Jump to content
IGNORED

Trying to make some PoKEY music!


VinsCool

Recommended Posts

4 hours ago, emkay said:

Sounds all interesting. 

 

Table Manuscrite sound very special, using the extreme distorted guitar sound from time to time. Even the reverb is working nicely. 

 

But, you are sure about the uridium sketch?

 

 

 

Thanks, these distorted sounds could have been more stable tbh... I probably have reached the limit of my own patience trying to get the tuning work anymore too, I'll just give 1 final check today then move on to something else I wanted to experiment :D 

Uridium I just liked the way the notes sounded together with my tuning change, but that's about it, it was not my own tune after all.

  • Like 1
Link to comment
Share on other sites

Finally got some very good results now.
This is possibly the best I could get the Distortion A tuning for now.
There are some things that still aren't that great, mainly sharps being a bit off, but it's not as bad as my earliest attempts.

I also made some interesting observations with that alternate tuning:
- The overall sound is more stable in certain conditions, such as high pass filter special sounds created with certain semitones interval, 7 semitones apart does some really nice power chord tones, for example.
- Chords now shouldn't sound as dissonant as they used to be, since the intervals are a bit more constant up to the 7th octave, most things shouldn't sound out of tune as much as a result, up to certain limits, that is.
- As the frequencies go higher in tone, it gets closer and closer to the "original" RMT table, which was based on some 1983 documentation specific for NTSC tuning, which was probably what made it sound so weird in PAL and 15khz mode. However, now it's actually purposefully scaled to these frequencies, so things are more "in-tune" between the lowest and highest octave. Not quite perfect but I think it sounds much better that way.
- Harmonies are much better with distortion C(E) basses, there is some really good resonance on many notes, and I noticed their "frequencies" are also very similar now... This cannot be a coincidence... and I am too curious to not check this out sometime later ?
- Running tunes in PAL with my new table makes about everything sound considerably better, in fact now the notes are incredibly close to a really good 440hz based tuning, which was actually not expected. Same behaviour as described above.
- Running that table in 15khz seems to have drastically improved the tuning as well, for the reasons mentioned above. I suspect 1.79mhz mode will also benefit? I need to confirm.
- My new table is now incredibly close to synthpopalooza's Distortion A 15khz table, even though it was actually intended to 64khz, based on a 443hz tuning, clearly there is a connection somewhere...
- Note B-2 is actually possible to get with the frequency FF... which is actually not officially supported by RMT for some obscure reason.
- Perfect octaves come from perfect divisions in the frequency value. for example, B-2 uses FF, B-3 uses 7F, and that makes a perfect 2:1 resonance. Same observation for every notes. Unfortunately, this logic didn't translate perfectly downward between notes, however, I was able to get as many octaves as I could to be "perfectly tuned" together. As long as a "good" tuning frequency exists in the higher range, dividing downward will always produce a perfect 2:1 resonance, but the opposite isn't always true. So I picked up as many notes as I could that were nicely tuned and made them go downward.
- Fun fact, I noticed that 00, the highest possible frequency, is actually a B-10 note, perfectly in-tune to FF, and no, I am not making this up, do the math! ? 

Also just for fun I tried out the "Merge" feature of RMT2LZSS... Works pretty nicely for making medleys!
I just did a sloppy conversion of my Raymaze 2000 tunes as a test, also using this tuning theory.
 


I still am testing out tunes for now, but so far I got some pretty conclusive results.
This makes me pretty happy... I was starting to believe I wasted my time for the last few days, but now I think it turned into my favour :D 

So now I'm just about to mass convert my .rmt files for like the 3rd time this evening, and I'll make sure to report back the results, hopefully I won't encounter yet another serious problem on a random tune... :P 

POKEY Table 64khz (Octave Pattern) Final.txt Raymaze 2000 Themes (Merged).xex

  • Like 8
Link to comment
Share on other sites

Okay, I got a little carried over and now it's really late lol

Pretty much everything in this folder sounds great now!
There's surely some room for improvements but it's a lot better than my last attempt.

