Search options (Page 11 of 16)

Thank you.

Turns out the OPL envelope is all kinds of funky (generally e.g. the attack is the exact opposite to what is in use here). I think the envelope generator in klystrack should overall be augmented so that you can pick if the attack/decay phase is exponential or linear or whatever.

That is correct.

The high nibble of the MULT param, duhhh! Seriously though, if you have MULT set to 3F, it means carrier is multiplied by 3 and modulator by 15 (and the whole instrument is magically transposed so it doesn't play out of tune, in this case the instrument would play a fifth higher but it's transposed down). You can shift-click the up and down buttons and it will change the high nibble. Generally, you can probably just randomly click the param to some setting until you find an interesting harmonic.

There needs to be a separate parameter for it but I didn't want to spend the whole Sunday afternoon working on silly projects!

Feedback should work better now btw. Also, I tried to create some sounds by ear and compared those to some Youtube videos of old games and in many cases you get very close to the original.

Look at the klystrack wiki, it has instructions how to get the sources (you need SVN) and how to compile.

http://code.google.com/p/klystrack/wiki/HowToCompile

Edit: Updated the article a bit.

Sound effects and drums and that stuff comes to mind when you might benefit from some weird ratio that has two prime numbers. Perhaps you can also "color" the carrier with an atonal modulator that's barely there?

Well, the OPL has a multiplier on all oscillators, modulator or not. Mathematically it is the same to have a CARRIER:MODULATOR ratio of like 3:1 and a modulator frequency of 0.333 * CARRIER. But my reasoning is that if you only multiply the frequency it gives you a stabler waveform (100 % my own speculation) since there is no rounding involved in the maths when you only multiply by whole numbers. Honestly, though, it might be overkill because the oscillators already have lots of accuracy and who cares if the oscillators eventually go out of sync after a minute of playing the same note...

The best part of feedback is the black magic! IMO it was very smart to include that in the chip.

It is ok, thanks. The most important thing is that the catometer is visible!

Ok, so FM: harmonics sets the frequency multiplier to a preset value, I just randomly put some values in there, on the OPL there are multipliers such as 0.5, 1, 2-10, 12, 15, that translate to some interesting scales. Though, on the OPL you can also set the multiplier for the carrier as well so I think my implementation misses some harmonics (e.g. put carrier to x5 and modulator to x7), so I need to add some more fractional values (now there only are multipliers for 0.125, 0.25 and 0.5, i.e. three octaves down, two octaves down and one octave down).

Please do tell if you can't make some kind of sound that was possible on the OPL chips, I hardly have any experience with them. I am not sure if you can get that bell-like sound that's a very charateristic FM sound. Interestingly, if you modulate a triangle carrier with the triangle modulator, it sounds like sync modulation due to the modulator sort of stopping and starting the carrier, because it goes either up at full speed or down at full speed.

The modulator simply modulates the oscillator frequency (or rather oscillator phase), so you can use the multiple waveform selection thing with it. Though, LFSR isn't compatible with phase modulation so it is not touched at all (that also means noise). But IMO sinewaves are the only way to fly!