesn’t keep up.

JS> After all, it turns out to be simple arithmetic: in 1 clock cycle the z80 will have time (at
JS> 66 MHz) to complete ~18 clock cycles. When executing a command in two JS cycles, everything turns out very beautiful. Ha four-cycle
JS> command z80 70 clock cycles. Lovely!
Yes, lovely, but if you don’t need to display it on the screen in real
time.

JS> It’s a couple of times easier to emulate a border, IMHO. Well, it depends on how you
JS> is done.
Well, it’s yours IMHO, it’s one thing to switch the palette, and another to track
the time of each switch, and then build a border using this table.

Best regards
┬ ┬┌┐┬┐ ┐┬┬
╨ґ╨╙ґ╨╙ ╨ ╨

24 Nov 97 21:25, Igor Gruzdev wrote to Vladimir Yudin:

IG> Study the dos extender. ;)

Yes, it’s time.
Sincerely
┬ ┬┌┐┬┐ ┐┬┬
╨ґ╨╙ґ╨╙ ╨ ╨

AH> Some kind of acid... Find a real speck, is it really difficult? How do you
AH> even make an emulsion without a real machine?
Yes, obviously some kind of acid. And I thought there was something wrong with the emulation! :)

Yes, I still want to buy a real spec, but for now everything is from memory! :)

Sincerely
┬ ┬┌┐┬┐ ┐┬┬
╨╜╨╙╜╨╙ ╨ ╨

VY> Well, this is yours IMHO, it’s one thing to switch the palette, and another to track
VY> the time of each switch, and then build a border using this table.

What’s so complicated about that? Tact counter. When writing to a port, a counter is written along with the color. Nothing complicated. It is much more difficult to emulate the main
screen along with attributes in a chunk. Or is it not?

VY>> track the time of each switch, and then use this
VY>> table to build a curb.

JS> What's so complicated about that? Tact counter. When writing to a port, a counter is written along with the
JS> color. Nothing complicated. It is much more difficult to emulate the main screen along with attributes in a
JS> chunk. Or is it not so?

It’s difficult to do here to make it all work quickly. :)
And the screen is of course a little more complicated, but also solvable.

Sincerely
┬ ┬┌┐┬┐ ┐┬┬
╨╜╨╙╜╨╙ ╨ ╨

By the way, even in STS EDEDEDEDEDEDED00 will be one command, and until the percentage
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
reaches the last byte, it will not go to any interruption.

C Best regards, Alex from FFC COMPUTERS.

AL> By the way, even in STS EDEDEDEDEDED00 will be one command, and so far the percentage
AL> not
AL> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
AL> will reach the last byte, it will not go to any interruption.

Maybe our STS are different, but for me EDED is NOP *. The EDED pack... is a pack of
NOP *.

In addition, in all documentation, including those that describe all sorts of
undocumented features of the z80, it is written that EDnn, where nn is in the range
#00-#3F #80-#FF, does nothing (except, of course, the command block
operations). Think about whether you are confusing ED with DD/FD.

JS> Maybe our sts are different, but for me EDED is NOP *. A pack of EDED... is a
JS> pack of NOPs *.

For the r register this is something like a pack of nops, and to leave for an interrupt, one
command, imho.

JS> Moreover, in all the documentation, including those where all sorts of
JS> undocumented features of the z80 are described, it is written that EDnn, where nn in
JS> ranges #00-#3F #80-#FF, does nothing (except, of course,
JS> block operation commands). Think about whether you are confusing ED with DD/FD.

I’ll try to check more precisely.