| Author |
Message |
blue hell
Site Admin
Joined: Apr 03, 2004
Posts: 24422
Location: The Netherlands, Enschede
Audio files: 297
G2 patch files: 320
|
 Posted: Sun Dec 31, 2006 10:58 am Post subject:
|
  |
|
Damn, you're right ... totaly forgot about the "normal" patches 
Problem is I can't currently patch for the Classic as I've got no working MIDI interface for it.
_________________
Jan
also .. could someone please turn down the thermostat a bit.
 |
|
|
Back to top
|
|
 |
qfingers
Joined: Nov 16, 2006
Posts: 186
Location: Tucson, AZ
G2 patch files: 2
|
| Posted: Sun Dec 31, 2006 11:21 am Post subject:
|
|
|
| Blue Hell wrote: | Damn, you're right ... totaly forgot about the "normal" patches
Problem is I can't currently patch for the Classic as I've got no working MIDI interface for it. |
Well, there are some things I can check for. I can check the Frequency settings, the FmMod input, and the Slv input. I also have the capability of finding the actual output. I build a netlist of connections that contains the single output and all the inputs it's tied to. This is how I get the Master Lfo frequency setting. It was kind of a hack because I had to sort the modules so the Master Osc/Lfos where processed first. I would like to add to the modules_table.txt I use to create my nord/g2/modules.py file a possible internal netlist for some of the simpler modules (one input and one output). Then I can search for the Master Osc or Lfo if necessary through the signal "modifier". Things like mixers, signal-signal multpliers, are more difficult to work with.
Another possibility is to allow the program to use multiple models for the NM1 module and allow some sort of configuration to tell the program to use a specific model for a specific NM1 module. This is going to require major changes to the converter.
Maybe there is a simpler solution.
q |
|
|
Back to top
|
|
|
cappy2112
Joined: Dec 24, 2004
Posts: 2494
Location: San Jose, California
Audio files: 2
G2 patch files: 1
|
| Posted: Sun Dec 31, 2006 11:41 am Post subject:
|
|
|
| Blue Hell wrote: | Damn, you're right ... totaly forgot about the "normal" patches
Problem is I can't currently patch for the Classic as I've got no working MIDI interface for it. |
What kind of patch do you need?
I've got a NM rack sitting here keeping my G2 company.
If you're really in a pinch, I could setup remote desktop access.
(if you promise not to type format C:\)  ) |
|
|
Back to top
|
|
|
3phase
Joined: Jul 27, 2004
Posts: 1189
Location: Berlin
Audio files: 13
G2 patch files: 141
|
| Posted: Sun Dec 31, 2006 12:17 pm Post subject:
Re: grey signal inputs... |
|
|
| qfingers wrote: | My question is when do I decide to apply this?
q |
I think the substitute for grey signals should be only used in special cases...the normal thing to go should be to patch the master/slave thing like its done now ( debugged of cause )...
So when the condition is that a grey cable goes to a non grey input or a non grey signal goes to a grey (slve) input soemthing has to be applied to handle that situation...
And thats probably not the ugly buildingblock i posted but i send it in to show the problem..
The grey input kondition is probably better patched in a way blue hell suggested... at least it sounds more logic to aply the scaling to the fm input instead... what hapens when the fm input is connected allready in the patch? i ve to check... actually it should be possible to just sum the signals... i ve a look how that behaves |
|
|
Back to top
|
|
|
qfingers
Joined: Nov 16, 2006
Posts: 186
Location: Tucson, AZ
G2 patch files: 2
|
| Posted: Sun Dec 31, 2006 1:05 pm Post subject:
Re: grey signal inputs... |
|
|
| 3phase wrote: | | qfingers wrote: | My question is when do I decide to apply this?
q |
I think the substitute for grey signals should be only used in special cases...the normal thing to go should be to patch the master/slave thing like its done now ( debugged of cause )...
So when the condition is that a grey cable goes to a non grey input or a non grey signal goes to a grey (slve) input soemthing has to be applied to handle that situation... |
Yes but I have seen patches that just scale the Osc Slv output thru a amp module to a Mst input. In this case the current MasterOsc (eventually fixed) solution is still appropriate. I don't know how many patches will make use of a non Slv output to Mst input and FmMod together.
q |
|
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004
Posts: 24422
Location: The Netherlands, Enschede
Audio files: 297
G2 patch files: 320
|
| Posted: Sun Dec 31, 2006 1:43 pm Post subject:
|
|
|
| cappy2112 wrote: | What kind of patch do you need?
