Jump to content

Korg DSS-1 upgrade finished - 16MB RAM, 1.44 MB FD, USB, & more


McHale

Recommended Posts

  • Members

Tom,

 

Excellent work!!! Nice to see you on here.

 

Any chance you might be able to do something about the E-MU Emulator II along these lines? I suppose it's a long shot at best, but I had to ask. :)

 

Even a simple RS-422 to USB or SCSI device would go a long way.

Link to comment
Share on other sites

  • Replies 345
  • Created
  • Last Reply
  • Members

Glad that you are enjoying the mod!


One of the limitations of the VCF board on the DSS-1 is that some of the control voltages are common for all voices. The only ones that are individual are VCA and VCF. The VCA is common for both inputs. There are 2 control voltages which I assume are for each audio input, LEV1 and LEV2, but they are all tied together.


I'll ask Jim for the system file format, he did say he would offer it to anyone who is interested.


There is also a default system load feature, any system in the root directory with the name DEFAULT is loaded automatically from USB upon boot.


-Tom

 

 

tom in the dss1 you can control the volumes of each osc layer i thought this was done by control voltage of the mixer in the 2069 ic so adding a sofware control off the osc mixer should provide a linear crossfade of the 2 layers adding the lfo to control this would give a ever evolving sound of the morphing layers

 

also is it possible to set a control of lfo amount by envelope in software for the dss1 so that you can modulate the lfo over time giving more expression to timbre of the sounds

 

free running oscilators would give the synth a more analog feel as it would simulate oscillater drift which gives that beating sound when more than 1 layer is used

 

if he can add loop start adjustment so that it can move slightly every so many cycles and a single cycle wave could be doubled so that the set sizes of the single cycle waves of 4 8 16 32 64 128 256 512 samples sizes could then change phase slightly if the loop size was constantly kept at these sizes but just moved slightly to the left or the right this would simulate the drift and would keep the pitch of the sound as the sample loop size would be constant

Link to comment
Share on other sites

  • Members

tom in the dss1 you can control the volumes of each osc layer i thought this was done by control voltage of the mixer in the 2069 ic so adding a sofware control off the osc mixer should provide a linear crossfade of the 2 layers adding the lfo to control this would give a ever evolving sound of the morphing layers


also is it possible to set a control of lfo amount by envelope in software for the dss1 so that you can modulate the lfo over time giving more expression to timbre of the sounds


free running oscilators would give the synth a more analog feel as it would simulate oscillater drift which gives that beating sound when more than 1 layer is used


if he can add loop start adjustment so that it can move slightly every so many cycles and a single cycle wave could be doubled so that the set sizes of the single cycle waves of 4 8 16 32 64 128 256 512 samples sizes could then change phase slightly if the loop size was constantly kept at these sizes but just moved slightly to the left or the right this would simulate the drift and would keep the pitch of the sound as the sample loop size would be constant

 

 

by the way tom im jammie from england who had 1 of the first kits from you have you any more of those keyboards left that you have been upgrading

Link to comment
Share on other sites

  • Members
Tom,


Excellent work!!! Nice to see you on here.


Any chance you might be able to do something about the E-MU Emulator II along these lines? I suppose it's a long shot at best, but I had to ask.
:)

Even a simple RS-422 to USB or SCSI device would go a long way.



I don't know too much about the EII apart from the fact that I really wanted one in my youth. I know there was a 'EII+HD' version so there was a SCSI version for sure. Whether it was an upgrade or a whole separate motherboard remains to be seen. I have a few projects in the pipeline first but I'll give it a look.

Link to comment
Share on other sites

  • Members

 

tom in the dss1 you can control the volumes of each osc layer i thought this was done by control voltage of the mixer in the 2069 ic so adding a sofware control off the osc mixer should provide a linear crossfade of the 2 layers adding the lfo to control this would give a ever evolving sound of the morphing layers

 

 

You can control the OSC1 and OSC2 levels independently so in theory the crossfading is possible. However because of the way the analog card is designed, the crossfading would be synchronized across all voices every time you press a key. So maybe not the exact effect you are looking for

 

 

also is it possible to set a control of lfo amount by envelope in software for the dss1 so that you can modulate the lfo over time giving more expression to timbre of the sounds

 

 

That's a question for Jim...It should be possible since the envelopes are done in software.

 

 

free running oscilators would give the synth a more analog feel as it would simulate oscillater drift which gives that beating sound when more than 1 layer is used

 

 

I think Jim has added this via the MONO key assign mode.

 

 

if he can add loop start adjustment so that it can move slightly every so many cycles and a single cycle wave could be doubled so that the set sizes of the single cycle waves of 4 8 16 32 64 128 256 512 samples sizes could then change phase slightly if the loop size was constantly kept at these sizes but just moved slightly to the left or the right this would simulate the drift and would keep the pitch of the sound as the sample loop size would be constant

 

 

The basic problem is that the voice generator hardware doesn't communicate where the current sample position is when it's playing back. So you would have to write different loop start values blindly and hope that it doesn't glitch. Jim mentioned that changing the loop start point is possible, so we'd have to experiment with this.

