Page 1 of 1
XMSNES - Now with SPC ENGINE AND C(++) SOURCE CODE!
Posted: Fri Jul 13, 2007 10:52 am
by Hamtaro126
This is not my utility, But it is more useful than before.
http://ekid.nintendev.com/xms/
I knew this is not dead, Plus it has a LOT of goodies!
Check the updated stuff out,
And as always: use your favorite XM tracker (Like Modplug tracker)
Posted: Wed Jul 25, 2007 9:46 am
by Bregalad
Seems interesting.
Will the WAV to BRR converter be released separately ?
Posted: Thu Jul 26, 2007 3:12 am
by dXtr
Bregalad wrote:Seems interesting.
Will the WAV to BRR converter be released separately ?
If you need one, there already exist one. Also started writing one myself, thought I never finished it (like with most my projects
), it could do brr->wav and wav->brr (thought without the filters). Was going to add them too, but other things got in the way and I havn't touched the code for a year or something like that. =/
Posted: Thu Jul 26, 2007 10:38 am
by Bregalad
Without the filters : Does it mean it only uses compression mode 0, where the sample data is 4 bit PCM ? If so that's not too much what I'm looking for, because filters 1 and 2 are more interesting.
Posted: Fri Jul 27, 2007 3:03 am
by dXtr
well in my converter there is only support for filter one (mostly b\c I didn't have the time to add the others back then).. but in the other converter I was talking about it's full support for all.
Can't remember where I originally found it, so I've uploaded it here if you need it, can't remember
clicky
Posted: Thu Aug 02, 2007 1:24 pm
by Bregalad
Note that once the converter supports filter zero it shouldn't be too hard to add the others : They only follows a slightly more complicated formula :
Filter 1 : Data = Desiredoutput - last sample * (15/16)
Filter 2 : Data = Desired output - last sample * (61/32) + 2nd last sample * (15/16)
Filter 3 : Same as filter 2 with different fractions ?
I think the best is to convert all 16 bytes of a block assuming all enabled filters, round up the results, and then re-convert in the other way to check with wich filter the last sample of the block (the 16th) ends up the closer to the original 16-bit sample, and use this filter. A good thing would be able to let the user chose wich filter to use and not to use. Another good thing would be to be able to force the looping point block to filter zero (wich is needed in most cases).