pcmplay - new dragon sound routine

[simon]

@stig

I WAS i c64 demo programmer back in the day.... now i'm a coco hacker...

cybernoid is way cool....

i think my favourite c64 track to date is green beret by martin galway...

/Simon

[Stig]

@stig

I WAS i c64 demo programmer back in the day.... now i'm a coco hacker...

cybernoid is way cool....

i think my favourite c64 track to date is green beret by martin galway...

/Simon

Great to see people hacking for other things besides the C64, it has enough people hacking it. The more people hacking for the coco / dragon the better !

Green Beret is another amazing tune, would love to hear that on the Dragon also, very haunting tune !

Thanks for commenting on my video also Simon

[sixxie]

You can lop the ORB #2 out of your loop if you set the appropriate bit in the DDR for that PIA as an input rather than an output. The A side of PIAs have an internal pull-up resistor, so should keep the output high, though I've no CoCo serial kit to test that with.

[simon]

@stig...

at the end of the day the c64 was an awesome machine.. (still is).... but the dragon / coco were very undersung...

hence the demo fury from my part.... (along with utz making awasome tunes for what ever strange stuff we cook)...

heck... no one thought we would ever cook a musical tapeloader ala ocean... but teamwork... my player, stewart orchards code... utz's xm converter....

/Simon

[sorchard]

using byte-length counters would speed up things a lot, but I find it limits the range of notes too much

What I was suggesting would still give you 16 bit counter resolution and less than 5 cents of pitch error in the region of A0. There is a catch though: I was wondering how you were getting the higher notes with a 24 bit counter and I now realise you have multiple cycles in your samples, not just a single cycle. To make the 16 bit counters work will require the tone samples to be reworked into a single cycle. Not a quick fix, but that looks like the path to a higher mix rate.

@sixxie
The Cybernoid 2 bass does sound off in CyD, but the C64 original doesn't sound spot on either, so I don't think the problem is entirely down to precision. If it is due to rounding errors then transposing up or down a semitone would make the errors fall in different places and change the pitch relationships. My SID knowledge is virtually nil but a quick search reveals the SID to have 16 bit counters so could have a similar magnitude of pitch error, assuming a similar range of values used to specify pitch. There could also be a psychoacoustic element to this though I can't offer much insight other than the perceived pitch depends on more than just the fundamental frequency.

@Simon
For me it's Arkanoid

[simon]

@stew...

would that be the actual arkanoid tune or the ocean loader tune ??

/Simon

[sorchard]

@simon

The Arkanoid theme is my favourite though the ocean loader in its various forms is up there as well.
Oh, I've just remembered the Wizball theme... Yet another Martin Galway masterpiece.

[sixxie]

You can lop the ORB #2 out of your loop if you set the appropriate bit in the DDR...

And I think I've scraped another cycle off, bit harder to describe though so: http://www.6809.org.uk/tmp/pcmplay/new.asm

[utz]

And my favourite C64 tune is another Galway classic - Parallax.

@Stig: Explanations of how to write music for pcmplay are in the readme. But if those are too confusing, let me know and I'll try to explain it better/easier.
I'll certainly write more tunes for this (hoping to do a proper musicdisk at some point), but it will take time. At the moment I'm tied up with quite a few other projects, so I don't expect to get around to it this year.

@pser1: Yeah, fastpoke will help, of course. Though it would be nice to combine pcmplay with some visuals, too

@sixxie: Wow, I wasn't aware of CyD. I should have a good look at that Also, what exactly do I have to change to implement your idea about changing the data direction of the PIA? Also, could somebody verify this on actual hardware?

@sorchard: Watch out, not all samples are repeating cycles, and even those that are don't always use the same cycle lengths. Look at 01-kick.smp, 02-noise.smp, or 19-chord05.smp, for example.

EDIT: @sixxie: Oh, looks you were faster than me I'll look at your new source tomorrow, am a bit too tired to properly "admire" it now

EDIT2: fixed the keyboard problem in the github release, btw. Now you can exit with Space on Dragon, too. Or with O. Or whatever else lies in that key column.

[sixxie]

@sixxie: Wow, I wasn't aware of CyD. I should have a good look at that

I bit the bullet and joined github - amazed my username was available! https://github.com/sixxie/cyd