KVR :: u-he » zebra mseg not keeping time [View Original Topic]
There are 30 posts in this topic.
iamyourfather - Thu Apr 05, 2012 9:53 am
Just started experimenting with the MSEG and quite dismayed to find that the timing drifts all over the place... TBH it's not a little off it's not even close.
Any ideas anyone please ??
iamyourfather - Thu Apr 05, 2012 9:59 am
Also while using the MSEG, the buttons on the rest of the synth strop working correctly, the seem to resist moving from where they are and spring back when you try to change them.
Bronto Scorpio - Thu Apr 05, 2012 9:59 am
Are the Attack, Loop and Release knobs of your MSEG set to 0?
Cheers
Dennis
Bronto Scorpio - Thu Apr 05, 2012 10:01 am
iamyourfather wrote:
Also while using the MSEG, the buttons on the rest of the synth strop working correctly, the seem to resist moving from where they are and spring back when you try to change them.
That sounds like an accidental midi learn.
Cheers
Dennis
iamyourfather - Thu Apr 05, 2012 10:28 am
OK the weird button action has gone since I reloaded the patch so it may well have been as you said, but sadly the MSEG still will not keep time. I have checked the attack and release and loop buttons and the are dead on zero or 1.0. I really think this is a bug.
iamyourfather - Thu Apr 05, 2012 10:32 am
Ya know I had a similar timing issue with the ARP a while back, I have a weird feeling this is a reaper issue. It may well be time to invest in a proper DAW like ableton... hmmm
justin3am - Thu Apr 05, 2012 10:37 am
iamyourfather wrote:
Ya know I had a similar timing issue with the ARP a while back, I have a weird feeling this is a reaper issue. It may well be time to invest in a proper DAW like ableton... hmmm
In Live and Numerology (the DAWs I have at hand) the MSEGs and Arp follow tempo perfectly. I'll try it in Reaper in just a moment.
iamyourfather - Thu Apr 05, 2012 10:40 am
Thanks a lot friend, would really appreciate that.
justin3am - Thu Apr 05, 2012 10:44 am
Yeah, it works fine in Reaper 4.2.1 as well. The Attack, Loop and Release parameters should all be set to Zero.
iamyourfather - Thu Apr 05, 2012 10:50 am
well I am stumped as I checked those multiple times just to makes sure. mebbe should try to make a basic patch just to check it more thoroughly.
Thanks again
iamyourfather - Thu Apr 05, 2012 11:10 am
OK I think I have found something. For the original (faulty) patch I was modulating the bus 1 volume to produce some gated noise. I tried switching the MSEG to osc volume and lo and behold it keeps time. it seems that bus volume modulation is the problem
http://www.box.com/s/d919ac3b0f49cc653bde
Here's the patch, there should be 2 MSEG mods... bus volume and osc volume, try both. I found OSC works fine and BUS volume doesn't. Also, play 1 long sustained note over several bars to test.
iamyourfather - Thu Apr 05, 2012 11:13 am
ignore that, now the osc volume is all out of time too, I am literally tearing my hair out over this lol
iamyourfather - Thu Apr 05, 2012 11:26 am
Ok tested the MSEG in alchemy and it is working fine.
xh3rv - Thu Apr 05, 2012 11:40 am
iamyourfather wrote:
ignore that, now the osc volume is all out of time too, I am literally tearing my hair out over this lol
A tricky technical point:
The OSC volume mod is additive. It's the volume knob plus the value of the MSEG.
The bus volume mod is a multiply rather than addition. The current envelope is scaled by the mod value rather than summed with it. With a negative mod values, it's inverting the mod source rather than subtracting.
In the patch you posted with both active the end result of these modulations on top of each other is definitely a little unsettled. Things sound more in sync to me with 0 bus modulation, 0 osc volume, and MSEG1 for osc volume modulation - essentially gating.
Hope this helps
iamyourfather - Thu Apr 05, 2012 11:44 am
they are not supposed to be active simultaneously, they were supposed to be for comparison of modulating each individually. I've since found both go out of time anyway. I checked with my alchemy MSEG and it works fine, so I am a bit stumped ATM.
Thx a lot for taking the time to look at it
xh3rv - Thu Apr 05, 2012 12:01 pm
iamyourfather wrote:
I've since found both go out of time anyway.
Turning off the LFO->tuning modulation made a perceptual difference to me as well, but it sounds like something else must be going on though. Just with what's in the patch, it does sync here and stays in sync for several minutes, which should rule out some things, I guess.
justin3am - Thu Apr 05, 2012 12:14 pm
Sounds more and more like accidental midi-learn.
Do this...
Urs wrote:
Open the file
..Vstplugins\Zebra2.data\Support\com.u-he.Zebra2.midimap.txt
in Wordpad and maybe erase any entries, or post the content here.
