| |
chatGPZ
Registered: Dec 2001 Posts: 11433 |
Testers needed: Saw+Pulse waveform samples have been wrong for 25 years
Quoting Dag Lem from here:
The 6581 pulse + sawtooth samples were obtained around 25 years ago. At the time, it was not known that the oscillator MSB can be zeroed via combined waveforms. The following program takes this fact into consideration, and obtains correct samples for the complete waveform:
https://github.com/daglem/reDIP-SID/blob/master/research/pulse-..
https://github.com/daglem/reDIP-SID/blob/master/research/pulse-..
It would be helpful if others could run the .d64 to obtain samples from different 6581s. The 6581 samples in reSID are comparatively strong (more non-zero values than average), and it would be nice if we could get samples which more or less match the existing ones (for MSB = 0). |
|
| |
ws
Registered: Apr 2012 Posts: 251 |
Would a Reloaded Board somehow influence the data sample of a SID? (In case i put an original 6581 back in it?)
Or is just a specific original board build with an original MOS6581 Chip desirable for this data sample?
Is the ARMSID (if in 6581 mode) actually generating a "legit" 6581 circuit signal or is it generating waves based on those old sample tables? And would an ARMSID in 6581 mode data sample actually be of any help? |
| |
chatGPZ
Registered: Dec 2001 Posts: 11433 |
The Board does not matter. And any sort of SID that isnt actually a SID does not help, of course. They are probably using those exact wrong samples from 25 years ago :) |
| |
ws
Registered: Apr 2012 Posts: 251 |
Quote:They are probably using those exact wrong samples from 25 years ago :)
That was exactly, why i was asking. Thanks. |