Author |
Message |
Hyde
Joined: Jul 15, 2004 Posts: 118
G2 patch files: 2
|
Posted: Sun Jan 21, 2007 6:55 pm Post subject:
|
|
|
qfingers wrote: | Hyde wrote: | thanks q.
i know all the patches work on my Micro because i use them often. |
I found the problem already. It's a NM1 file parser problem. I don't want to create a release until a few more of these things are found. Because there was only 2 sets of eye looking at this before. I'll create a updated release later. The number of bugs and complexity will determine when it gets released.
The problem was from modules being labeled with '[' and '] within them. The parser used these characters for determining the beginning and ending of sections. It's fixed in my version now.
q |
THANK YOU Q for finding it and fixing it. i will just wait for the new release then. you are the man |
|
Back to top
|
|
|
ian-s
Joined: Apr 01, 2004 Posts: 2669 Location: Auckland, New Zealand
Audio files: 42
G2 patch files: 626
|
Posted: Sun Jan 21, 2007 8:21 pm Post subject:
|
|
|
g2ian wrote: | Thanks for the DX2G2 tool.
What exactly is it expecting as input?
It just returns quickly to a prompt with no message on sysex bank files, with or without headers. |
OK to answer my own question, looking at the source code, the program is looking for .syx specifically and will process single patch or 32 patch bank dumps. It will not work with RAW files which have no sysex header, and some of the banks from the internet have extra sysex manager headers which will also cause failure.
Looks like it will create 4 pch2 files per bank, nice.
I don't have a single patch dump to test but the bank conversion results in
ROM2A.SYX 32 voice
ROM2A.SYX 32 voice
ROM2A.SYX 32 voice
ROM2A.SYX 32 voice
being repeated till you press Ctrl+Break, and no output, anyone had any luck with this? |
|
Back to top
|
|
|
qfingers
Joined: Nov 16, 2006 Posts: 186 Location: Tucson, AZ
G2 patch files: 2
|
Posted: Sun Jan 21, 2007 9:18 pm Post subject:
|
|
|
g2ian wrote: | g2ian wrote: | Thanks for the DX2G2 tool.
What exactly is it expecting as input?
It just returns quickly to a prompt with no message on sysex bank files, with or without headers. |
OK to answer my own question, looking at the source code, the program is looking for .syx specifically and will process single patch or 32 patch bank dumps. It will not work with RAW files which have no sysex header, and some of the banks from the internet have extra sysex manager headers which will also cause failure.
Looks like it will create 4 pch2 files per bank, nice.
I don't have a single patch dump to test but the bank conversion results in
ROM2A.SYX 32 voice
ROM2A.SYX 32 voice
ROM2A.SYX 32 voice
ROM2A.SYX 32 voice
being repeated till you press Ctrl+Break, and no output, anyone had any luck with this? |
Well what should happen is each patch name displayed with a number preceding it. Post the ROM2A.SYX and I'll run a test on it. Maybe it's something I overlooked in the SYX file.
q |
|
Back to top
|
|
|
3phase
Joined: Jul 27, 2004 Posts: 1183 Location: Berlin
Audio files: 13
G2 patch files: 141
|
|
Back to top
|
|
|
ian-s
Joined: Apr 01, 2004 Posts: 2669 Location: Auckland, New Zealand
Audio files: 42
G2 patch files: 626
|
|
Back to top
|
|
|
qfingers
Joined: Nov 16, 2006 Posts: 186 Location: Tucson, AZ
G2 patch files: 2
|
Posted: Mon Jan 22, 2007 6:44 am Post subject:
|
|
|
g2ian wrote: | qfingers wrote: | Post the ROM2A.SYX and I'll run a test on it. Maybe it's something I overlooked in the SYX file.
q |
Ta q
Here it is, Factory Rom 2A |
It converted fine under linux. Which OS are you running? There must be a difference in reading binary files. Under windows I would have to change the line:
Code: | f = open(fname,'r') |
to
Code: | f = open(fname,'rb') |
It may be as simple as that. I'm adding it to my version and will be releasing a bugfix release today.
q |
|
Back to top
|
|
|
ian-s
Joined: Apr 01, 2004 Posts: 2669 Location: Auckland, New Zealand
Audio files: 42
G2 patch files: 626
|
Posted: Mon Jan 22, 2007 10:36 am Post subject:
|
|
|
qfingers wrote: | It may be as simple as that. |
Yes it was
Seems to work very well, I'll see if I can figure out why the velocity sens is not working and post back.