I've got a NM rack sitting here keeping my G2 company.
If you're really in a pinch, I could setup remote desktop access.
(if you promise not to type format C:\) :-)) |
That's very kind of you Tony, but I'm affraid it would not be very practical if I can't have my hands on it ... it needs experimentation and sound comparisons ... should not have given away my good old 2x2 ...
_________________
Jan
also .. could someone please turn down the thermostat a bit.
|
|
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004
Posts: 24422
Location: The Netherlands, Enschede
Audio files: 297
G2 patch files: 320
|
|
|
Back to top
|
|
|
qfingers
Joined: Nov 16, 2006
Posts: 186
Location: Tucson, AZ
G2 patch files: 2
|
| Posted: Sun Dec 31, 2006 2:02 pm Post subject:
|
|
|
| 3phase wrote: | | 2 bugs found... on oscillators with master slave connection modulation don't gets applied to the master oscillator...see attached patches..the source..the result on g2ools 0.11 and the corrected version... |
How do I handle 2 pitch mod input case? Just move it to the OscMaster Pitch input? Or should I move it to the PitchVar input of the OscMaster?
| Quote: |
2nd bug... input of fx area gets set to +6 db even when not required..leading to unwanted distortions...
|
fixed.. next version...
q |
|
|
Back to top
|
|
|
qfingers
Joined: Nov 16, 2006
Posts: 186
Location: Tucson, AZ
G2 patch files: 2
|
| Posted: Sun Dec 31, 2006 2:14 pm Post subject:
|
|
|
| 3phase wrote: | regarding the inverted output..
In the end of the day it seems that all waveform outputs of the NM1 are inverted... Therefore it could cause bigger problems in the conversion if some oscilators get inverted while others not..
The question is if a model shall be exact... with all phase relations..
or just sound the same...
I suggest to leave inverted waveform outs out of the conversion for the moment..just with saw lfo´s´they are important...
|
I could add it. I've got initial support for low resource convertion. I haven't used it for anything yet, but for all the pitch modulations, pw modulations, fm moduations, etc.., I can remove all the Mix2-1B modules (or more if necessary) when the low resource mode is selected. Same with waveform level converter, I can remove it when low resources are selected.
I also working on fixing the PW mod stuff but it's not done yet.
q |
|
|
Back to top
|
|
|
3phase
Joined: Jul 27, 2004
Posts: 1189
Location: Berlin
Audio files: 13
G2 patch files: 141
|
|
|
Back to top
|
|
|
3phase
Joined: Jul 27, 2004
Posts: 1189
Location: Berlin
Audio files: 13
G2 patch files: 141
|
| Posted: Sun Dec 31, 2006 2:33 pm Post subject:
|
|
|
| qfingers wrote: | How do I handle 2 pitch mod input case? Just move it to the OscMaster Pitch input? Or should I move it to the PitchVar input of the OscMaster?
q |
I would suggest to utilize the second pitch var input than...
best is to create a nm1 patch having that condition..let the converter do the job and optimize manually to the desired result..
anyway..next year..i am late..
happy 2007 again... |
|
|
Back to top
|
|
|
qfingers
Joined: Nov 16, 2006
Posts: 186
Location: Tucson, AZ
G2 patch files: 2
|
| Posted: Sun Dec 31, 2006 3:22 pm Post subject:
|
|
|
| 3phase wrote: | I have a fix for the condition of normal control signals applied to a slave oscilators grey input now..works perfect..100% match..
However... enough nording for 2006... have a happy 2007...
more next year... ähhh..tomorow  |
Happy New Year!!!. It's only 2:30pm here in Tucson, AZ (I love this place).
Anyway, is this the fix you want for "all" Mst input connections? Obviously I don't have to apply it if there is no Slv output connected or Mst input connected. |
|
|
Back to top
|
|
|
qfingers
Joined: Nov 16, 2006
Posts: 186
Location: Tucson, AZ
G2 patch files: 2
|
| Posted: Sun Dec 31, 2006 3:47 pm Post subject:
|
|
|
| 3phase wrote: | I have a fix for the condition of normal control signals applied to a slave oscilators grey input now..works perfect..100% match..
|
Sorry for the double quote...
Why do you need the extra Constant module? If you set the Osc to Part mode and set it to 0, you get the same thing. Plus anything setting you set on the Constant module, can be done to the Frequency setting. Unless of course you would want one of the other Osc frequency modes: Semi, Freq, Fac. Then I could see it's usefulness. On the other hand, it's a converter and those mode are unnecessary for a converter.
