IX Magic 1.18 - July 22nd, 2024

[IXix]

Buzz can't seem to find the dll when I try to load Magic in a previous Buzz project. It then crashes Buzz. It says "%1 is not a valid Win32 application" when I try to drop it into a new project.

I'm not sure I'm following what you're saying. Could you give step by step instructions?

[Candle]

Sure. The following situations happened with v. 1.15d x86

First instance:

1. Opened previously saved BMX containing Magic as one of its Buzz machines.
2. Buzz hangs when it begins to load Magic into the project.
3. Buzz crashes & closes.

Second instance:

1. Open a new, blank Buzz project
2. Add Magic to the Machine View
3. Error message window with "%1 is not a valid Win32 application" as an explanation appears.

I reverted to v. 1.15c x86 and attempted to follow the second instance. No error message appeared on load, however Magic was loaded in as "Crashed". I reverted back to v.1.15b x86 & had no problems loading Magic into a new porject or from a previously saved BMX. I am currently using v. 1.15b x86 as it seems to be working as expected.

Hope that clears things up & helps you troubleshoot. Thanks for looking at this.

See You In The Shadows…

[IXix]

Sure. The following situations happened with v. 1.15d x86...

Thanks. That's weird. I hate programming.

[Candle]

I know you're deep in the code for this wonderful machine, IX, but I just stumbled across a bit of a feature request (bug?) while you're re-coding things (hopefully it's an easy one). So I've been busy re-programming my Foot Controller & Magic for a new version of my "Live" Buzz Setup (it's a little smoother of an interface & gives me a few more options than I had before). What I've noticed is that as I'm adding new Sequences into Magic or updating Step or Event data, if I have more sequences that fit in the page column & I require a scroll bar (which I do with my current setup ), then every data update makes the list of sequences jump back up to the top. It's somewhat annoying to have to scroll back down to the sequence/step/event I'm editing just to update another piece of data only to have it jump back up to the top. I'll post some pics to help explain:


Editing Data


After Editing the Data

So if there was a way to stop Magic from jumping back up to the top of the window (second picture) that would make editing smoother. Thanks for all your hard work IX

See You In The Shadows…

p.s. As you might notice from the pics, I couldn't do what I do without your wonderful machines (Patchbay (I think I'm using five different ones ), Split & of course Magic ). So again, thank you for making my performance dreams a reality!

[IXix]

Yeah, I can see why that might be a bit annoying. It's sort of a known issue but I never really expected it to be that much of a problem as I rarely get into scrollbar territory. I'll get around to it soon.

Love the screenshots!

[Candle]

I'm sure Tinga is someone else who probably "gets into scrollbar territory" I'm just nuts with all my presets, but Magic works so perfectly for mapping my Foot Controller to Buzz.

Thanks for having a look

See You In The Shadows…

[IXix]

Thanks for having a look

No problem, thanks for pointing it out. It's fixed in the next release but I can't let it out until I catch a tricky bug and update the docs so it'll probably be a couple of weeks. If you're desperate (and brave) you can be a beta tester.

[Candle]

Well, I got it programmed as is, despite the scroll jump. But yeah, if you'd like me to Beta Test, I could give it a shot. I'll probably re-name the DLL so it doesn't mess with my "Live" setup (now that I've got it all put back together).

See You In The Shadows…

[IXix]

Oh my goodness how I love writing docs.

zzzzzzzzzzz

[IXix]

The most annoying thing about writing docs is that I tend to discover that the machine isn't behaving as the docs I'm writing say it should. Seriously, if you want to find bugs in your machine, just start trying to document how it works!

[IXix]

2021.05.04 IX Magic 1.16 x86, IX Magic 1.16 x86
+ Improved value parser - better input validation, bad values highlighted in red.
+ Allow complex values for delay times (see documentation)
+ Improved/extended peer/MIDI value syntax
+- new ping-pong/wrap modes
+- additional range specifiers for velocity/random modes
+- new %% value specifier
+ Assorted minor bugfixes that I can't quite remember
+ Machine version 1.16, file version 14

May the 4th be with you! The force is strong with this one.

You wouldn't know it from the changelog but I have been hacking away at this for absolutely ages, completely rewriting several major internal systems. The result is a VASTLY superior input parser which is much easier to debug and much easier to modify, so I no longer have to deal with hideous old spaghetti code whenever I need to change something. I can't tell you how happy I am about that.

