IX Magic 1.19 - October 13th, 2024

[IXix]

No, the only change from one to the other was unplugging my laptop from my Audio Interface (USB) & changing Buzz's Audio Output from ASIO to MME.

As for Magic, I only added pages, sequences, steps & events. I didn't touch the resources page, iirc.

Thanks for the info, it all helps. This is very strange though. I don't think there's much chance of finding out what happened unless it happens again.

[Candle]

Well, fortunately it hasn't happened again. And I've backed-up my setups to XML so if it does I don't have to start from scratch. Again, that's an awesome feature. Magic is so awesome. It's allowed me to setup my new MIDI foot controller exactly the way I want. I'm essentially using it as a MIDI to Peer controller to control various Buzz Machines (like your PatchBay, for example). It's making my playing & recording methodology so much smoother.

See You In The Shadows…

[IXix]

It's making my playing & recording methodology so much smoother.

Glad to hear it!

[IXix]

2020.05.17 IX Magic 1.12d - x86
+ Dialog resizing improvements

Little update. Window resizing is better now.

Probably going to tackle the 64bit build next. So many errors and warnings!

[tinga]

Yes, resizing is better. thanks
A random page demo, 3x2 sequences, with accumulator, I was not able to do that just with Magic. Also fade in/out for smooth change.
About the random I use page 1 sequence G maj to change qsamo wave, I get always the same random sequence,I have to reset to change ?

[IXix]

I get always the same random sequence,I have to reset to change ?

There's no true random in Magic, so you get the same sequence every time. I need to expose the seed values to the user and maybe add a true random (technically pseudo-randomly pseudo-random) mode.

[IXix]

A random page demo, 3x2 sequences, with accumulator, I was not able to do that just with Magic. Also fade in/out for smooth change.

Cool sounds.

[tinga]

There's no true random in Magic, so you get the same sequence every time. I need to expose the seed values to the user and maybe add a true random (technically pseudo-randomly pseudo-random) mode.

That's not a problem, Accumulator and PeerLFO solve that, these 2 great machines are now under Magic control, same thing for PeerState, combined with Magic state offers great flexibility, so good news for me if this great trio (.Acc,LFO,State) will have a big role to play, anyway ScaleWalker is the perfect machine for random, Magic is perfect to drive gigantic modular structures, and so many other things.

[Candle]

Very , Tinga.

See You In The Shadows…

[IXix]

2020.05.31 IX Magic 1.12e x86
+ Fixed broken Event force note-off modes

Turns out the two 'force note-off' modes have been broken for absolutely ages but they're fine now, better than ever.

The x64 conversion is done but I had to make lots of changes so I want to test drive both versions it a bit more before I release it.

[IXix]

Opening up Magic in the "Live" setup that I was modifying today, I found it completely empty

I have half a suspicion of what might have happened. I know it's a while back but can you remember, did you have two instances of Magic and do you have multithreading enabled? I just encountered a similar error where the data in the file was corrupt for no apparent reason but I had just been testing a new feature using two instances and it occurred to me that it could be a threading issue.

[Candle]

Opening up Magic in the "Live" setup that I was modifying today, I found it completely empty

I have half a suspicion of what might have happened. I know it's a while back but can you remember, did you have two instances of Magic and do you have multithreading enabled? I just encountered a similar error where the data in the file was corrupt for no apparent reason but I had just been testing a new feature using two instances and it occurred to me that it could be a threading issue.

I'm pretty sure I only had one instance. I don't think I've ever used more than one instance of Magic in any of my Buzz Projects. And yes, I do have Multithreading enabled. Two Processors with Thread Affinity Max set & Time Critical Priority (according to my Buzz settings).

I haven't had it happen since - but I have been saving my Template as I modify it. Perhaps having the backup has exterminated that particular gremlin on my machine

I'll experiment with it tonight - maybe play with multiple instances - and see what happens.

Could having an instance of IX Split effect it in any way since Magic has Split's functions built in? I know I have used both Magic & Split in projects before - & that I probably had Magic & Split in that project at some point. Maybe even when the data loss happened..... But I can't remember for sure.

On another note (on a different machine), I've been having problems with Patch Bay retaining it's assigned outputs. I'll get it setup, save the file & then the next time I load the file, the outputs have been swapped. I'm using a 1 Input x 2 Output instance & the two outputs keep getting swapped when I reload the file. It's not a huge issue (it only takes a moment to re-assign them), but I thought I'd let you know. I also have a regular 8x8 Patch Bay in the same project but it seems to not have problems retaining its assignments. It only seems to be the 1x2 version that is having the issue. Again, not a big deal (or priority). Just thought I'd let you know.