..Vstplugins\Zebra2.data\Support\com.u-he.Zebra2.midimap.txt
in Wordpad and maybe erase any entries, or post the content here.
ho66it - Thu Apr 05, 2012 1:21 pm
@iamyourfather
If you have a SoundCloud account, you can get Ableton Live 8 Lite for free now. At least you could test your patch in Ableton using the Lite version.
http://www.ableton.com/get-free-live-lite?sn=57FE-CC4A-D1DC-8EC9-896F-7925&utm_source=sc-free-lite-mailing&utm_medium=email&utm_term=sc-free-lite-mailing-b&utm_campaign=free-lite-for-sc
[/url]
iamyourfather - Thu Apr 05, 2012 10:13 pm
justin3am wrote:
Sounds more and more like accidental midi-learn.
Do this...
Do this...
Urs wrote:
Open the file
..Vstplugins\Zebra2.data\Support\com.u-he.Zebra2.midimap.txt
in Wordpad and maybe erase any entries, or post the content here.
..Vstplugins\Zebra2.data\Support\com.u-he.Zebra2.midimap.txt
in Wordpad and maybe erase any entries, or post the content here.
// Global MidiCC Map
1=PCore:X_1 //Core:X1
69=XMF1:Cut //XMF1:Cutoff
71=XMF1:Res //XMF1:Resonance
72=XMF1:XFM //XMF1:FilterFM
73=XMF1:FOff //XMF1:FreqOffset
84=OSC1:WNum //Oscillator1:WaveWarp
I thought I'd let someone have a peep before I just go on and erase it
iamyourfather - Fri Apr 06, 2012 2:18 pm
OK not sure what's going on. I deleted the midimap data and the patch still won't keep time, however a new test patch I made does keep time as does the HS Flapper preset from the factory bank. I guess I can just go on and remake what I originally wanted to do as it seems ok. Not sure how any of the data listed in that midimap .txt would affect the timing tho... if at all.
Still thanks for your help chaps.
bmrzycki - Fri Apr 06, 2012 4:02 pm
iamyourfather wrote:
OK not sure what's going on. I deleted the midimap data and the patch still won't keep time, however a new test patch I made does keep time as does the HS Flapper preset from the factory bank. I guess I can just go on and remake what I originally wanted to do as it seems ok. Not sure how any of the data listed in that midimap .txt would affect the timing tho... if at all.
Still thanks for your help chaps.
You could double click all the mseg source and destination knobs to reset them to zero. It's possible you used shift+mouse to add a tiny bit to one of the knobs and that will add up over time. It's also really hard to see by just looking at the little knob arc. Still thanks for your help chaps.
iamyourfather - Sat Apr 07, 2012 3:01 am
Sadly the issue is not fixed at all. I just loaded up the SM LFOed XY preset from the factory bank and it was all over the place. It simply is NOT an issue of error in my patch creation... I really wish it was.
Could Urs please take a look at this as it's really cramping my style. Thanks.
IAYF
Urs - Sat Apr 07, 2012 5:17 am
How long do you hold the notes?
The MSEGs may run out of sync after 30 seconds or so, and mostly if Swing is used. I've tried to wrap my head around this, but the only possible cure is to not Swing the MSEGs at all.
iamyourfather - Sat Apr 07, 2012 8:05 am
There's no swing applied and the longest notes held were maybe 8 beats at 145 bpm. It doesn't so much drift out of time so much as randomly start off each note in and out of time. It sounds more like the envelope start point is being applied with some random factor involved. I will see if I can get an mp3 together of an example to give a better picture.
Thx Urs
iamyourfather - Sat Apr 07, 2012 8:42 am
OK ... here's an mp3 showing the timing in the simplest possible form just a gated saw against a kick and hat. I used repeat notes of 8 beats in length, you can hear that the timing accuracy seems to come in discreet chunks suggesting that the start time of the mod envelope is hitting at a different time for each note. At least that is what it sounds like to me.
http://www.box.com/s/81b26c45e11c5068d681
Also here's the patch I used to generate the saw / gate
http://www.box.com/s/f8423a55444804486fae
Thanks, IAYF
iamyourfather - Sat Apr 07, 2012 8:49 am
BTW sorry the mp3 has some weird glitches in it, I recorded the file to wav in reaper first tho so it's nothing to do with the issue
justin3am - Sat Apr 07, 2012 10:23 am
the end point for MSEG1 in the patch you posted wasn't on beat. Try this...
http://www.3amnoise.net/noise_strobe-fix.h2p
iamyourfather - Sat Apr 07, 2012 11:47 am
good grief, how dumb do I feel right now. Dunno how it happened with presets but I can only assume that I touched something and did not realise. Thanks to all for sorting this out for me. I can only apologise for my error and you wasted time.
You guys are simply brilliant
justin3am - Sat Apr 07, 2012 7:05 pm
No problem, dude. Glad it's all worked out.
Urs - Sun Apr 08, 2012 2:14 am
Gah - I should have caught that myself
There are 30 posts in this topic.