Edit:
I forgot here is my TODO list, please post anything else:
- Morphs uuhhh... eventually
- PitchMod input when slave is used (fixed already for OscC but not optimal)
- All Waveforms should have inverted output
- Add pitchmod tuning to mixer settings (without the fine tuning)
X KeySplit upper and lower swapped
X Fx-In pad always at +6Db
- PWMod broke
- non-Mst/Slv broke
The ones with the X at the beginng are finished already. The ones with the - at the beginning I'm either working on or haven't touched yet.
q |
|
|
Back to top
|
|
|
cappy2112
Joined: Dec 24, 2004
Posts: 2494
Location: San Jose, California
Audio files: 2
G2 patch files: 1
|
| Posted: Sun Dec 31, 2006 4:53 pm Post subject:
|
|
|
| Blue Hell wrote: | | sound comparisons ... |
I can run the output of the NM into my laptop- only 2 channels.
Remote Desktop also has an option to transfer the remote system's audio to the client. Yes it would be tedious though, and frustrating not to have the unit in front of you. I do have a webcam |
|
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004
Posts: 24422
Location: The Netherlands, Enschede
Audio files: 297
G2 patch files: 320
|
| Posted: Sun Dec 31, 2006 5:44 pm Post subject:
|
|
|
| qfingers wrote: |
- All Waveforms should have inverted output
|
For audio osc's ? IIRC the classic labels/glyphs did not match the actual polarity of the wave. Maybe wait til someone verifies with a scope. Tony, does your webcam do scope shots as well ?
Happy newyear to everyone !
_________________
Jan
also .. could someone please turn down the thermostat a bit.
|
|
|
Back to top
|
|
|
qfingers
Joined: Nov 16, 2006
Posts: 186
Location: Tucson, AZ
G2 patch files: 2
|
| Posted: Sun Dec 31, 2006 7:56 pm Post subject:
|
|
|
| Blue Hell wrote: | | qfingers wrote: |
- All Waveforms should have inverted output
|
For audio osc's ? IIRC the classic labels/glyphs did not match the actual polarity of the wave. Maybe wait til someone verifies with a scope. Tony, does your webcam do scope shots as well ?
Happy newyear to everyone ! |
I thought they are free running, the phase would be random in that case. The only case I could see it an issue is when sync is used. That's the only way that it would matter.
q |
|
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004
Posts: 24422
Location: The Netherlands, Enschede
Audio files: 297
G2 patch files: 320
|
| Posted: Sun Dec 31, 2006 8:23 pm Post subject:
|
|
|
Hmm, in that case I meant polarity I guess and not phase 
You said "All waveforms should have inverted output"
I was wondering if (and then why) you would want to invert the output polarity of the audio oscs. The editor's visual represenation of the shape of the audio osc waves might be inverted for the NM Classic's saw waveform.
_________________
Jan
also .. could someone please turn down the thermostat a bit.
|
|
|
Back to top
|
|
|
qfingers
Joined: Nov 16, 2006
Posts: 186
Location: Tucson, AZ
G2 patch files: 2
|
| Posted: Sun Dec 31, 2006 8:57 pm Post subject:
|
|
|
| Blue Hell wrote: | Hmm, in that case I meant polarity I guess and not phase
You said "All waveforms should have inverted output"
I was wondering if (and then why) you would want to invert the output polarity of the audio oscs. The editor's visual represenation of the shape of the audio osc waves might be inverted for the NM Classic's saw waveform. |
It matters if the sync'd waveforms are 180 degrees out of phase (sine, triangle, square). It also matters if the Sawtooth is falling instead of rising. Plus it's useful to handle <50% pulse width. For the last 2 cases, I alread have a polarity reversal because 3phase showed that it mattered to some sounds. So validating the phase/polarity of the waveforms output when they are sync'd will answer the question. And I can code for it either way.
q |
|
|
Back to top
|
|
|
cappy2112
Joined: Dec 24, 2004
Posts: 2494
Location: San Jose, California
Audio files: 2
G2 patch files: 1
|
| Posted: Mon Jan 01, 2007 2:09 am Post subject:
|
|
|
| Blue Hell wrote: | | qfingers wrote: |
- All Waveforms should have inverted output
|
Tony, does your webcam do scope shots as well ?
|
No- but I can borrow a scope over a weekend and make digicam images.
If someone makes a complete list of what they want to see, I can do that.
