Brendan, this is perfect! Dear and explanatory as always! I think I then realized that maybe it is better to use it without quantization ... but now that I know well the operation knowingly exploit it to the fullest! Thanks again, I hope there will soon be more news!
I'm not really planning on any big changes for a few months, a few bugfixes and UI streamlining maybe, and get a few more explanatory videos out. Mostly need to finish my PhD
Hello Brendan, I noticed something ... I do not know if it depends on the custom firmware or is a bug from the original firmware, but when loading samples no symbol appears on the display to indicate that charging is terminated or been successful. ..sarebbe very useful to see something! )
Hello Brendan, I noticed something ... I do not know if it depends on the custom firmware or is a bug from the original firmware, but when loading samples no symbol appears on the display to indicate that charging is terminated or been successful...are very useful to see something!
Hi Brendan, first of all, your firmware is amazing! It really takes the lxr to a whole new level! But i've noticed a bug that crashes my lxr everytime: as soon as there is modwheel input on channel 1, the lxr says goodbye. I've tried it with different midi settings and even assigned a morph target, but the problem stays the same... It would be great if you could fix that bug (:
thanks for letting me know! Yeah, I've crashed it a lot during testing by sending too much morph data, I'll likely have to put in a limit where it can only morph once per step or something. You mention modwheel input "on channel 1". Does it not crash if the LXR is assigned to other channels? Is that your global channel? What are the individual channels assigned to? (Voice channels morph individually with mod wheel, and the global channel morphs all, but I did combine them for cases where multiple channels are stacked up)
Maybe my description was a bit misleading... The lxr crashes only when i send modwheel data (from my blofeld keyboard) on channel 1. If i set the blofeld to 2-16 the lxr keeps working.
Edit: I did some more testing and it seems to be a problem with the individual voice midi channel settings. If one or more of them are set to "0" the problem occurs. Global midi settings seem to have no influence on this.
Edit 2: I can also crash it when the voices are on the same channel (e.g. 16) and i send on ch 16. Maybe i'm experiencing the same issue as you and there's just too much morph data... Would be nice if i could switch off external morph input in the settings.
ok, I'll look into this - there should certainly be a limit to how many external morph commands the LXR will process at a time, but it might take a little while to get the balance right. It's going to be a little while before I can work on this...
Just started playing with Morph for the first time :P
I've noticed something weird: I tried "turning off" Morph for Voice 1 by going through every parameter in Voice mode and using your shortcut trick of pressing the encoder and then shift to see the Morph destination value, and then I would set this to be the same as the current voice parameter value. The idea is then that when Morph is engaged, everything morphs except Voice 1. This seems to work great for low Morph values, but when I get to higher values my Voice 1 (bass drum) starts to sound like sample rate reduction is kicking in, extremely so for very high Morph values.
I checked if maybe some other voice was modulating Voice 1 and maybe that other voices LFO was Morphing, but nope, it definitely has something to do with the sample rate reduction. With the global sample rate fixed to 127, if I crank the Morph but quickly tweak the sample rate slightly down and then back to 127, everything sounds great, but as soon as I tweak Moprh again my Voice 1 gets some sample rate reduction
I'm using your firmware Brendan (why would you not??) so I thought I'd post here. Any help would be greatly appreciated!
I've had my LXR for a bit more than a year and just stumbled across this which I just got to try out!
I also have feature request regarding the MIDI routing which I can't seem to get around in the original SW.
I'd love to sync my LXR from a MIDI source, send the clock over USB to my iPad, send CCs and notes back from my iPad to both the LXR and the the MIDI out port. Are you following?
Just started playing with Morph for the first time :P
I've noticed something weird: I tried "turning off" Morph for Voice 1 by going through every parameter in Voice mode and using your shortcut trick of pressing the encoder and then shift to see the Morph destination value, and then I would set this to be the same as the current voice parameter value. The idea is then that when Morph is engaged, everything morphs except Voice 1. This seems to work great for low Morph values, but when I get to higher values my Voice 1 (bass drum) starts to sound like sample rate reduction is kicking in, extremely so for very high Morph values.
I checked if maybe some other voice was modulating Voice 1 and maybe that other voices LFO was Morphing, but nope, it definitely has something to do with the sample rate reduction. With the global sample rate fixed to 127, if I crank the Morph but quickly tweak the sample rate slightly down and then back to 127, everything sounds great, but as soon as I tweak Moprh again my Voice 1 gets some sample rate reduction
I'm using your firmware Brendan (why would you not??) so I thought I'd post here. Any help would be greatly appreciated!
P.S. Good luck with your thesis!