I hope I am not becoming tone deaf with all that time spent listening to the tunes tonight ?

tunes tests 3.zip

  • Like 1
  • Thanks 1
Link to comment
Share on other sites

On 2/28/2021 at 12:22 AM, VinsCool said:

 


 dta frqtabpure-frqtab,$00
 dta frqtabpure-frqtab,$20
 dta frqtabpure-frqtab,$40
 dta frqtabbass1-frqtab,$c0       
 dta frqtabpure-frqtab,$80
 dta frqtabpure-frqtab,$a0    
 dta frqtabbass1-frqtab,$c0
 dta frqtabbass2-frqtab,$c0

 

Looking at this table, something was missed with replacing the distortion settings and set to the note table. Only one bit pattern is needed for "Volume Only, Wave File PlayBack, or Digital Sounds" like $1v. V is the column level. Can $30, $50, $70, be used indicate different pokey/table settings? Like make $F0, be the 16bit note table merger 2 channels together. I need to look it up, can the tables be larger than 64 bytes and use 2 tables for the upper and lower bytes and get 8 octaves.


I acquired these note tables awhile ago for possible solutions for 88 notes, 8 octaves. 

 

Quote

;         C  C#   D   D#  E   F   F#  G  G#  A    A#  B
MUSIC_NOTE_AMS_TONE_TABLE: ;Pure/Buzz Tone  from AMS  (NTSC 64 khz)
    .byte $8E,$87,$82,$79,$73,$6C,$66,$61,$5B,$57,$F5,$E9 ;O1  00
    .byte $C5,$B9,$AD,$A7,$9B,$92,$89,$83,$7A,$74,$6E,$FF ;O2  12
    .byte $F3,$E6,$D9,$CC,$C1,$B6,$AD,$A2,$99,$90,$88,$80 ;O3  24
    .byte $79,$72,$6C,$66,$60,$5B,$55,$51,$4C,$48,$44,$40 ;O4  36
    .byte $3C,$39,$35,$32,$2F,$2D,$2A,$28,$25,$23,$21,$1F ;O5  48
    .byte $1E,$1C,$1A,$19,$17,$16,$15,$13,$12,$11,$10,$0F ;O6  60
    .byte $0E,$0D,$0C,$0C,$0B,$0A,$0A,$09,$09,$08,$08,$07 ;O7  72
    .byte $07,$06,$06,$05,$05,$04,$04,$04,$04,$03,$03,$02 ;O8  84
MUSIC_NOTE_AMS_DIST_TABLE: ;Distortion      from AMS  (NTSC 64 khz)
    .byte $C0,$C0,$C0,$C0,$C0,$C0,$C0,$C0,$C0,$C0,$C0,$C0
    .byte $C0,$C0,$C0,$C0,$C0,$C0,$C0,$C0,$C0,$C0,$C0,$A0
    .byte $A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0
    .byte $A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0
    .byte $A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0
    .byte $A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0
    .byte $A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0
    .byte $A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0,$A0
;         C  C#   D   D#  E   F   F#  G  G#  A    A#  B           
MUSIC_NOTE_16BIT_HIGH:
   .byte $d5,$c9,$be,$b3,$a9,$a0,$97,$8e,$86,$7f,$77,$71 ;O0  00
   .byte $6a,$64,$5f,$59,$54,$50,$4b,$47,$43,$3f,$3b,$38 ;O1  00
   .byte $35,$32,$2f,$2c,$2a,$28,$25,$23,$21,$1f,$1d,$1c ;O2  12
   .byte $1a,$19,$17,$16,$15,$13,$12,$11,$10,$0f,$0e,$0e ;O3  24
   .byte $0d,$0c,$0b,$0b,$0a,$09,$09,$08,$08,$07,$07,$07 ;O4  36
   .byte $06,$06,$05,$05,$05,$04,$04,$04,$04,$03,$03,$03 ;O5  48
   .byte $03,$03,$02,$02,$02,$02,$02,$02,$02,$01,$01,$01 ;O6  60
   .byte $01,$01,$01,$01,$01,$01,$01,$01,$01,$00,$00,$00 ;O7  72
   .byte $00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00,$00 ;O8  84
;         C  C#   D   D#  E   F   F#  G  G#  A    A#  B   
MUSIC_NOTE_16BIT_LOW:
   .byte $c1,$c1,$6e,$bd,$a6,$21,$23,$a7,$a5,$16,$f4,$38 ;O0  00
   .byte $dd,$dd,$33,$db,$d0,$0d,$8e,$50,$4f,$88,$f6,$99 ;O1  00
   .byte $6b,$6b,$96,$ea,$64,$03,$c4,$a5,$a4,$c0,$f8,$49 ;O2  12
   .byte $b2,$32,$c8,$72,$2f,$fe,$de,$cf,$cf,$dd,$f8,$21 ;O3  24
   .byte $55,$96,$e0,$35,$94,$fb,$6c,$e4,$64,$eb,$79,$0d ;O4  36
   .byte $a7,$47,$ed,$97,$46,$fa,$b2,$6e,$2e,$f2,$b9,$83 ;O5  48
   .byte $50,$20,$f3,$c8,$a0,$7a,$56,$34,$14,$f5,$d9,$be ;O6  60
   .byte $a5,$8d,$76,$61,$4c,$39,$27,$16,$06,$f7,$e9,$db ;O7  72
   .byte $cf,$c3,$b7,$ad,$a3,$99,$90,$88,$80,$78,$71,$6a ;O8  84


