1st Alpha Version of Theta Music Composer 2.0

[Heaven/TQA]

i just found the 1st alpha version of TMC 2.0 on jaskiers site which is now available in english as well:

 

http://jaskier.atari8.info/index_en.html

 

i have attachted the 1st build which you should load with a DOS...unfortunatly it has no example music file included (as jaskier is not an musician) so we have to wait for raster, emkay, sack, xray, samurai etc... to write some nice tunes with it...

 

if you have questions then please adress them directly to jaskier who is popping to this boards as well from time to time...

tmc20a.zip

[Jaskier]

Hi Heaven!

 

You are late The beta version is coming:

tmc20b.zip

[raster/c.p.u.]

Hi Heaven!

You are late  The beta version is coming:

Good work, Jaskier!

I have some few questions only:

1.) command 5 (P+filter -> filter) doesn't work yet, am I right? (I tried to use it in loop, but it has no effect.)

2.) command 6 (P -> filter) ... I think it's redundant (IMHO command 5 is fully sufficient). Don't you think?

3.) will be possible to show "Instruction" by Help k.... OH?! HEY! It works already! Oh Yes, it seems to me be intuitive behaviour.

[Heaven/TQA]

jaskier/raster

 

can you please make a comparison chart what is the difference between RMT & TMC 2.0??? so it would be easier to understand which tracker is "better" in which area? f.e. can TMC play samples etc?

[Jaskier]

1.) command 5 (P+filter -> filter) doesn't work yet, am I right? (I tried to use it in loop, but it has no effect.)[

 

You have to manually turn on the filter. Write 6 to 4-th parameter in 1-st row.

 

2.) command 6 (P -> filter) ... I think it's redundant (IMHO command 5 is fully sufficient). Don't you think?

 

I don't know, but the actual accent code looks very funny:

; accents

a0     adc frq0

a1     sta frq0

a6     jmp grajreturn

a2     adc aktwys0

a3     sta aktwys0

a7     jmp grajreturn

a4     adc filter

a5     sta filter

      jmp grajreturn

 

3.) will be possible to show "Instruction" by Help k.... OH?! HEY! It works already! Oh Yes, it seems to me be intuitive behaviour.    

 

I've just noticed that when you press help the actual instrument/pattern is not saved to memory, so you loose your changes. Next bug

[raster/c.p.u.]

You have to manually turn on the filter. Write 6 to 4-th parameter in 1-st row.

Yes, I have set AUDCTL to 6 (filter is turned on), but command 5 doesn't P+filter->filter .... Aha! There is bug in help text. Command with number 4 is missing and command 6 is twice there.

Ok, I tried to use command 4 with parameter 01 in loop now, but it doesn't work too. Or I don't understand how it should works?!

 

I don't know, but the actual accent code looks very funny:

; accents

a0     adc frq0

a1     sta frq0

a6     jmp grajreturn

a2     adc aktwys0

a3     sta aktwys0

a7     jmp grajreturn

a4     adc filter

a5     sta filter

      jmp grajreturn

Very nice!!!

[analmux]

@ Jaskier:

 

Did you implement the asymmetric squarewaves, or does it still give too much trouble with the timer-IRQ and the DLIs?

 

I've read about another way to generate asymmetric squarewaves on the Polish forum (atariarea), but that requires two voices, but no interrupts, so it's very time-economic.

[Jaskier]

Yes, I have set AUDCTL to 6 (filter is turned on), but command 5 doesn't P+filter->filter .... Aha! There is bug in help text. Command with number 4 is missing and command 6 is twice there.    

Ok, I tried to use command 4 with parameter 01 in loop now, but it doesn't work too. Or I don't understand how it should works?!

 

I don't know what you expect, but for me it works. Try something like that in loop:

aaa

444

f00

c04

 

The sound is much different that without accents.

 

@Analmux:

TMC for now is only typical POKEY player. The waves are created by the POKEY generators without changing them. It is very possible that the asymetric waves or PWM invented by Saint will be implemented in the next versions of TMC but not now. My main goal for now is to create a demo and I'm very happy that I've made first step: I've created (ok, almost created ) the programm for my musicians to make music for my demo.

 

@Heaven:

About comparison chart: generally TMC and RMT are created for making sound the same way: intensive and very advanced use of standard POKEY generators. So the comparison will be like comparison MS Office and OpenOffice. First is under Windows, second has source code published

[raster/c.p.u.]

