preenfm Forum
PreenFM => preenfm2 and preenfm3 => Topic started by: Xavier on May 22, 2020, 11:01:22 AM
-
2.12 Beta2
https://drive.google.com/file/d/1chpzhsbCt1kFxVuPfmdk3dedF1DcR0Yj
Contains the 3 fixes from Toltekradiation. Thanks :)
And the midi CC7 & CC10 implementation as suggested by SFM for volume and left/right balance.
The new regular firmware is now the overclocked version : 192 mhz instead of the 168Mhz recommended.
I never saw a stm32F405 with problems at 192mhz. No need for the "o" version anymore.
Please report if you see any instability.
Thanks for you help.
2.12 Beta1
https://drive.google.com/open?id=1FYsYKkh9P41uj_SkQGeggpk7EU1fPktR
Please test and post feedback.
What's working and what's not working : both are usefull.
Thanks a lot :)
1. Double the power of the Matrix
Each row now has 2 destinations.
So you can modulate 24 destinations instead of 12.
Big thanks to Toltekradiation (Patrice Vigouroux) for the following enhancements that i merged from his firmware fork.
2. Enveloppe loop
Modulator operator have now loopable enveloppe.
Set Release time to 0 and release level to 1.0.
3. Glide
2 more glide values (11 and 12) for slower glide
4. Step sequencer new usage
1 new destination for each step sequencer : the starting point
Interesting to use with Note1/Note2 sources.
Example :
Step SEQ1 : 0 / 0.5 / Your pattern 1 here (_85F_85F _85F_85F)
Redirect the Step seq to your destination of choice :
Matrix row : Seq1 3.00 Mix3 None
Now let's use midi note to select the Step.
Matrix row : Not1 / 1.00 / sq1S / None (or sq2S for more fun)
Set Note1 to : Flat / 0 / +ln8
Not really straightforwared but that's it.
Each note will trigger a different step and will result in a different sound.
5. Matrix new source
Use rndK to trigger a new random number [-1:1] each time a note is started (key pressed).
6. Filters rework
Merge last modifications from the filter code.
-
Current Pull Request with all the code modification is here :
https://github.com/Ixox/preenfm2/pull/4/files
The branch is here :
https://github.com/Ixox/preenfm2/tree/2.12Beta
Patrice, if you have time to take a look , that would be great ;)
-
double matrix power !
what a great idea, i have to add this in my version :D
i have reviewed the code modifications and tested the firmware : all seems ok so far
-
Each update since i got the preenfm2 a few months back has been like receiving a fantastic present. Thank you so much guys :) Stunning work, so glad to see it continue to evolve.
-
this just gets better and better
-
omg. this is exciting! thank you! :-*
-
Will update right away. Double matrix destination is amazing! Thanks a lot for the hard work.
-
Thanks for yet another great update!
-
Here is the standard zip of 2.12 Beta1 with all versions inside.
So he Eurorack version is included for those who can try (look for "_cv" in the bin name)
https://drive.google.com/open?id=1FYsYKkh9P41uj_SkQGeggpk7EU1fPktR
Thanks
-
I fear i found a bug :-\
when using the step seq start point, in some case, the matrix values are not well calculated.
i joined an example patch to test this :
in this one, i defined a keyboard zone start between C3 and G3, this zone is used to increase INDEX_ALL_MODULATION, so playing in this zone is clearly heard.
if i play some notes except G3 or F#3, then play G3 or F#3 (and only these two), the modulation is not always well set.
this happen in 2.12Beta versions (standard and overclocked), and in my version as well.
i don't understand yet, but continue to search...
(edit: previous patch used user waveform, replaced it)
-
I fear i found a bug :-\
Great that's what beta are for ! ;)
-
excellent stuff, I will have to give the step sequencer thing a test, if only because I still can't understand quite how it works.
:o
-
to summarize the step seq start thing : it expect 0 < val < 1 to start at step val*16
the typical use is with a keyboard source :
if you use note 1 scaling, set to flat/0/+linear (default), it will generate a 0 to 1 value across the keyboard.
With this as a source in the matrix, seq start as destination, the step sequencer will then start at step 1 for leftmost notes of the keyboard, and start at step 16 for rightmost note of the keyboard.
if you set the step seq speed to zero bpm, it's no more a sequencer but it's used as bank of value only.
This way, you have 16 usable keyboard zones : just use the step seq as source in the matrix.
i hope it's clearer.
-
thanks, I think I get that part now (I hope!)
As I've only had the Preen a little while and not really played with the step sequencer I am a bit unclear on how much change to a sound is possible per step, I probably just need to have a try and load up your drumkit patch to see what's happening in it!
