Özkan Sezer
Özkan Sezer
You are right, there are several others than display significant differences.
> You are right, there are several others than display significant differences. Took all the files from http://ftp.modland.com/pub/modules/Ultratracker/ and ran a test on them printing the finetune and calculated speed...
I don't know. Playing with the non-zero finetune values, I don't know how I can get an acceptable result without using pow(): ```diff --- original.txt +++ with-pow.txt @@ -1,45 +1,45...
@AliceLR: what do you suggest?
> I still think a numerical approximation could be used here depending on the domain of this function. Indeed - If you have a solution, that'd be nice. > edit:...
With Alice's version, `Break The Beat` sounds very close to libxmp's output. Thomas, what do you say?
Alice: apart from the pow approximation (which I intend to use yours, waiting to hear from Thomas), is the rest of the patch OK?
OK, I pushed ed3e5b3394a07669055a694f0ce3e60ebb905c48 for the sample speed issue. With this, I have only two modules giving different results than pow(): cyboccultation.ult finetune: 678 -> 8373 (pow) -> 8372 spherical...
Waiting for the solution for portamento issue and the rest (i.e. items 2 and 3 on Thomas' original list).
> I have a working tone portamento fix for libxmp that handles various minor quirks. Presumably this could be ported to MikMod. It might be more work since, as far...