No one using Brendan's firmware can confirm this behaviour? Can anyone refute it?
I'd love to sync my LXR from a MIDI source, send the clock over USB to my iPad, send CCs and notes back from my iPad to both the LXR and the the MIDI out port.
plus 1.
I'd be great if the LXR could function as some sort of usb host to at least sync usb-only synths/hardware (like the op-1). it's a pain in the ass if you're short on money, decide to buy such a device and have to spent another 100 euro or more for a "professional" solution. (there's a 60€ diy solution, thou). the connectivity isn't the problem, cables or adapters are round about 10 euro or so (connecting a mini usb to the usb hub socket form the lxr). is that even possible to implement in your custom firmware, brendan?
@marek_messer the LXR can't be a USB host as it's got the wrong type of USB port, this is not something that can be solved with firmware. Best to get a separate host for the OP-1.
or did I miss the point and it's because the LXR has this kind of socket USB port? there would be/should be a adapter for that, too, right? See I don't get this different USB port types/host stuff, so excuse me if I'm being wrong and dumb ; )
You're not being dumb, it's just confusing. Basically- USB devices always need a host to translate the messages, and neither the OP-1 or the LXR are set up to be one. Without a host, it's a bit like trying to connect a computer's keyboard directly to the mouse. However you can build a USB host fairly easily- here's a link that might help.
ok I did know that : ) that's why I thought the LXR could work as a host if it's possible due to changes in Brendan's firmware. I didn't get that the hardware is the problem. but thanks for the very good link, comes in handy. unfortunately I don't have an ardurio programmer so I am going for the raspberry pi solution.
Hello Brendan! Just upgraded and I really love it, lots of great new features. I have a problem! When instant pattern switching I randomly get tracks within patterns playing asynchronously. Stopping and starting the sequencer re-alignes then. I'm using an external clock. The 'press pattern when in perf mode' feature doesn't fix it, only start/stop, and it happens randomly! Thanks again for a great OS update Ben
Hey all! Sorry I haven't checked in for a while - as I've mentioned, it's going to be crunch time for my PhD for the next few months, but be assured I will come back to the LXR when I can.
I have tried to correct persistent and annoying bug where the LXR would miss a lot of channel-specific MIDI messages when synced to MIDI clock from the MIDI input jack (syncing from USB did not have this problem). If you have been having trouble receiving notes or CC parameter while synced to a sequencer from the MIDI jack, this may fix it for you.
To address some of the things mentioned above:
Hi, is there a manual for your firmware version? Thanks
There isn't a manual, but I track the changes I've made at brendanclarke.com/files/LXR in a readme. You can consider it a supplemental for the main LXR manual, which mostly still applies to the firmware.
I'd be great if the LXR could function as some sort of usb host to at least sync usb-only synths/hardware (like the op-1). (connecting a mini usb to the usb hub socket form the lxr). is that even possible to implement in your custom firmware, brendan?
I don't want to confuse people, but this may not be impossible, just very very difficult. As @borfo (IIRC) has mentioned, MIDIbox has an implementation for the same processor (STM32F4) and USB configuration that DOES have USB hosting implemented. I have no idea how to go about actually porting that to the LXR. It may be a huge job, it may require modifying the bootloader. I just don't know, but if someone wants to dive down that rabbithole, by all means....
When instant pattern switching I randomly get tracks within patterns playing asynchronously....
Hmm. do you have any pattern scaling or length set? Keep in mind that scaling and length is per pattern. It's strange - double tapping the pattern button doesn't re-align the pattern? are you doing any pattern chaining or pattern rotation?
I'll look into @Donderdag's bug also. I'd like to smooth out a couple of ways morph works. It's a lot of data to deal with - could be done better than it is right now.
I've noticed something weird: I tried "turning off" Morph for Voice 1 ... definitely has something to do with the sample rate reduction.
so... the global sample rate reduction is a separate parameter from the individual voice sample rates, and when morph is applied, it morphs all of them. If you want to 'zero out' your morph target and only have a few parameters that morph, the better way to do this is to save the whole drum kit, then load that as the morph target. You will then have the same sample rate reduction for both the kit and the morph, and the kit will sound the same regardless of morph value.
This can get tedious, but keep in mind that in the current firmware, 'performances' and 'all' also store morph data, so that's a better place to keep things if you have a morph arrangement you really like!
Can confirm that external midi clock now works perfectly from troublesome devices before (Midibox). I'm also pretty sure the click on deep drums has gone as well. Well done mate, awesome work.