-
beware ! the drumkit patch will only work with the fw version available in the corresponding thread, the destination list is not the same in the 2.12Beta. otherwise, you will have to fix the seq start destination in the drumkit patch.
-
As I've only had the Preen a little while and not really played with the step sequencer I am a bit unclear on how much change to a sound is possible per step, I probably just need to have a try and load up your drumkit patch to see what's happening in it!
It can ve very usefull for rythmic part, sync it to midi clock (bpm after 240) and give it a try.
Use the matrix destination=IM* to get what it does.
And play with the StepSeq Gate from the menu.
Toltekradiation's patch allow to assign a key per step instead of having the step seq moving forward alone over time.
-
ok, i found the "bug" cause : the gate value of the step seq !
in my test patch, gate value was 0.8, so 20% of the rightmost notes of the keyboard zone could not work :P.
in addition, the result was not consistant because of a missing reset :
https://github.com/pvig/preenfm2/commit/c0482e639ec8d89419dd0c73e34c76622fa095b6
-
beware ! the drumkit patch will only work with the fw version available in the corresponding thread, the destination list is not the same in the 2.12Beta. otherwise, you will have to fix the seq start destination in the drumkit patch.
ah! I wondered what was going on, thanks.
:)
-
Haven't used my preenfm2 in almost 3 months. Plugged her in again last night and been messing with it again last night and now. Updated to new firmware! So far so good!
-
mmm, another one :
the attack modifications from a stepseq source are not always calculated early enough : one have to press the key again to get the right value..
-
ok, the fix was close to previous one :)
here is the commit : https://github.com/pvig/preenfm2/commit/b13f0cf9a5d3c0e87f31c3711e362e4901e7e461
-
Wow, this sounds great! Thank you for improving the PreenFM2!
Will flash it now and play around. The added Dst2 in the Matrix is an immediate big-like!
-
i found another bug :
only with algo 28, when env loop is activated on Op6, notes are no longer started...
(edit) bug fix here : https://github.com/pvig/preenfm2/commit/9f452a282170c1f4e9fabe5512dba92d73327734
-
Is there a release with these fixes?
-
Hello,
I use the PreenFM2 quite a lot for my music. A real annoyance is that it does not have channel-volume and channel-pan. (At least not on the standard-CCs) So I have to use filter-gain or the CCs for the mixer to achieve the desired changes. While this is nice if you want to alter the sound and stereo-seperation completely, this is totally annoying if you just want a fade-in and/or fade-out.
So, please, make these midi-CCs (Channel-Volume/-Pan) work, before adding more filters or other more CPU-intense but seldomly used things.
all the best and TIA!
Stefan
-
@Drexy No release with the fix yet... I don't have any excuse except the lack of time.
Hello,
I use the PreenFM2 quite a lot for my music. A real annoyance is that it does not have channel-volume and channel-pan. (At least not on the standard-CCs) So I have to use filter-gain or the CCs for the mixer to achieve the desired changes. While this is nice if you want to alter the sound and stereo-seperation completely, this is totally annoying if you just want a fade-in and/or fade-out.
So, please, make these midi-CCs (Channel-Volume/-Pan) work, before adding more filters or other more CPU-intense but seldomly used things.
all the best and TIA!
Stefan
I'm not sure i get it.
Why doesn't the MixerGain (CC73) work for fade-in and fade-out ? It's available even when no filter is selected.
And can you clarify what you mean or want to achieve with "stereo-separation" ?
-
Why doesn't the MixerGain (CC73) work for fade-in and fade-out ? It's available even when no filter is selected.
According to this List: https://ixox.fr/preenfm2/manual/midi/ CC73 is "filter-gain" and not "mixer"-gain. ;) This is why I didn't even think it would work without a filter being set.
Now, that I have tried it, it seems very well suited for the job. Only weird thing remaining in that context is, that it is on CC73 (just curious, was there any particular reason for this?) and not on CC7 (which would be Midi-Channel-Volume).
According to that same Midi-CC-List for the Preen (is it complete?) CC7 is not used. If so, would it be possible to double CC73 on CC7? So, that the PreenFM2 would react on both? CC73 for backwards-compatibility and CC7 for all sorts of "Standard"-Midi-Sequencer-Software? CC7 for Channel-Volume is pretty much an agreed standard...
And can you clarify what you mean or want to achieve with "stereo-separation" ?
I will try to...
