many sliders to multislider with offset value
I am using max midi patches with many sliders in live to send artnet information to another patch.
When using a lot of channels and many universes the CPU goes crazy even when using only a few sliders with automation.
Tried to change slider’s latency to 20 but there are times that the information is so much that gets an 8-core macbook to its knees.
So trying to redesign my patches I was wondering…
Is there a better and lower cpu intensive way of doing this?
----------begin_max5_patcher---------- 1458.3oc4b9saihCEF+5jmBDZtayF4+Cr2M6d+9BTUUQCtMth.UfS2Nyn9tu fMzP5j.FRvJInQpoXHIe9GeGygygN+Z9L2GSemm657WN24La1ulOalZnxAlU s8L2MguuJNLWcXtwh23KS1to7nVn2+agYIga3Ga2ulwy4IxPoHM4gL9Jo9qC hQKgDjOEuvAxPKoKbvrkfhMnKAN2W+lCkqVKRdtwaDCCTGFDW9xgdOM9BKeK vpwEQJIl93K+I5SwElUHcIO6AdR3iw7lGewzPjDyk4eYvzsx5QQUipGR9iW4 ZM55tvw8o3zvBMWKq7v23QODJkYhG2J469s7JbWw6RfFukm9T8v0i2TswoIO 2Bz26XyWmlIM8fqmCfCruMENgx8QgnkGZ+aSDxb4OhO5GfH4MQtnFy58+w74 6dQ8yOluvPyXB++JNaVetTxeWpmhogQOFl7b8NdJMQlK9o9aETK8V7VdvR6H kprV9MrVMrPAM+z+jteOSDF672owQtl6hfGyEolDN2O.xrgmmG9L+2PCe05T G3AACtUv3QTQnXUPGEnHSvgIC63j4fPA0OnbdARNOtXR57Mjy2F.VP9JRTgE eXaXgBmPbgnPAFRLfKfIDW.kqb5fBncyERvzgKPlJ.B4wLfK9SHtfTAPHpmA bwa5vEeU7CB6a.VtHubzQxb4mwKeIUjz+DWptPDzWk+BQkN8wxbgPFRpKn9m .7EBZzWKpFMvVQCdZgF8kiLCMnIEZpthzNzzxsBPfSKznunTMZ.s4ZvAWin4 Ob7Fv8NVE+vTWaB25RvX+yGVN5UlDIxyLUXCfJ.r4Tw6pjJzArvKqGTgcURE x.nBhXNUnWkTAOfka0I6ZFUHWkTAM.pPnl6UPWkTAN.p.Xl6UfWgTQ5HZ9u9 SHFSwDBPee0r1xqCCNe030qMBsv3WtHR.rZAoOy+qEDh7lToFSXMHCNnsvOD 6hlLiScaz7Ao6xR60sAcYVt7swRQdrHhm0Q+Mwz5FSVLSU1APiIZNWVH8p9o cmS0pQcz4SM8v5hjhQ5u.Bo4Ga0YAJrlFqJfWTnL7HMBExZqSnfyTmPGFqUc pbeV2rUyv+Y8cr66pKyzhTJ8fDDqrbg5hFpvFlzNq+ryNUQxE+LnmMYFBuo5 x7d7tkFLWbbs0XYXKMVVEAzYikO89FafyhdeOd9E5mypp2XmhyBby4rnSFmE oSmEHXY4RV9d80YoKm6o3r7u4LVjIiwB2kwx2aYw5Udv9thUUsvOEek2MmuB OY7Unt7ULxxRaEouqWU0HgSwWQu47UnIiuB1kuhfVVlfkGom9Jc8INEaE4ly VAuMrUGofM4NxzumI+Wtbn0rgR0OYiAsVPvyW8.AiRLUGATHvtnHCikpp4vo byJ2ZwRWzARJDT3TR95i0u5ba436avxS2lspljUmtb1c9MhmKEIed58tc0tr wAsVDEwSZVvnHQd4Y3ni6zMVN.CzS4isisziI7o7IrxV5gYfdrGdPlnFPGxY iH50TQhrxAiQ9pBoizOa9phot+F5ZspqKpZicJ37MyL4DOj0uoFU0DSjtGJM 1XLzuIFkuLIGUmhmIAR1agFLvzE9f1SOvtBkf1kOntzCxt5A2kdH1UOjtzC0 t5g1kdrb7EqK83YO8XThNd1Kd2nKcFXO8PLY8Yn8zC0D8frmdXllXpkzimoI lZI8DXhdrW7NwzvcKIGitOBe6oGSV9gDXO8fMPOTf8ziI9Gp8RGivFgaGIP+ G4qtcO5aGQswHb6HDuQP+PL1dS.+wXB3gr2DHXDl.HHzZS.poILYmHRJXL3I EXOdNFU3.4GXsIvXrjH1W2iRaneOSy31NFZeSS31NxwjkqtrTCaP04LP+vCq 8cAAF3zz07O70WeimkWIAkvc2D9RZV4lrEpMEI5MUcfvMiW9eGK5impFILa0 ZgjuRtMS2Ii28YtyK+d9X9+SF62qN -----------end_max5_patcher-----------
Thank you in advance
Please post the other part of your patch that’s on the receiving end of the [s toArtNet] – the problem is that every time you send even the smallest message you’re outputting a whole universe of data – that’s rather silly, you’ll want to only transmit the data entry that’s different from last time. So while what you have there is somewhat sound code, the rest of the patch very well isn’t.
I still think feeding into a large multislider is actually a cool way of handling things in a way, since it provides some neat pattrstorage properties, but in order to get that going properly you’d need some feedback from the multislider out to your live sliders. It’s worth considering, even though it doesn’t affect the efficiency of the patch.
Thank you for your kind feedback. Plz forget the rest of the patch. Even if i delete the sending patch the main CPU footprint comes from this (sliders, list and select patches).
That is what I ask, if there is a better way of organising this mesh and if using jit or not will make any difference cpu wise :)
woah. I’m inclined to think there’s something wrong with your computer almost – you’re saying that a few sliders and a single big multislider is enough to almost kill your pc?
I can’t currently throw anything at your patch to make my mbp struggle, or even get warm.
Are you using the latest version of max and all that?
Well, multiply it by 100 devices automate each like crazy and then tell me how is ableton running while playing all of these. It is a sensitive cpu question, if you can give any good hint please do.
…you have a hundred of these going at once? So how do you know that it’s not just Live struggling to keep up with automation?
What’s your ram situation when running that abomination? And you’re saying that you have the problem even if you "only" run the slider patches and you close the actual artnet sender?
So when you say:
>When using a lot of channels and many universes the CPU goes crazy even when using only a few sliders with automation.
you mean you’re using hundreds of sliders?
I’m wondering if it’d help to have fewer devices with more sliders instead.
Use [change] or anything that would not put data through if you have no change.
You may want to use [speedlim], [defer] and/or [deferlow]
Try to use as less UI object as possible. (message is an UI object)
If you need UI for visual feedback, you may want to have it refresh not really often, but have the data going to whatever that is doing.
Have a look on the forum for slow interpolation, even on the Max forum. I used [pattrstorage] for interpolation of hundreds of UI objects and I got similar issues.
Maybe you’d want to use your multislider directly, with [pattrstorage].
Thank you Musinou that really helped. You gave me the idea of using only one multislider than many sliders. I ll check to see how automation reacts with this. cheers
Do you think gen will do better, should I purchase it as to create an AU device and use it instead? any other recommendations would be really appreciated
I would try having only one device with many sliders instead.