Thanks again for this, between the old NM and DX7, the G2 now has access to more patches than any other synth in history. |
|
Back to top
|
|
|
qfingers
Joined: Nov 16, 2006 Posts: 186 Location: Tucson, AZ
G2 patch files: 2
|
Posted: Mon Jan 22, 2007 10:50 am Post subject:
|
|
|
g2ian wrote: | qfingers wrote: | It may be as simple as that. |
Yes it was
Seems to work very well, I'll see if I can figure out why the velocity sens is not working and post back.
Thanks again for this, between the old NM and DX7, the G2 now has access to more patches than any other synth in history. |
So you looked at dx2g2.py. Pretty simple hey? I want to write a NordLeadX converter as well. It should be as simple. The major problem is the convertion table for the various parameters.
As far as the Velocity Sens, LFO, and PitchEG, those parameters need a convertion table. There are only 10 to 20 parameters, but alot of them have limited ranges under 10. I am also having problems with the LFO waveforms. I have .syx files tha crash setting the LFO waveforms. I searched the web but didn't find an accurate table of waveforms for the DX7. That will also have to be addressed.
q |
|
Back to top
|
|
|
cappy2112
Joined: Dec 24, 2004 Posts: 2465 Location: San Jose, California
Audio files: 2
G2 patch files: 1
|
Posted: Mon Jan 22, 2007 11:32 am Post subject:
|
|
|
g2ian wrote: | qfingers wrote: | It may be as simple as that. |
Yes it was
Seems to work very well, I'll see if I can figure out why the velocity sens is not working and post back.
Thanks again for this, between the old NM and DX7, the G2 now has access to more patches than any other synth in history. |
I'll start clearing some space on my hard drive for the G2DX patches now |
|
Back to top
|
|
|
ian-s
Joined: Apr 01, 2004 Posts: 2669 Location: Auckland, New Zealand
Audio files: 42
G2 patch files: 626
|
Posted: Mon Jan 22, 2007 12:04 pm Post subject:
|
|
|
qfingers wrote: | So you looked at dx2g2.py. Pretty simple hey? |
I can sort of follow it, python looks like a child of C so I'm almost at home.
All the operator level stuff is 1:1 (or should be) so a table should not be required.
The LFO and Pitch EG are not so important, apart from adding vibrato, they were used very rarely for a few lame sfx patches. To get it exactly right, Clavia would need to add a DX style, stand alone EG. The rates on these depend on relative levels so R1 = 95 is not a specific time, it depends on the difference between L1 and L2 (scaled by velocity). I'll try to post more info on the LFO later. |
|
Back to top
|
|
|
jksuperstar
Joined: Aug 20, 2004 Posts: 2503 Location: Denver
Audio files: 1
G2 patch files: 18
|
Posted: Mon Jan 22, 2007 1:12 pm Post subject:
|
|
|
g2ian wrote: |
Thanks again for this, between the old NM and DX7, the G2 now has access to more patches than any other synth in history. |
Very concisely put! I feel like He-Man now. "I have the power!!" hehe.
|
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24075 Location: The Netherlands, Enschede
Audio files: 277
G2 patch files: 320
|
Posted: Mon Jan 22, 2007 1:31 pm Post subject:
|
|
|
3phase wrote: | here another fix of Jan´s patch...
They bugs are allready fixed...but it didnt sound wright...
but than i found what is wrong...