At my last job, I used a Lecroy scope that could save screen images as bmp's over GPIB or ethernet. That would have been ideal for this. |
|
|
Back to top
|
|
|
3phase
Joined: Jul 27, 2004
Posts: 1189
Location: Berlin
Audio files: 13
G2 patch files: 141
|
| Posted: Mon Jan 01, 2007 6:22 am Post subject:
|
|
|
| qfingers wrote: | It matters if the sync'd waveforms are 180 degrees out of phase (sine, triangle, square). It also matters if the Sawtooth is falling instead of rising. Plus it's useful to handle <50% pulse width. For the last 2 cases, I alread have a polarity reversal because 3phase showed that it mattered to some sounds. So validating the phase/polarity of the waveforms output when they are sync'd will answer the question. And I can code for it either way.
q |
the new PW modulation block dont needs the inversion anymore because you now can directly copy the initial pw 1:1 and don have to bother with modulations that appear on <50% pw...therefor this module is better ...
because there is no need anymore to have output polarity in sync with the control signal polarity. Its possible to forget about audio waveform inversions for the G2.. especially because that fucks up the sync connections...
the waves on the G2 might have inversed polaritys in relation to the NM1..but the sync input is changed aswell...
so after discovering that i suggest to dont do polarity changes on audio waves for now...
It still have to be fixed on lfos..an uprising filtersweep shouldnt become a falling one...
regarding the grey input...
For all master osc/ slave osc conditions the patch should be converted in a way you have done it allready... that is much more convinient and resource safing on the G2... its just not necessary to simulate grey signals...
exept the patches where grey signals are used for other purposes... where a control or audio (blue or red) kabel goes to a slave input...
there the patch i posted after blue hells suggestion should be applied...
The constant module for the -64 offset i only added for the convinience that you just translate the initial freq of the osc 1:1...
of cause you can set the osc to zero... but that would probaly not match patches where the oscilator is tuned down..so you get to the lower border of the note range quicker within a modulation..what probably would result in a different sound...
but..maybe ..on such extrem fm mod modulations as they appear via the grey signals its possible that the initial osc setting is over blown somehow... but this is theorie..with the -64 offset constant modules its a proofed situation...
I would say that when you ve done the next version we check again the conversions of such critical patches and i can do a 1:1 sound comparison...
and tell you if it needs to be further modified or not...
regarding your dsp safe mode.. we should see where this is causing problems..one for sure... for pitch modulated patches you get probably not in the nice sounding area without the tables... the difference within the response curves of nm1 and g2 is often about 20 cent detune..that is in many cases enough to let something sound nasty that was supposed to be beautifull...the ear is very sensetiv on pitch...
However..the converter starts to get work done..i havent tested so much patches..but now its one in ten that is totally of or not converted while a few days ago i had 2 in 10 converted...
thats really some progress..i am personally convinced now that the endresult will so damn close that you really acces the old lib without sonical loss... what is a great thing...
people really have done some work for the nm 1..there are good sounds in the lib..much more than today availible for the g2...
Happy new year to all of you... |
|
|
Back to top
|
|
|
qfingers
Joined: Nov 16, 2006
Posts: 186
Location: Tucson, AZ
G2 patch files: 2
|
| Posted: Mon Jan 01, 2007 7:56 am Post subject:
|
|
|
| 3phase wrote: | | so after discovering that i suggest to dont do polarity changes on audio waves for now... |
I'll remove the level converter from the convertion.
| Quote: |
It still have to be fixed on lfos..an uprising filtersweep shouldnt become a falling one... |
I'll keep it here.
| Quote: |
regarding the grey input...
For all master osc/ slave osc conditions the patch should be converted in a way you have done it allready... that is much more convinient and resource safing on the G2... its just not necessary to simulate grey signals...
exept the patches where grey signals are used for other purposes... where a control or audio (blue or red) kabel goes to a slave input...
there the patch i posted after blue hells suggestion should be applied...
The constant module for the -64 offset i only added for the convinience that you just translate the initial freq of the osc 1:1...
of cause you can set the osc to zero... but that would probaly not match patches where the oscilator is tuned down..so you get to the lower border of the note range quicker within a modulation..what probably would result in a different sound...
but..maybe ..on such extrem fm mod modulations as they appear via the grey signals its possible that the initial osc setting is over blown somehow... but this is theorie..with the -64 offset constant modules its a proofed situation... |
I can look that the output connection that's tied to the Mst input. If it's not a gray rate output, I'll apply the new model, otherwise I'll apply the old model. If it becomes a problem, I'll come up with another way to resolve it.