You can place the output of the four mixer-channels of a given PreenFM2-Voice in the stereo-position separately -- which is a very cool feature, btw -- like you can do with the separate audio-levels of these. This is (according to the list mentioned above) controlled for the pan-position of each mixer-output with CCs (23 + (n-1) *2), and for the volume of each mixer-output with CCs (22 + (n-1) *2). So CC22 would control Mix1-Volume and CC23 would control Mix1-Pan-Position... This text is just an attempt to clarify of what I am speaking about (you will know these CCs by yourself for sure... ;) )
What I would like to have additionally would be some means of a "global" Pan-Position (Standard-CC for this would be CC10).
So what would be the desired behaviour of CC10? Imagine CC23 (pan MIX1) is set to totally left and CC25 (MIX2) is set to totally right. CC10 then would be reacting as follows: CC10 set to left: MIX1 and MIX2 mapped completely to the left position. CC10 set to right: MIX1 and MIX2 mapped completely to the right position. CC10 set to center: MIX1 left, MIX2 right as set by CC23 and CC25.
Why do I bother and ask?
I use the preenFM2 mainly in a DAW-Sequencer-Setup in which I use all four of its midi-channels with different instruments. Most DAWs/Midi-Sequencers provide Channel-Volume and Channel-Panning on CCs Nr. 7 and Nr. 10 inside of some sort of channel-mixer in that scenario. And "all" (despite the Waldorf Rocket ... which is monophonic and doesn't feature midi-controlled volume at all...) my synths feature them on these CCs. Except the PreenFM2.
all the best,
Stefan
-
Thanks, that makes total sense.
CC7 would copy CC73 i agree, that's easy to do.
Keep in mind that's a digital volume, when decreasing it reduces the number of bits used by the DAC. So keep it at 1.0 for regular use.
For CC10, I see, that's the reason i created the "Mix" Filter (that's the first one) and its "Pan" parameter, but when using it you cannot use any other effect.
It would have to be a feature available through CC only as i don't find any nice place to put that in the already crowded UI.
-
For CC10, I see, that's the reason i created the "Mix" Filter (that's the first one) and its "Pan" parameter, but when using it you cannot use any other effect.
It would have to be a feature available through CC only as i don't find any nice place to put that in the already crowded UI.
Having this available through CC, only, would be -- at least from my point of view -- absolutely perfectly OK.
I must admit, that I do not know of any synth which features more than this (Channel-Pan only via CC10). I may be wrong but I believe that there is pretty much no synth out there having it directly accessible from within the UI...
-
2.12 Beta 2
https://drive.google.com/file/d/1chpzhsbCt1kFxVuPfmdk3dedF1DcR0Yj/view?usp=sharing
See first message of this thread for details.
SMF can you give a try to firmware 2.12 B2 i just uploaded. You posted the good day when i have some free time :)
Toltekradiation, Beta 2 should contain your 3 fixes. Thanks :)
-
Xavier,
I tried 2.12 Beta 2 and CC7/CC10 do work like a treat! Very, very nice! :)
But it seems like I stumbled across a bug when selecting banks/patches in the DX7 section. If I do accidently select patch Nr. 33 then the PreenFM2 freezes on playing the current waveform and needs to be restarted for recovery (my sequencer starts numbering patches at Nr. 1, so this is exactly one patch above the allowed range for DX-7 banks -- just for clarification. This looks like a missing range-check leading to out-of-bounds memory-access at a first glance.).
With PreenFM2-banks, however, this doesn't happen.
-
But it seems like I stumbled across a bug when selecting banks/patches in the DX7 section. If I do accidently select patch Nr. 33 then the PreenFM2 freezes on playing the current waveform and needs to be restarted for recovery (my sequencer starts numbering patches at Nr. 1, so this is exactly one patch above the allowed range for DX-7 banks -- just for clarification. This looks like a missing range-check leading to out-of-bounds memory-access at a first glance.).
Nice catch.
Thanks, it's fixed for next version.
-
I guess an updated Editor that reflects these change will come when out of beta? because this won't work with the latest editor i reckon?
-
I guess an updated Editor that reflects these change will come when out of beta? because this won't work with the latest editor i reckon?
Correct.
Unless i miss something, the only modification needed by the editor is the second matrix destiation.
I have to modify the editor before releasing an official 2.12.
But don't hold your breath ;)
-
I guess an updated Editor that reflects these change will come when out of beta? because this won't work with the latest editor i reckon?
Correct.
Unless i miss something, the only modification needed by the editor is the second matrix destiation.
I have to modify the editor before releasing an official 2.12.
But don't hold your breath ;)
Ok.. pretty happy with it as it is, though wouldn't mind having the update :D