Blue mixers have to stay blue...otherwise your feedback point gets lower ! |
Hey, that's bright thinking man !! And it's really really close now.
Probably that means that the perc osc problem re pitch modulation is solved now as well - the signal should just be made blue ... should have thought of that myself ... _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
3phase
Joined: Jul 27, 2004 Posts: 1183 Location: Berlin
Audio files: 13
G2 patch files: 141
|
Posted: Mon Jan 22, 2007 1:54 pm Post subject:
|
|
|
There is a sonic missmatch in certain situations that shouldnt be there...
But the majortity of diffrences in the the sound of the 2 machines are related to parameter settings..when you get them wright booth machines sound the same..exept in certain conditions...but maybe this certain conditions can aswell be explained by para meter mismatchings...
but i dont think so... I ve somekind of instinkt for bugs...and as i wrote here before..as more i do 1:1 listenings tests as more this special g2 behaviour sounds like an error to me... Its just difficult to point at..its more a structural thing than something that is related to a single module..it builds up on bigger patches...at least i got the impression...
As better the converter gets, as closer the conversions become as more you can look into the details... |
|
Back to top
|
|
|
ian-s
Joined: Apr 01, 2004 Posts: 2669 Location: Auckland, New Zealand
Audio files: 42
G2 patch files: 626
|
Posted: Mon Jan 22, 2007 2:22 pm Post subject:
|
|
|
So on an operator level, what is working 100% is the envelope rates and levels, KBlevScale including Output level and AMod.
So that just leaves Coarse,Fine,Detune,Vel,RateScale,Ratio/Fixed which all are set to zero in all cases. I'm still not following the code enough to be able to fix it.
The LFO waves seem to be S+H, Triangle, Saw Down, Saw Up and Sine but thats just from the DXManager Editor software and I can't confirm that it is correct.
Edit: Spoke too soon, the KBLevScale is fine except for some scrambling of the LR curve type, here is a patch (1 variation only) that has been hand tweaked to match the original.
Description: |
|
Download |
Filename: |
magazineb1.pch2 |
Filesize: |
3.16 KB |
Downloaded: |
1791 Time(s) |
Last edited by ian-s on Mon Jan 22, 2007 2:49 pm; edited 1 time in total |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24075 Location: The Netherlands, Enschede
Audio files: 277
G2 patch files: 320
|
Posted: Mon Jan 22, 2007 2:46 pm Post subject:
|
|
|
g2ian wrote: |
The LFO waves seem to be S+H, Triangle, Saw Down, Saw Up and Sine but thats just from the DXManager Editor software and I can't confirm that it is correct. |
http://www.maths.abdn.ac.uk/~bensondj/dx7/manuals/dx7-man.pdf looks like the original manual to me, but it's been decades since I last saw the real thing.
It mentions an aditional square wave form. _________________ 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: Mon Jan 22, 2007 3:04 pm Post subject:
|
|
|
g2ian wrote: | So on an operator level, what is working 100% is the envelope rates and levels, KBlevScale including Output level and AMod.
So that just leaves Coarse,Fine,Detune,Vel,RateScale,Ratio/Fixed which all are set to zero in all cases. I'm still not following the code enough to be able to fix it.
The LFO waves seem to be S+H, Triangle, Saw Down, Saw Up and Sine but thats just from the DXManager Editor software and I can't confirm that it is correct.
Edit: Spoke too soon, the KBLevScale is fine except for some scrambling of the LR curve type, here is a patch (1 variation only) that has been hand tweaked to match the original. |
I would need a little more to see the problem. A fixed patch is fine, but how does it relate to the original? I need to be able to match parameters. I would need the original patch to see what parameters it was that was incorrect. Plus I don't want to fix it for just one case, I need all cases.