I don't know what you expect, but for me it works. Try something like that in loop:

aaa

444

f00

c04

The sound is much different that without accents.

For example, lets loop contains:

a

4

0

1

And I expect, then filter value will be ...,1,2,3,4,5,6,7,8,.... (+1 for each this command)

a

5

0

1

Now I expect the filter value will be .....1,1,1,1,1,1,.... (=1 for each this command)

(And BTW - I think command p+filter->filter is enough and there is no need of p->filter, because it is possible to do everything we need by command p+filter->filter)

[Jaskier]

a

4

0

1

And I expect, then filter value will be ...,1,2,3,4,5,6,7,8,.... (+1 for each this command)

 

No, the AUDF of filtered channel will be AUDF(primary)+1

 

a

5

0

1

Now I expect the filter value will be .....1,1,1,1,1,1,.... (=1 for each this command)

 

The same here. The difference is when you are using special types of sound like vibrato etc. These types can also modify filter, but with accent 5 you can forget about these changes and store exactly what you want. (Maybe it is redundant, but I've still 2 accents codes to spare )

[samurai]

i think, that demo of new possibilities should be done (some demo short songs) - as raster did in rmt ...something which make my try to do something on TMC.

 

For now RMT is enough for me ...as you will hear in Trencin ...or after downloading the stuff from Forever5

[raster/c.p.u.]

No, the AUDF of filtered channel will be AUDF(primary)+1

Of course, I thought that AUDF of filtered channel will be AUDF(primary)+1,+2,+3,+4 ... (i.e. difference between AUDF primary and AUDF filtered will be increased about +1 for each command in loop). If it will works in this way, it give much more variation potential of loops at all.

 

Similarly, command 0 should really increase current frequency (see the description: P+F->F !). So, if we use

A

0

0

1

in loop, then resultant Frequency should be increased about +1 for each step (+1, +2, +3, ...) , no constantly +1 forever as it does now.

[Jaskier]

I'll see what I can do. Grayscale team requested to make major changes in TMC so I can change some more.

[raster/c.p.u.]

I'll see what I can do. Grayscale team requested to make major changes in TMC so I can change some more.

I'm sorry, Jaskier, I'm not sure if I expressed clearly:

I expected that TMC2.0 command "0" (P+F->frq) does the same as TMC1.0 command "2" (P+A->AUDF).

So, If envelope looks like this:

Step         

AAAA

0000

0000

1111

and note C-1 will be used (i.e. frq 243), then there will be played frequencies: 244,245,246,247 and if last command (in column 4) will be in loop, then it will be performed in the same way (248,249,250,...).

 

And I have one important remark:

Command 1 (P->frq) should set AUDF to direct "P" value, but now it works like frq+P->AUDF and there is missing some (totally needfull) command for setting direct value to AUDF.

(this instrument feature is wholly important, in TMC1.0 it works of course and it is used very very often)

 

Oh... I'm confused... Maybe I don't understand how it works in TMC2.0 (what is "F", "frq" (source frq? destination frq? AUDF?) etc... ). I'm sorry...

[Jaskier]

I'm sorry, Jaskier, I'm not sure if I expressed clearly:

I expected that TMC2.0 command "0" (P+F->frq) does the same as TMC1.0 command "2" (P+A->AUDF).

 

I think that now it does. I'm sure that the increase/decrase of frequency were not permanent (I've checked it in code now). But I can add the accents for permanent shift of frequency and filter.

 

Oh... I'm confused... Maybe I don't understand how it works in TMC2.0 (what is "F", "frq" (source frq? destination frq? AUDF?) etc...   ). I'm sorry...

 

I think that I don't understand it either. Now I've ralized that during my "happy optimalization" I've changed the way how accents works. I have to change it back.

[Jaskier]

O.K. Next version ready to test.

 

Bugs removed:

-HELP key doesn't clear instrument and pattern data.

-keys for set/clear tag moved to SHIFT/CTRL K (conflict with volume setting)

-no sound at the start of playing the whole song

 

Bugs added:

-new accent engine (as sugested by Raster)

-new 16 bit sound engine (proposition by Emkay)

-file format changed, I'm preparing for some HUGE MODIFICATIONS proposed by Greg

 

more to come, beware!!!

tmc20c.zip