Link to comment
Share on other sites

  • Members

by the way tom im jammie from england who had 1 of the first kits from you have you any more of those keyboards left that you have been upgrading

 

 

I thought it was you Jammie. I have about 6 or 7 DSS-1's left, just sold one to a customer in Hawaii.

Link to comment
Share on other sites

  • Members

if he can add loop start adjustment so that it can move slightly every so many cycles and a single cycle wave could be doubled so that the set sizes of the single cycle waves of 4 8 16 32 64 128 256 512 samples sizes could then change phase slightly if the loop size was constantly kept at these sizes but just moved slightly to the left or the right this would simulate the drift and would keep the pitch of the sound as the sample loop size would be constant

 

Loop end cannot be updated while the sample is playing, only Loop Start. Updating only Loop Start would change the pitch, so that is not desired.

 

-Jim

Link to comment
Share on other sites

  • Members

As Tom mentioned, the OSC 1 / OSC 2 Mix Ratio control is hard wired to all 8 voices. I do let you assign this parameter to one of the sliders, so you can morph between the two multisounds as you play.


It sounds like you want an envelope to control the envelope. I'm not sure how that would be presented to the operator. We already have a setting where the velocity of the key pressed can affect some of the LFO parameters.


I'm looking at adding the Random Sample Start that Juno6 had suggested. This should give the desired affect.


Loop end cannot be updated while the sample is playing, only Loop Start. Updating only Loop Start would change the pitch, so that is not desired.


-Jim

 

 

so theres no way of changing loop position

 

i know the 8 voices are tied in the crossfade which is fine

 

what i ment about the lfo is say its speed or amount of effect could be modulated by a envelope the effect would be a pulse width lfo which when set to morph the mixer or filter modulation would give an exciting effect

 

its done on the korg ms2000 you can use lfo 1 to modulate lfo 2 or envelope to modulate the lfo obviously the mg would also work on the effects

Link to comment
Share on other sites

  • Members

Since Jim is making many revisions to the OS, I was wondering if it is possible to add more waveforms for the LFOs (MGs). Currently there is only the sine wave LFOs. Having a square, triangle and S/H wave LFO would be nice!

 

Are the LFOs in the DSS-1 analog or digital? I would think if they were digital, it would not be a big deal adding the code to add the additional waveforms.

Link to comment
Share on other sites

  • Members

Since Jim is making many revisions to the OS, I was wondering if it is possible to add more waveforms for the LFOs (MGs). Currently there is only the sine wave LFOs. Having a square, triangle and S/H wave LFO would be nice!


Are the LFOs in the DSS-1 analog or digital? I would think if they were digital, it would not be a big deal adding the code to add the additional waveforms.

 

 

Yes, they are digital. I could add different shapes. What do you mean by "S/H wave"?

 

-Jim

Link to comment
Share on other sites

  • Members

Since Jim is making many revisions to the OS, I was wondering if it is possible to add more waveforms for the LFOs (MGs). Currently there is only the sine wave LFOs. Having a square, triangle and S/H wave LFO would be nice!

 

 

Good idea!

 

S&H means a sample & hold giving new random values at a rate set by the LFO

Link to comment
Share on other sites

  • Members

No, the LFOs are per voice. The real limitation is that the filter control voltages are handled by CPU-2, which I have so far not changed.

So for now, I would only be adding features to the Oscillator MG.

Any shape that can be put in a look-up table and used in place of the sine wave table should be fairly straight forward to do.

I guess I can add a program parameter 20 to select the OSC MG waveform.


-Jim

 

 

Ohhh... that would be great! I don

Link to comment
Share on other sites

  • Members

 

No, the LFOs are per voice. The real limitation is that the filter control voltages are handled by CPU-2, which I have so far not changed.

So for now, I would only be adding features to the Oscillator MG.

Any shape that can be put in a look-up table and used in place of the sine wave table should be fairly straight forward to do.

I guess I can add a program parameter 20 to select the OSC MG waveform.


-Jim

 

 

 

Jim this is awesome! I haven't bought a DSS-1 from Tom yet, but this indeed would make the synth as versatile as it can be as far as modulation....now if it had a Mod Matrix with at least four slots....now that would be killer!

 

Tom's mods made me really look at the DSS-1 and it's your OS updates Jim that are going to make me buy this synth.

Link to comment
Share on other sites

  • Members

No, the LFOs are per voice. The real limitation is that the filter control voltages are handled by CPU-2, which I have so far not changed.

So for now, I would only be adding features to the Oscillator MG.

Any shape that can be put in a look-up table and used in place of the sine wave table should be fairly straight forward to do.

I guess I can add a program parameter 20 to select the OSC MG waveform.


