Clip Mapper 1.0

[S4racen]

Ahhhh thats where i was going wrong, i hadn't set my pattstorage properly!!

Cheers
D

[concrete_cow]

Hey Guys,

Sadly no luck with 0.9.9. Still getting device mapping bleed. I'm going to rebuild my set from the ground up with 0.9.9 tonight. The current DJ set is the product of a couple of months work and there may be something i've set up previously that's causing issues with clip mapper... i'll report back later tonight after some more testing...

Appreciate the updates Bjorn... i'll be back!

Cheers,
Sunil

[concrete_cow]

Sorry fellas not had a chance to rebuild my set yet. Been putting it off as i'm making a mix for our next promo CD. Will hopefully get a chance later in the week when it's done

Darren, how is the device bleed issue for you now, are you still seeing it with 0.9.9?

Cheers,
Sunil

[S4racen]

Device bleed happens occasionaly but not at all often as it used to... I might get one value change every 15 buttons press, but that said i'm trying to break it but bashing through the clips within milliseconds.....

It works stably enough for me to build it in permanently....

I'm a couple of days away from finishing my patch now....

Cheers
D

[concrete_cow]

Cheers D. I'll prioritise rebuilding my set in that case

[hoffman2k]

What I reckon to be the problem is the API call that is made for the names, min, max and value each time a new parameter is changed.

I've already reduced API calls a lot and figured these 16 wouldn't be taxing in a real-time situation.
And the data of the last call should always pop up.
However, now that I hear you do these changes within milliseconds, I can see how a little chocking is possible. But it still doesn't explain the last values bleeding thru.

I could build a system that replaces that API call with a coll to read from, but I can't think of a system to keep all parameter values up to date all the time.

Maybe I can streamline this by storing the name, min and max in a coll and just make the API call for the values.
Its definitely possible. Wetter it'll be easy and not break things is another question.

[concrete_cow]

Bjorn, I remember getting device bleed even longer periods between clip launches, i.e. 5 seconds or more. So maybe timing isn't that significant to the bleed problem? Using col to store device values sounds like it may be a solution, although I can see that 'refreshing' col efficiently may be tricky with millisecond changes lol! Is this what you mean? Let me know if I can test anything...

Off topic a bit, but hopefully a quick answer section! Bjorn, I've been looking into making an offshoot patch of the clip mapper. Basically, i want to create a version that features only 16/20 buttons and no knobs. I tried modifying one of your patches around 0.9.6, but it didn't work. May have been because of overlapping clip mappings, and i didn't really try hard to get it working because device bleed was making the original patch too unstable. Do you know if this is a simple thing to implement in 0.9.9? Copy/paste and edit some message box values? If so i'll have another crack at it!

Darren. I currently use a 4x4 clip launch section from tracks 5/6/7/8 on my apc40 to control loopers in my 4 'deck' channels. I'd like to use the the modified clip mapper mentioned above to implent the same looper section, along with a couple of other ideas i have for master FX (effectrix). The problem I have is that currently M4l buttons in the clip launcher seem to act as toggles only. I think this is a limitation of M4L afaik. Do you know of a way to get the M4L buttons to inherit the behaviour of the device they're mapped to or is it a case of work around? Also, do you know if the APC40 clip launcher LEDs would behave correctly if midi mapped or would you use a different method to map the APC40? I hope this explanation is clear (it's probably not lol!)

[hoffman2k]

Would you be able to capture the problem with video?
I still haven't managed to reproduce it, but before anything I want to make sure I can see the problem before I go modifying the device.

In regards to modifying knobs/buttons, just "paste replace" the bpatchers that hold the GUI elements. Copy a button bpatcher and paste replace it over the knob ones.

[concrete_cow]

Cheers, I'll get a video over to you tonight. Btw, are either of you using the beta's or are you using standard releases of live? I've been using 8.1.1 standard only, so don't know if the problem is still there in the betas

Thanks for the tip btw...

[S4racen]

Hi ya,

On beta6 at home.... The lights update perfectly for me, i've had the clip mapper as a floating window since the start and so can watch the dials update, sometimes it takes a while to update the screen but the knobs jump to the right values immediately!!

Cheers
D

[concrete_cow]

Cool! I like the idea of a floating window. I'll see if i can figure out how to do that!

Btw Darren... How many clips do you have mapping back to your Clip Mapper? I think I've been seeing a pattern where the more clips i mapped, the more device bleed I saw... I have 14 clips mapped in my current DJ set... with a total of 108 knob mappings... Also all of the mentioned clips are on the same midi track...

[concrete_cow]

Hi Bjorn

See the movie below

This was the firt attempt made to demonstrate the issue

http://dl.dropbox.com/u/121914/Recording2.mov

Btw. The vid was taken on live 8.1.2b6... behaves same as 8.1.1

Thanks
Sunil

[hoffman2k]

Thanks for that. I'll see what I can do.

Cheers

- Bjorn

[hoffman2k]

Good news, I think I nailed it.

http://covops.dreamhosters.com/uploads/ ... 1.amxd.zip

Turns out it wasn't API calls after all. I was able to reproduce the bug and now with this fix I can't reproduce it even using follow actions on 1/4th of bar at 300bpm.

[concrete_cow]

You're sh1tin me!!!

Checking it out right now