emusonacid! legacy

I have been trying for sometime to create a string sound that is convincing using test tones. I have played around with using saws and triangles for this, and have had some okay results. but nothing all that spectacular. if one is to use a triangle or saw, the slow-ish attack and the long decay just aren't enough to convience the listener that this is suppose to be strings. they are simply too buzzy. so a lowpass filter would have to be used to somewhat smooth out the sound. and perhaps q added in an attempt to retain the higher frequencies. Like I said, I haven't had the greatest of luck with this technique. That is of course until now...

this is by no means perfect, but i believe it is getting damn close to how a string sounds. it is a bit dull in the higher harmonics, as if it where lowpassed, but, assuming this could even work, with proper eqing after recorded into an audio program, it might just be possible to high shelf the dull frequencies to brighten the sound. Or perhaps there might be just one other wave that could be added to this "patch" to boost the upper harmonics. this is something i haven't played with. Okay, here we go. this test tone i got from a demo of wavelab that came with my pro audio 8 some years ago. That demo has an extraordinary amout of tones and wave shapes you can sample. Hopefully the new versions have this as well... i haven't played with those yet. Plus the tones form that program are razor sharp compared to the tones out of the emu. Any way...

what you need to do is get a hold of a Rectified Sine Wave. I now for some reason can't remember what it looks like. If I remember right it is just the positive bits of the sine wave, with no negative deeps. but I think that is describing a halved sine wave, of which i sampled from wavelab.

once you have the rectified sine wave, take it and copy the voice four times. You might get the same results with lower voices, but the idea with four voices is that the detuning isn't quite so destructive when applied. I have also found that placing samples at the sample manage screen or the section just after you sample a waste of time. If you go right into the multi voice edit screen in a preset (alt A) you can set your sample that you want on voice one, and then set your origin, your range is automatically the full range. this may not be useable for some, but for others...

once you have copied the wave four times you want to go to voice two and set the ftune to 7, the third voice to 11 and the fourth voice to 21. now move the cursor to the group and edit them all to have an adsr setting (for the amp) 20 for the first stage attack, and 50 for the decay, or whatever you like.

you should now have a nice-ish sounding string section. just need to get clarity with it now.

other setting for the detuning offers other interesting results. For a more electronic sounding string sound, try this: voice 2 set to 3, voice 3 to 6, and voice 4 to 9. you get more flange in this setting and a bit brighter sounding high end. also try 5, 10, and 15 for detune settings. One thing i would stay away from are, unless you want this sound, ftune settings higher then 30 or 40, you start getting the minor second kind of sound. remember 64 on the ftune is equal to a half step so setting your second voice at 64 is about the same as setting the ctune at 1.

oh and remember to either sample in mono or convert down to mono, saves polphony. this is a four voice patch running at mono, when it comes to stereo, you are of course up to 8. so a triad would be 24 voices in stereo...

if you guys can't get this wave type, say so and i will post the waves up on this thread, perhaps 1/2 to 1 second lenghts.

hope you guys find this useful. closest i have gotten so far... at least with a result that i like..

cheers for the wavlab test tone heads up, i knew that prog had to be good for something