 

  • Thanks 1
Link to comment
Share on other sites

4 hours ago, NeoNZJ_Slayer said:

Looking at this table, something was missed with replacing the distortion settings and set to the note table. Only one bit pattern is needed for "Volume Only, Wave File PlayBack, or Digital Sounds" like $1v. V is the column level. Can $30, $50, $70, be used indicate different pokey/table settings? Like make $F0, be the 16bit note table merger 2 channels together. I need to look it up, can the tables be larger than 64 bytes and use 2 tables for the upper and lower bytes and get 8 octaves.


I acquired these note tables awhile ago for possible solutions for 88 notes, 8 octaves. 

 


 

That would certainly make a lot of sense that way I believe.
When I was making own observation regarding the notes mapping, it surprised me how so many things could have been done differently, just for having more settings, and cases for different table use.

I especially really like the mapping of your table above, being able to pick up what distortion to use on specific notes would make a lot of things faster since it would remove the necessity to add an additional instrument, however I don't know how well things like vibrato and portamento would be handled from that point, I suppose a check to not have these effect used on these distortions would make sense.
From what I could tell 16 bit was already done in a way similar to yours, having tables for high and low bytes, but this is pretty much absolutely necessary in the first place, since doing notes manually is pretty slow directly in the music tracker.

As far as I know, things were done in the way they are currently done in order to save as much memory as possible, so I can understand the reasoning behind it.
I am not a programmer however, so my own observation as an outsider may be absolutely wrong. hahaha :D 

Link to comment
Share on other sites

On 3/5/2021 at 7:23 AM, emkay said:

Back then, I proposed to have the "clean" Note and Pattern editor , as RMT is now. 

The instruments might need a "synthesizing tool" , to develop the resultings sounds. 

 

But it all needs the possibility of doing pitch adjustments, when the tune is running at the patterns.

It can be used to avoid cancelling, or to release from some "deaf clamped tuning" that happens due to physical restrictions. 

 

On 3/5/2021 at 7:53 AM, VinsCool said:

This, especially, is driving me crazy at the exact moment, lol ?

 

Time to start a new thread regarding a new tracker ?

  • Haha 1
Link to comment
Share on other sites

13 hours ago, VinsCool said:

Okay, I got a little carried over and now it's really late lol

Pretty much everything in this folder sounds great now!
There's surely some room for improvements but it's a lot better than my last attempt.

All that music theory is flying way over my head ? 

 

I tried those latest tunes and while they sound good I didn't notice any difference (but I'm not very familiar with them)

 

So I checked aurora and SAPRed both the original and your version, and while there lots of differences, all the values vary by a single unit, surely that can't have a big impact ?

 

I'm curious if you may have been wandering for too long and gone back to your starting point ? ?