You don't care though, all you care about is what's new, so you'll be glad to know that random delay times for Events are now possible, as are friendly fractions of units like "1/4". There are also a few additions to the value syntax which make it more powerful and flexible. The main change is in how value ranges are handled.

I've tried really hard to make sure that the new parser is completely compatible with pre 1.16 songs, testing side by side with the old version. I'm confident that all is as it should be but if you spot red text when you load a song or see some behaviour that you don't think is correct, please let me know. I had to increment the file version again meaning 1.16 files won't work in older versions so make backups before you save if that worries you.

I did have to make one change that isn't backwards compatible but the machine will detect the issue when loading and fix it automatically so there's nothing to worry about. The change is related to use of % for relative values in combination with a value range. If it finds a value like "+%5:50-100" it will change it to "+%%5:50-100" to preserve the correct output.

Enjoy and please report any problems (though hopefully you won't have any)

[rav]

You are huge

You don't care though, all you care about is what's new, so you'll be glad to know that random delay times for Events are now possible, as are friendly fractions of units like "1/4". There are also a few additions to the value syntax which make it more powerful and flexible. The main change is in how value ranges are handled.

Awesome, i was waiting for it

[thOke]

IX ROX
Will try it as soon as possible! Thanks a lot!

[Candle]

Looking forward to trying it. Thanks for all the hard work IX

See You In The Shadows…

[tinga]

Great, more than six month that I didn't do anything in Buzz , thank you for this motivating update .

The result is a VASTLY superior input parser which is much easier to debug and much easier to modify, so I no longer have to deal with hideous old spaghetti code whenever I need to change something. I can't tell you how happy I am about that.

Good news, I'm happy for you, and I will feel less guilty if by an improbable chance I find a bug .

[Buzztler]

I'll have to have to check it out ... (is this correct english?) ...



anyway ...

Dankeschön. Thx.

[IXix]

if by an improbable chance I find a bug .

Haha, yes, no chance you'll find any bugs at all. Definitely no bugs, honest.

I did forget to mention one thing. A few times there was a crash, either when closing Buzz or when pressing the new song button. I couldn't reliably reproduce it so I'm not totally certain I fixed it, so if it happens to any of you please try to remember what you were doing before it happened. It's strictly about what happens when resources are being released though, so it's not going to happen while you're working on a song, only when you close Buzz or start a new song.

I'll have to have to check it out ... (is this correct english?)

Yes, perfectly correct.

[River Cricket]

Is it possible to have Magic route Poly Aftertouch (or Channel Aftertouch, for that matter) to PVST synths?

I've got Magic controlling a pVST synth in a rudimentary fashion (Polac MIDI In > Polac MIDI Bus (right-click > attributes > mode = 0), Event type: PeerCtrl, Note = 'P), but aftertouch messages aren't being received by PVST.

They show up, kinda-sorta, in Magic MIDI view - though it's a bit weird, for instance, Channel Pressure messages just spam the same CC #252 message in the Input window, which seems incorrect, since according to the MIDI specification CC #252 is a "stop" message. PolyAT shows a velocity value moving as I'd expect (along CC #253, which is allegedly an undefined/reserved slot), but nothing in the Input window to show which note is being affected. Probably just display errors, since both channel and poly AT work fine when routed directly to a PVST synth.

[IXix]

Is it possible to have Magic route Poly Aftertouch (or Channel Aftertouch, for that matter) to PVST synths?

Not at the moment.

They show up, kinda-sorta, in Magic MIDI view - though it's a bit weird...

I'll look into that.

[bozz]

It would be cool to create a collective book, every one could purpose a single pattern or a complex thing in one page, a global state, Magic is a modular pattern editor perfect to illustrate Buzz power.

Let us start this collective book idea. Or did it start already? I would be happy even reading just the table of contents, with the titles only. Something like: x and y using technique z for doing zzzub. It would be cool having a simple set of rules, keeping those collective book pages as simple as possible, just demonstrating the core idea, so it can be easily understood by others later. Some other rules could contain which machines to use. I would like something limiting to 1 bass machine, 1 sequence machine, 1 pads machine, 1 kick machine, 1 hihat machine, 5 samples, or so. Just all very simple.