-Jim

 

 

I feel Jim can get pretty creative with the LFO shapes. Off the bat I can name Triangle, Sawtooth, Pulse, Sine (already there), Random (sample and hold), Noise and if another parameter can be added for Reverse (Positive/Negative)...wow, that's about as flexible as anything I've seen on any synth.

 

Jim, are you okay with us putting you to work like this? I would be happy to paypal you some bucks for these updates.

 

:-)

Link to comment
Share on other sites

  • Members

I feel Jim can get pretty creative with the LFO shapes. Off the bat I can name Triangle, Sawtooth, Pulse, Sine (already there), Random (sample and hold), Noise and if another parameter can be added for Reverse (Positive/Negative)...wow, that's about as flexible as anything I've seen on any synth.


Jim, are you okay with us putting you to work like this? I would be happy to paypal you some bucks for these updates.


:-)

 

 

if adding lfo tables can you use any digital single cycle wave

 

as i have thousands which would make good modulators

 

you could get barber effects

 

free running would be good aswell playing any key with the lfo not retrigering

 

also would it be possible to add human feel which is when the lfo slightly speeds and slows to give it a more natural effect

Link to comment
Share on other sites

  • Members

As Tom mentioned, the OSC 1 / OSC 2 Mix Ratio control is hard wired to all 8 voices. I do let you assign this parameter to one of the sliders, so you can morph between the two multisounds as you play.


It sounds like you want an envelope to control the envelope. I'm not sure how that would be presented to the operator. We already have a setting where the velocity of the key pressed can affect some of the LFO parameters.


I'm looking at adding the Random Sample Start that Juno6 had suggested. This should give the desired affect.


Loop end cannot be updated while the sample is playing, only Loop Start. Updating only Loop Start would change the pitch, so that is not desired.


-Jim

 

 

ive been doing tests on wave forms from the sq80 which does free form oscillators

 

the drift is so slight as long as the loop start position is applied in very small increments then back again the change in pitch is very small but adds excitement to the waveform

 

i also did an analysis on a violin wave and its the same type of drift due to the performers movement off the bow which gives it very slight pitch variation

 

as long as the loop start is not extreme the pitch change should not be noticed to much by the listener

 

and would add the effect of free form oscilators

 

ive been looking at this at depth and the function of moving loop start will cause clicks as the wave form will start not from zero crossings

 

the only way it would work would be for the sample loop size to be moved forward and back word so that the waveform start will always match the waveform end i know this cant be done in the dss1 as theirs no loop potion markers

 

so i think to achieve the same effect would be to use the the DE-tune parameter for the whole of the multisound layer to be slightly moved in small increments at a set frequency

 

this small adjustment would add the drift effect its the only way to achieve it with out wave modulation

 

can you program the detune of the layer like this for this effect this way each note would have the same variation which is what analogs do

Link to comment
Share on other sites

  • Members
Jim, are you okay with us putting you to work like this? I would be happy to paypal you some bucks for these updates.


:-)



It's fine, I asked for ideas! This is just a hobby for me, I never intended to make money off it. :thu:

-Jim

Link to comment
Share on other sites

  • Members


also would it be possible to add human feel which is when the lfo slightly speeds and slows to give it a more natural effect

 

 

What if I expand the aftertouch parameter to be able to control the MG frequency instead of just intensity? Would that work?

 

 

-Jim

Link to comment
Share on other sites

  • Members

What if I expand the aftertouch parameter to be able to control the MG frequency instead of just intensity? Would that work?



-Jim

 

it needs to be able to go back and forth at very small increments so not to add vibrato

Link to comment
Share on other sites

  • Members

When I say that they are per voice, I mean that the hardware registers are there for individual control. The original operating system may have done things in software that applied a single parameter across all 8 voices (or 16 oscillators).

As an example, the original parameters for autobend was a single variable for starting offset and a single variable for direction. When I added Portamento, I needed these variables to be per voice, so that as you play, the direction and offset of the autobend affect for each voice was based on the previous note played.

The oscillator MGs are all done in software via an interrupt routine, and there are 16 sets of counter registers that control the rate that the samples clock out for each of the oscillators.

There are global parameters such as from the joystick and aftertouch that get added to each of these parameters.


So since it is just a software change, yes, I should be able to change the starting phase of the LFO or let it free run.


-Jim

 

 

can you add different wave forms for the lfo and what size in samples do you need for this

Link to comment
Share on other sites

  • Members
i have thousands of transwaves as i have been making them for 15 years and have a small program that i resynthesis samples which turns it into spectra data of 256 sample lenghts in a table of 64 cycles like the wave and microwave xt and blofeld



Wow, this is really cool! I just bought an Ensoniq TS10 for making my own transwaves, and two days ago I found a guy from Germany who did a program for transwave creation and coincidentally you also did! Is there any chance of having access to that program and the trasnwaves??? (please, please, please! :love:)

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.


×
×
  • Create New...