 

Link to comment
Share on other sites

9 hours ago, NeoNZJ_Slayer said:

I am sure there are different ways to encode and compact the music data. There are many good programmers out there that are experience with bitwise operations and binary encoding files. Make use of unused bits. 
 

That would be another patch and a pretty complex one too... modifying the RMT code (beyond replacing tables) isn't a good idea for all the reasons mentioned previously.

 

Don't want to prevent anybody from trying of course, I just think it'd be a wasted effort.

Link to comment
Share on other sites

1 hour ago, rensoup said:

I tried those latest tunes and while they sound good I didn't notice any difference (but I'm not very familiar with them)

 

I prefer this over making something sound worse! lol
The idea I had was to make things sound as good as they could be, if I am even able to, that is.
Technically the higher octaves have barely any difference, it's what lies below that was mostly changed, in order to make everything a lot more consistent.

 

1 hour ago, rensoup said:

So I checked aurora and SAPRed both the original and your version, and while there lots of differences, all the values vary by a single unit, surely that can't have a big impact ?

Well the best way to know is to listen to each version side by side, that really was all I did to know if what I tried to do helped or broke something.

1 tiny change can make a big difference, especially in the higher tones, so yeah, hopefully it really did help a bit overall :P 
There is also giai_phon_mien_nam, or computer_world, or castle_von_krumpen_arp_fix, or hopefully most of the tunes that made use of the Distortion A the most, that might see major improvements, or fail miserably if my idea was shit, lol. 

Hopefully I was able to improve the tuning in a good way.

For the ones I personally have listened several times with comparison in mind, many notes seem to work better together now, but some others are a bit debatable, mainly sharps, and E and G that feel slightly off in some situations, so it's a lot of trial error, so far the combination I have is currently the nicest of the bunch, but it's definitely possible to improve it, there are 256 tones that can be chosen after all!
 

I also noticed a few combinations that also made many of the Distortion basses tones work much better in the mix, and I noticed a certain pattern that I wanted to experiment a bit later tonight or maybe another day lol
If my theory turns out to be actually true, that means there may another potential improvement using the same approach, this time in the Distortion C ? 

I mean, what I had in mind here wasn't to make the perfect tuning, but to making as many things as possible sound in-tune together, which ultimately makes the whole thing seem correct, instead of having many notes that are sounding wrong, or out of tune.
I like to believe if I make everything equally out of tune, the resulting sound might be in-tune! :D

 

1 hour ago, rensoup said:

I'm curious if you may have been wandering for too long and gone back to your starting point ? ?

 

I honestly felt that way the other day, and I really was ready to give up since I felt like I was wasting my time on some stupid ideas, and would eventually make a fool out of myself, lol
But now I believe I am finally starting to understand, and so far got to a point I wouldn't have expected, so maybe it was actually worth it? :D 

At worst, it would be experience and no one is harmed, or at best, I would be able to contribute something useful in the Atari 8-bit scene.

  • Like 3
Link to comment
Share on other sites

2 hours ago, VinsCool said:

For the ones I personally have listened several times with comparison in mind, many notes seem to work better together now, but some others are a bit debatable, mainly sharps, and E and G that feel slightly off in some situations, so it's a lot of trial error, so far the combination I have is currently the nicest of the bunch, but it's definitely possible to improve it, there are 256 tones that can be chosen after all!

 

Well, yes, but of those 256:

  • 128 are in the bottom octave. You don't have to be particularly clever to make this sound good.
  • 64 are in the second octave. The conventional table already goes awry here.
  • 32 are in the third octave. It is tricky to get this right.
  • 16 are in the fourth octave. Some notes are going to be seriously out of tune no matter what you do.
  • 8 are in the fifth octave. That's not even enough distinct values to make a full 12-tone scale, and some of the ratios that are available are not close to any of the desired intervals.

Your latest table is very interesting!

  • Thanks 1
Link to comment
Share on other sites

On 2/2/2021 at 8:20 AM, emkay said:

Hey Vins

 

How about something completely different. 

Have a listen to this one:

 

 

 

 

It's sometimes unbelievable , what this old chip can do, using all available channels. 