Great to hear it! Strange that this change would result in a (lack of) audio effect. I basically moved where realtime messages get processed further up the chain of functions that deal with MIDI. I *think* what was happening was that the LXR was expecting data bytes after realtimes, which never arrived. I wonder if spamming huge amounts of invalid CC data would produce an audio effect as well...
But maybe best let sleeping dogs lie for now
At the moment, I'm just finishing up some testing on being able to load individual tracks from pattern/perf/all files.
Hi Brendan, thanks for getting back. Firstly your OS is absolutely killer. It's really nice for playing live, especially rolls, loops and macros. So, I still get the asynchronous tracks when instant patten switching, but I CAN clear them now that I turned follow on!? It happens in both slave and master mode. The silly thing is I now quite like it! You get the strangest versions of your patterns, then you realign them with the appropriate pattern button. Cheers!
So, I still get the asynchronous tracks when instant patten switching, but I CAN clear them now that I turned follow on!?
aha! yes, thanks for bringing that up. I almost always leave follow on, it's not surprising that realign doesn't work with it off. That's definitely a bug, though, and I'll get that fixed.
If you want to test something out for me, turn follow back off and try to 'double-view' (shift-tap-tap the pattern button in perf mode) instead. This would be a little wiggy, definitely not the way I'd want it to work, but if it does/doesn't work, it would remind me where in the code things are lurking!
I've noticed something weird: I tried "turning off" Morph for Voice 1 ... definitely has something to do with the sample rate reduction.
so... the global sample rate reduction is a separate parameter from the individual voice sample rates, and when morph is applied, it morphs all of them. If you want to 'zero out' your morph target and only have a few parameters that morph, the better way to do this is to save the whole drum kit, then load that as the morph target. You will then have the same sample rate reduction for both the kit and the morph, and the kit will sound the same regardless of morph value.
This can get tedious, but keep in mind that in the current firmware, 'performances' and 'all' also store morph data, so that's a better place to keep things if you have a morph arrangement you really like!
Hope that helps!
This isn't nearly as tedious as the way I was trying to do it, and this procedure actually works, so I've adopted it. Thanks Brendan!
Good to hear! I mean setting the global sample rate for the morph patch isn't optimal - it would be nice if you could shift-change it like for other sound parameters. Small thing, but I'd like to put that in eventually, as a special case. As of now the 'perf' shift function overrules doing that.
Would it be possible to add velocity randomization? Something where you could select a percentage of randomization. Higher value = larger difference from the default value. Would speed up in humanizing percussion sequences and such.
Note randomization would be cool too, but I'm not sure how realistic that one might be.
My ZAQuencer (sequencer firmware for a bcr2000) has pattern fill functions very useful like fill steps by waves&math sequence of notes (or mute or velocity and other parameters) like sin, saw, square waves or random.. all with range parameters min/max... I think that are developed by not so long math code..and the code I think (I hope that) is opensource...
Would it be possible to add velocity randomization?
You can do this now! Just use one of the LFO's with a random waveform and modulate volume or whatevs. Pitch might be a bit trickier - I think if you modulate pitch, it doesn't end up quantized, but you could, if you wanted, put in some automation-only (velocity 0) notes on other tracks with randomness and a square LFO set to a harmonic interval. That would give you some random lower-pitched notes. I would love to, if possible, add a second LFO destination, there's a huge number of things you can do with them.
Comments
Dear and explanatory as always!
I think I then realized that maybe it is better to use it without quantization ... but now that I know well the operation knowingly exploit it to the fullest! Thanks again, I hope there will soon be more news!
I'm not really planning on any big changes for a few months, a few bugfixes and UI streamlining maybe, and get a few more explanatory videos out. Mostly need to finish my PhD
first of all, your firmware is amazing! It really takes the lxr to a whole new level!
But i've noticed a bug that crashes my lxr everytime: as soon as there is modwheel input on channel 1, the lxr says goodbye. I've tried it with different midi settings and even assigned a morph target, but the problem stays the same...
It would be great if you could fix that bug (:
The lxr crashes only when i send modwheel data (from my blofeld keyboard) on channel 1. If i set the blofeld to 2-16 the lxr keeps working.
Edit: I did some more testing and it seems to be a problem with the individual voice midi channel settings. If one or more of them are set to "0" the problem occurs. Global midi settings seem to have no influence on this.
Edit 2: I can also crash it when the voices are on the same channel (e.g. 16) and i send on ch 16.
Maybe i'm experiencing the same issue as you and there's just too much morph data...
Would be nice if i could switch off external morph input in the settings.