As far as the LFO waves, I need to know what number stored in the syx file matches those types. I found a dx7-sysex description and used that for reference, but I'm not sure it's totally correct.
q |
|
Back to top
|
|
|
ian-s
Joined: Apr 01, 2004 Posts: 2669 Location: Auckland, New Zealand
Audio files: 42
G2 patch files: 626
|
|
Back to top
|
|
|
3phase
Joined: Jul 27, 2004 Posts: 1183 Location: Berlin
Audio files: 13
G2 patch files: 141
|
Posted: Mon Jan 22, 2007 4:39 pm Post subject:
|
|
|
so..you made me currious... i just hooked on the DX7...
but... i ve to find a freeware dx editor now..do such things still exsist for os x? i will start a search... |
|
Back to top
|
|
|
qfingers
Joined: Nov 16, 2006 Posts: 186 Location: Tucson, AZ
G2 patch files: 2
|
Posted: Mon Jan 22, 2007 6:57 pm Post subject:
|
|
|
g2ian wrote: | here is the original bank, variation 7 is the example |
I forgot to set several parameters for each operator. It's fixed and will be in the next release.
q |
|
Back to top
|
|
|
3phase
Joined: Jul 27, 2004 Posts: 1183 Location: Berlin
Audio files: 13
G2 patch files: 141
|
Posted: Mon Jan 22, 2007 7:18 pm Post subject:
|
|
|
the dx converter is setting wrong algorythms |
|
Back to top
|
|
|
qfingers
Joined: Nov 16, 2006 Posts: 186 Location: Tucson, AZ
G2 patch files: 2
|
|
Back to top
|
|
|
ian-s
Joined: Apr 01, 2004 Posts: 2669 Location: Auckland, New Zealand
Audio files: 42
G2 patch files: 626
|
Posted: Mon Jan 22, 2007 7:43 pm Post subject:
|
|
|
3phase wrote: | the dx converter is setting wrong algorythms |
They look spot on with DXManager software, the ones I checked anyway, so it is different with a real DX7?
Edit:
Also I did the Yamaha stepped glide in the CS80 emulation patch, just put a note quant on the pitch output, set notes=off normal notes=1 stepped. Adjust the glide time in patch settings as usual. Probably will sound different unless the poly voice allocation is the same.
Edit more:
Looks like glide mode is not part of the patch setting anyway so never mind Last edited by ian-s on Mon Jan 22, 2007 7:57 pm; edited 1 time in total |
|
Back to top
|
|
|
3phase
Joined: Jul 27, 2004 Posts: 1183 Location: Berlin
Audio files: 13
G2 patch files: 141
|
Posted: Mon Jan 22, 2007 7:52 pm Post subject:
|
|
|
I was wrong..algorythms are set correct but lots of the other parameters are not...
I just set them manually for the Brass1 the first sound in rom bank 1a
i had to add a velocity offset of 32..but sounds close now..just the modulations is missing |
|
Back to top
|
|
|
qfingers
Joined: Nov 16, 2006 Posts: 186 Location: Tucson, AZ
G2 patch files: 2
|
Posted: Mon Jan 22, 2007 7:58 pm Post subject:
|
|
|
3phase wrote: | I was wrong..algorythms are set correct but lots of the other parameters are not...
I just set them manually for the Brass1 the first sound in rom bank 1a
i had to add a velocity offset of 32..but sounds close now..just the modulations is missing |
Then the way they are stored in the G2 is different. The internal parameters are not the same number because I set them exactly to the same number in the syx file. More analysis is needed.
Edit: I did forget 5 or 6 parameters for each operator though.
q |
|
Back to top
|
|
|
ian-s
Joined: Apr 01, 2004 Posts: 2669 Location: Auckland, New Zealand
Audio files: 42
G2 patch files: 626
|
Posted: Mon Jan 22, 2007 8:02 pm Post subject:
|
|
|
3phase wrote: | i had to add a velocity offset of 32..but sounds close now.. |
May just be down to the G2 velocity curve, a lot of people have complained about it requiring too hard a touch, do the sounds match if you play the DX7 from the G2 keyboard or visa versa? |
|
Back to top
|
|
|
|