| Quote: |
regarding your dsp safe mode.. we should see where this is causing problems..one for sure... for pitch modulated patches you get probably not in the nice sounding area without the tables... the difference within the response curves of nm1 and g2 is often about 20 cent detune..that is in many cases enough to let something sound nasty that was supposed to be beautifull...the ear is very sensetiv on pitch...
|
It will be a command-line option so you can run it and check the result:
| Code: |
./nm2g2.py --low patch.pch
|
This would remove the fine tuning (which is yet to be determined). The fine tuning mixers we added to the inputs of the modulation could be removed. Then if the result wasn't satisfactory, you run the converter the normal way:
| Code: |
./nm2g2.py patch.pch
|
Then it will be the normal processing with the fine tuning included.
| Quote: |
However..the converter starts to get work done..i havent tested so much patches..but now its one in ten that is totally of or not converted while a few days ago i had 2 in 10 converted...
thats really some progress..i am personally convinced now that the endresult will so damn close that you really acces the old lib without sonical loss... what is a great thing...
people really have done some work for the nm 1..there are good sounds in the lib..much more than today availible for the g2... |
Are you saying 9 output 10 patches convert correctly or 1 output 10 patches convert correctly? If it's 9 out of 10, that's 90% of course, I was hoping for 50% to 60%. If it's 1 out of 10, I gotta a lot of work to do.
q |
|
|
Back to top
|
|
|
3phase
Joined: Jul 27, 2004
Posts: 1189
Location: Berlin
Audio files: 13
G2 patch files: 141
|
| Posted: Mon Jan 01, 2007 11:04 am Post subject:
|
|
|
it s giving usefull results with 9 from 10 patches now...but there are still details... I havent had the time to do much conversion because i reported steadily points i came along... its sometimes not so easy to spot why patches sound different..in the end all the little tweaks will add up to a real conversion of the sonical content of a patch...
I am surrprised allready how good some things can be translated to the G2 when you dig out the details...
I ve to check the pitch table again..it appears that Nm1 and G2 are slightly out of pitch to each other... at least my machines
not much, around 1cent, but i think the result of the last translation where 127 to 127 gets a slight minus offset on the G2 is result of that... i ve to confirm if its enough to to change all fine tune values by 2 positive ticks to get it wright... i spend quite some time to get as little beating as possible... most of the 127 values i got to zero beat with the oscilloscope... very little tuning diffrences are rather to see than to hear...when it needs over a minute for one cycle its hard...
however ...its a little missmatch within 1 cent detune or so..but when i check a few values i might be able to confirm if its just enough to re offset all 127 values... its defenetly a mistake that the high value has a negative offset ...the tuning cheme of the nords is doing clean octaves...
so in any case the 127 should translate to 127 without any tunings |
|
|
Back to top
|
|
|
dasz
Joined: Oct 16, 2004
Posts: 1644
Location: victoria, canada
Audio files: 29
G2 patch files: 56
|
| Posted: Mon Jan 01, 2007 11:28 am Post subject:
|
|
|
I haven't poked my cat head in this discussion for a while, but after reading some posts, I am once again blown away by the effort and attention to detail from you guys. A big thanks again!
a heartfelt happy new years to you all!
/Dasz |
|
|
Back to top
|
|
|
G2DREAM
Joined: Apr 27, 2005
Posts: 171
Location: Athens,Greece
Audio files: 5
G2 patch files: 3
|
| Posted: Tue Jan 02, 2007 2:31 am Post subject:
|
|
|
Sorry to interrupt you all my friends but i just found the time(yyyyyeeeyy!!) to try the g2ools!
I installed python 25 but now all i am getting when i try to run the nm2g2.py is ms-dos windows that they open and close really fast!
How exactly i can use this software?
Thank you very much guys and sorry for the newbie question! |
|
|
Back to top
|
|
|
cappy2112
Joined: Dec 24, 2004
Posts: 2494
Location: San Jose, California
Audio files: 2
G2 patch files: 1
|
| Posted: Tue Jan 02, 2007 9:07 am Post subject:
|
|
|
| G2DREAM wrote: | Sorry to interrupt you all my friends but i just found the time(yyyyyeeeyy!!) to try the g2ools!
I installed python 25 but now all i am getting when i try to run the nm2g2.py is ms-dos windows that they open and close really fast!
How exactly i can use this software?
Thank you very much guys and sorry for the newbie question! |
you need to open a cmd console first, then type
nm2g2.py patchname.pch2
it wont work by double clicking. It sounds like this is what you are doing, |
|
|
Back to top
|
|
|
|
Forum index » Clavia Nord Modular » Nord Modular G2 Discussion
FAQ
Calendar
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
Chat Room