Again thanks for all your hard work & these amazing machines! I don't think my "Live" setup would even be possible without them

See You In The Shadows…

[Candle]

So I tried a project with two instances of Magic - as well as an instance of Split (just to see). I had one of the Magics send note data to a QSamo synth while the second one modulated the feedback of an FSM PhatMan. I purposely did not save either of the Magic templates before saving the BMX. I then shut Buzz down & then reopened it before reloading my Magic test BMX. The settings were still there. So in short, I have yet to reproduce the problem with the lost template data. I'll continue to test it out & let you know if I come across anything.

See You In The Shadows…

[IXix]

I'm pretty sure I only had one instance. I don't think I've ever used more than one instance of Magic in any of my Buzz Projects. And yes, I do have Multithreading enabled. Two Processors with Thread Affinity Max set & Time Critical Priority (according to my Buzz settings).

So much for that idea.

Could having an instance of IX Split effect it in any way

Nope, not at all.

I've been having problems with Patch Bay

I'll put it on the list of things to do.

Again thanks for all your hard work & these amazing machines! I don't think my "Live" setup would even be possible without them

You're welcome! Thanks for testing.

[IXix]

I purposely did not save either of the Magic templates before saving the BMX.

That wouldn't make a difference. Of course everyone knows that not having a backup makes data loss two hundred times more likely but that's another matter.

I then shut Buzz down & then reopened it before reloading my Magic test BMX. The settings were still there.

Thanks for testing. Threading problems can often be hard to reproduce though, so that doesn't necessarily prove anything. I've put stuff in place to hopefully make it more threadsafe anyway, just in case.

[Candle]

Alright, I'm glad I could help. If it does happen again, I'll try & figure out what was going on & let you know.

See You In The Shadows…

[Candle]

I purposely did not save either of the Magic templates before saving the BMX.

That wouldn't make a difference. Of course everyone knows that not having a backup makes data loss two hundred times more likely but that's another matter.

Murphy's Law of File Backups?

See You In The Shadows…

[IXix]

2020.07.31 IX Magic 1.14 x86, IX Magic 1.14 x64
+ Added x64 version
+ Prevent (hopefully) possible multithreading file corruption
+ Added new note-cut modes for Peer/MIDI Note Events (see documentation)
+ Added version info to Help menu
+ Updated docs
+ Machine version 1.14, file version 11

Update bump!

I had to do some fairly serious changes to get the 64bit working so let me know if you come across any problems. Having said that, I've used it a fair bit and haven't noticed anything strange.

The other big news is new note-cut modes for the note sequence type Events. You can now cut held notes on the target machine/device that were initiated by other Events, which makes it much easier to spread music across multiple objects, no need to force note-offs manually before sending new notes. See the docs for a full explanation.

Enjoy!

[Candle]

<DOES A DANCE>

See You In The Shadows…

[IXix]

New Year Update bump!

2021.01.02 IX Magic 1.15 x86, IX Magic 1.15 x64
+ Added Test/Reset/Enable controls for Page load events.
+ Fixed note-cut for Page load events.
+ Allow fractional values for Page load event delay/cut.
+ Improved status bar text for Sequences and Steps.
+ Display Event accumulator state in control label.
+ Event accumulator threshold is now included in calculation when Event is used as source for Step repeats.
+ Added random seed options to various context menus.
+ Added 'true random' mode for all randomisable things.
+ Various bugfixes
+ Updated documentation.
+ Added a Page load event to the demo bmx (randomises the BPM )
+ Machine version 1.15, file version 12

This has been far more work than the above changelog might suggest! I ran into a limitation of the rather old dialog updating system and had to rewrite tons of stuff, which of course introduced tons of new bugs which needed to be found and squashed. I think I've caught them all now but stay alert and save often, just in case.

I also had to bump the file version and although I've been using it myself without any problems for a while, it might be a good idea to keep a backup copy of any tracks you care about.

Anything that can have a random value now has it's own dedicated random number generator. Right-clicking the control for any of the following will give you options for setting the seed and for 'true random' mode:
- Event probability
- Peer/MIDI Event on/off values
- Peer Event track
- Step rule

Let me know if you find any bugs or have any questions/suggestions for improvements. Happy Buzzing!