Archive of the former Yahoo!Groups mailing list: Korg Poly800/EX800 Users

A new software release for the HAWK-800/AtomaHawk will be coming out within a week or so.

It includes a very important improvement to the AtomaHawk FM-800 mod features.

The result of the improved software is that the operating range of the FM-800 modulation is now (internally) 1-192 (instead of 1-128) and it is now possible to use the LFO's or EG3 to smoothly sweep the FM-800 mod through its entire range.

I created a bunch of patches last night that highlight the FM-800 AtomaHawk features and (IMHO) they sound absolutely fantastic and are far superior to what could be achieved before now.

I will package up the code and the patches I created into a zip file and will post it very soon. For now, you will all just have to salivate.

For those of you who like technical detail...

The digital potentiometer that was used in the AtomaHawk is actually a quasi audio pot (Dallas DS1666+ 10K). It uses two resistor ladder networks to fake being a logarithmic pot. The range of the pot is 0-127 where the first half of the range has a slope X and the second half has a slope 2 times X. This gives two different gradients that approximate a crude log scale. It works well in car radios but the problem with the AtomaHawk is that the FM-800 mod is feeding the DCO signal into the VCF on a linear control. This causes the modulation (LFO and EG3) to operate the range of the FM-800 at two different rates.

Up until now, that has caused the FM-800 mod to be a bit of a problem when trying to modulate using an LFO or EG3. For example, you couldn't get EG3 to sweep smoothly when operating the FM-800 mod. So I have rewritten the FM-800 modulation code so that from 0-63 the digital pot moves one to one but from 64 to 192 it moves only on odd steps. This overrides the original 2x slope of the second half of the pot wiper.

Now you may be wondering why I used a logarithmic digital pot in the first place. Well, the DS1666+ is one of the only digital pots around that supports a voltage rail that is negative from the logic signals. In the Poly-800 the logic operates at 0V/+5V while the analog circuits operate at -5V/0/+5V so the DS1666 was able to drop straight into the Poly-800 without any complex logic level conversion electronics. Now that the quasi logarithmic wiper response has been dealt with, we have a "perfect" solution.

Mike.