I've noticed something weird: I tried "turning off" Morph for Voice 1 by going through every parameter in Voice mode and using your shortcut trick of pressing the encoder and then shift to see the Morph destination value, and then I would set this to be the same as the current voice parameter value. The idea is then that when Morph is engaged, everything morphs except Voice 1. This seems to work great for low Morph values, but when I get to higher values my Voice 1 (bass drum) starts to sound like sample rate reduction is kicking in, extremely so for very high Morph values.
I checked if maybe some other voice was modulating Voice 1 and maybe that other voices LFO was Morphing, but nope, it definitely has something to do with the sample rate reduction. With the global sample rate fixed to 127, if I crank the Morph but quickly tweak the sample rate slightly down and then back to 127, everything sounds great, but as soon as I tweak Moprh again my Voice 1 gets some sample rate reduction
I'm using your firmware Brendan (why would you not??) so I thought I'd post here. Any help would be greatly appreciated!
P.S. Good luck with your thesis!
I'd be great if the LXR could function as some sort of usb host to at least sync usb-only synths/hardware (like the op-1). it's a pain in the ass if you're short on money, decide to buy such a device and have to spent another 100 euro or more for a "professional" solution. (there's a 60€ diy solution, thou). the connectivity isn't the problem, cables or adapters are round about 10 euro or so (connecting a mini usb to the usb hub socket form the lxr).
is that even possible to implement in your custom firmware, brendan?
See I don't get this different USB port types/host stuff, so excuse me if I'm being wrong and dumb ; )
but thanks for the very good link, comes in handy. unfortunately I don't have an ardurio programmer so I am going for the raspberry pi solution.
I have a problem! When instant pattern switching I randomly get tracks within patterns playing asynchronously. Stopping and starting the sequencer re-alignes then. I'm using an external clock. The 'press pattern when in perf mode' feature doesn't fix it, only start/stop, and it happens randomly!
Thanks again for a great OS update
Ben
I have tried to correct persistent and annoying bug where the LXR would miss a lot of channel-specific MIDI messages when synced to MIDI clock from the MIDI input jack (syncing from USB did not have this problem). If you have been having trouble receiving notes or CC parameter while synced to a sequencer from the MIDI jack, this may fix it for you.
To address some of the things mentioned above: There isn't a manual, but I track the changes I've made at brendanclarke.com/files/LXR in a readme. You can consider it a supplemental for the main LXR manual, which mostly still applies to the firmware. I don't want to confuse people, but this may not be impossible, just very very difficult. As @borfo (IIRC) has mentioned, MIDIbox has an implementation for the same processor (STM32F4) and USB configuration that DOES have USB hosting implemented. I have no idea how to go about actually porting that to the LXR. It may be a huge job, it may require modifying the bootloader. I just don't know, but if someone wants to dive down that rabbithole, by all means....
I'll look into @Donderdag's bug also. I'd like to smooth out a couple of ways morph works. It's a lot of data to deal with - could be done better than it is right now.
so... the global sample rate reduction is a separate parameter from the individual voice sample rates, and when morph is applied, it morphs all of them. If you want to 'zero out' your morph target and only have a few parameters that morph, the better way to do this is to save the whole drum kit, then load that as the morph target. You will then have the same sample rate reduction for both the kit and the morph, and the kit will sound the same regardless of morph value.
This can get tedious, but keep in mind that in the current firmware, 'performances' and 'all' also store morph data, so that's a better place to keep things if you have a morph arrangement you really like!
Hope that helps!
But maybe best let sleeping dogs lie for now
At the moment, I'm just finishing up some testing on being able to load individual tracks from pattern/perf/all files.
So, I still get the asynchronous tracks when instant patten switching, but I CAN clear them now that I turned follow on!? It happens in both slave and master mode.
The silly thing is I now quite like it! You get the strangest versions of your patterns, then you realign them with the appropriate pattern button.
Cheers!
aha! yes, thanks for bringing that up. I almost always leave follow on, it's not surprising that realign doesn't work with it off. That's definitely a bug, though, and I'll get that fixed.
If you want to test something out for me, turn follow back off and try to 'double-view' (shift-tap-tap the pattern button in perf mode) instead. This would be a little wiggy, definitely not the way I'd want it to work, but if it does/doesn't work, it would remind me where in the code things are lurking!
I mean setting the global sample rate for the morph patch isn't optimal - it would be nice if you could shift-change it like for other sound parameters. Small thing, but I'd like to put that in eventually, as a special case. As of now the 'perf' shift function overrules doing that.
Note randomization would be cool too, but I'm not sure how realistic that one might be.
I think that are developed by not so long math code..and the code I think (I hope that) is opensource...