You know, I'm not good ( ;) at  ) creating arrangements. 

Here is a proposal for a 4 channel POKEY version. 

64kHz

channel 1 and 2 playing the lead, but channel 2 is playing one octave higher, while channel one keeps the lowest possible octave. 

Channel 3 and 4 used for bass, fx, and drums. 

Some nifty AUDCTL usage given ;)

 

https://streamable.com/ak8rf0

 

50 KHz, no filters, some wierd arps.

IRQbass pocedure by @R0ger aplied.

 

No fancy effects nor drumline here... yet.

 

  • Like 1
  • Thanks 1
Link to comment
Share on other sites

Been playing around a bit more tonight, and now applied the same idea into the bass tables.
I have to say, this is some incredibly good results, compared to how these tunes originally sounded.
Some of these tunes got considerable improvements with these changes, some things may also have been broken, there's many things that could go wrong too.
Currently, even when there is a bit of dissonance somewhere, it feels a lot more fitting, if that makes any sense, lol. I suppose anything could happen.

I also really liked how my tune, still unfinished, sounded like with these tuning changes, so here's a video :D 


I also attached a bunch of executables I had listened at random earlier, I hope you guys enjoy how things sound like!
Credits go to their own authors when they aren't my own tunes, of course!

tunes tests 4.zip POKEY Table 64khz (Octave Pattern, C) v10.txt

  • Like 7
Link to comment
Share on other sites

4 hours ago, VinsCool said:

Been playing around a bit more tonight, and now applied the same idea into the bass tables.
I have to say, this is some incredibly good results, compared to how these tunes originally sounded.
Some of these tunes got considerable improvements with these changes, some things may also have been broken, there's many things that could go wrong too.
Currently, even when there is a bit of dissonance somewhere, it feels a lot more fitting, if that makes any sense, lol. I suppose anything could happen.

I also really liked how my tune, still unfinished, sounded like with these tuning changes, so here's a video :D 


I also attached a bunch of executables I had listened at random earlier, I hope you guys enjoy how things sound like!
Credits go to their own authors when they aren't my own tunes, of course!

tunes tests 4.zip 162.54 kB · 2 downloads POKEY Table 64khz (Octave Pattern, C) v10.txt 11.42 kB · 1 download

now I like that more than the video before... why? this doesnt sound like typical POKEY tune... and I like that distorted bass sound...

  • Like 1
  • Thanks 1
Link to comment
Share on other sites

9 hours ago, Heaven/TQA said:

now I like that more than the video before... why? this doesnt sound like typical POKEY tune... and I like that distorted bass sound...

Thank you very much!

 

I believe there is still a lot of very cool sounds this chip can produce, and I do my best to understand how they work and also make use of them :)

 

 

  • Like 2
Link to comment
Share on other sites

By the way if anyone is interested, here's my own edited version of RMT, using the Alternate Tuning tables I posted in this thread.
Rmt Vin Tuning Patch V9.exe
I also edited RMT2LZSS the same way, it actually was how I was able to test most of my tunes en masse easily, otherwise it would have taken at least 3 eternity to even get something lol
It replaced the table for RMT 1.28, the other ones were not touched.
RMT2LZSS Vin Tuning Patch A and C and E V9.exe
I hope you are okay with that rensoup? ? If there's anything I shouldn't have posted I'll delete upon request.

There is 1 particular change in the Bass 1 table however, in order to get proper notes below G#1, it uses the "Distortion E" bass timbre for those notes. This is only a personal preference, but otherwise everything should be virtually compatible with tunes made in 1.28(1.30).
Hopefully I did not break as many things as I tried to improve a bit :D 

Edited by VinsCool
Fixed a dumb hex editing mistake, oops
  • Like 4
Link to comment
Share on other sites

9 hours ago, VinsCool said:

Thank you very much!

 

I believe there is still a lot of very cool sounds this chip can produce, and I do my best to understand how they work and also make use of them :)

 

 

Not just the sounds. 

The better the control of pokey gets, the sharper the correct note can get .

 

 

Edited by emkay
  • Like 1
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...