Jump to content

MIDI latency


scube

Recommended Posts

Hello,

A couple of years ago I got an Evolution UC-33E midi controller to control Ableton Live parameters in "realtime"...a disaster. The faders were so slow that it was impossible for me to use them in my live sets..the latency was horrible. So I opted for an audio mixer.

Sources -> Soundcard -> Audio mixer

This method is fine for 2/4 stereo channels but the analog way is impossible for 16-24 stereo channels (too large mixer + lot of audio OUTS). So..it seems that I have to use MIDI controllers. I'd love to have a "quasi-realtime" MIDI controller.. how fast is the response of MIDIbox faders? Can it simulate analog faders and fast volume changes?

Hope it's clear and that no one will take my post amiss  :P

Scube

Link to comment
Share on other sites

Hi Scube,

shortest Pot/Fader conversion latency was always in focus during MIOS development. Basically analog inputs are converted faster than events can be transmitted via MIDI. There is a "dynamic priority" feature, which converts the two last moved faders/pots more often than the others; this improves latency even more.

So, you can expect a latency of ca. 1 mS (or 640 uS if the application gets use of the "running status" feature) on the two last moved pots/faders. And this is caused by the transmission time of a MIDI event.

If you move more faders the same time, it mainly depends on the number of scanned analog inputs, but even in worst case (64 pots) the latency isn't higher than 12 mS, which you won't really notice, as the transmission of 64 MIDI events takes ca. 64 mS...

My tip: open the UC-33E case, and replace the microcontroller board by a MBHP_CORE :)

Best Regards, Thorsten.

Link to comment
Share on other sites

My tip: open the UC-33E case, and replace the microcontroller board by a MBHP_CORE :)

Best Regards, Thorsten.

Ahaha :) Honestly I already opened that controller to do a sort of work of art  ::)

Anyway...1-64 ms are a great timing I think.. I'd like to use ECLER faders

www.eclerdjdivision.com - products - spare parts

It's a serious spanish company...do you think I can use their faders? Sorry, I know that I have to study better the midibox documentation  :-[

Scube

P.S. thanks for your detailed info

Link to comment
Share on other sites

I'd love to have a "quasi-realtime" MIDI controller.. how fast is the response of MIDIbox faders? Can it simulate analog faders and fast volume changes?/quote]

Don't forget the block size dependency if you're running with a PC based native system. A big disadvantage of the PC DAW compared with DSP based processing is, that the PC works in blocks (otherwise it couldn't manage to transfer the data internally to process) and the reaction of a fader move depends on the time until the block is processed. So if you configure your DAW with a big block size, the reaction on your user input will raise.

... couldn't find a datasheet, so I can't help.

Greets, Roger

Link to comment
Share on other sites

Hi, I have a Ecler NUO 5 and its faders are PERVERT GOOOOD ;) The eternity crossfader is the best i had in the hand ever. Channel faders are very good quality too, but you have to watch to take the right values. I don't know what they have actually, but I suppose them to be log and not lin.

greetz

Link to comment
Share on other sites

Don't forget the block size dependency if you're running with a PC based native system. A big disadvantage of the PC DAW compared with DSP based processing is, that the PC works in blocks (otherwise it couldn't manage to transfer the data internally to process) and the reaction of a fader move depends on the time until the block is processed. So if you configure your DAW with a big block size, the reaction on your user input will raise.

Greets, Roger

Thanks for your suggestions Roger. Unfortunately my PC was well configured..the problem was the UC-33 controller. I remember that the rotary encoders worked very well, with a pretty fast response..the s**it were the faders. Anyway, I know and I agree with you that DSP are better than general purpose CPU..my first audio card was a glorious TripleDAT..I still use it for its realtime spectrum analyzer ;)

Scube

Link to comment
Share on other sites

Ahaha :) Honestly I already opened that controller to do a sort of work of art  ::)

Pics please! :)

Anyway...1-64 ms are a great timing I think..

64ms? noooooo... TK said worst case is 12ms... perhaps you mis-read, but 640us (best case) is 0.64ms :D

I'd like to use ECLER faders

As said above, It's unlikely that they will work as they are audio faders and will be a logarithmic taper not linear (search the forum/google to find out what that's all about)

Link to comment
Share on other sites

I know and I agree with you that DSP are better than general purpose CPU..

... yes, better concerning this issue. On the other hand, the DSP-processing is not very flexible when it comes to changing dynamically what it has to do.

Both technologies have ups and downs. At this time, the Latency thing just weights much heavier than others... but the time will come, when... ;)

Greets, Roger

Link to comment
Share on other sites

but the time will come, when... ;)

Sad but true... these days people have made it so that it is cheaper to use a fast chip to compensate for performance faults... so *most* of the time, the bad stuff is hidden.... Crappy, bloated, poorly engineered code on fast chips is the way of the future.

It's the electronic equivalent of:

guy: "my car is slow. the tyres are flat"

tech: "install a bigger engine"

Link to comment
Share on other sites

As said above, It's unlikely that they will work as they are audio faders and will be a logarithmic taper not linear (search the forum/google to find out what that's all about)

Can you suggest me the best linear 10K faders? I can spend 4-5 euros each one. Honestly I searched for fader implementation and I found motorized sliders only..I understood that I have to connect the sliders to AIN modules but I did not found infos about soldering and connections  :'(

Some suggestions for this also?

Scube

Link to comment
Share on other sites

I'm not sure about faders but I do know that alps is a good brand... Finding the perfect mix of price and quality is a tricky thing which can only really be solved by looking around a lot ...

As for the connections, you can find the diagrams/schematics on the ucapps.de site :)

Link to comment
Share on other sites

64ms? noooooo... TK said worst case is 12ms... perhaps you mis-read, but 640us (best case) is 0.64ms :D

Sorry Stryd, but I think you may have ...eep... missed something.  :o

I think TK's remark was that the worst case scenario of moving all 64 pots at once (quite tricky for your standard 2 handed, 10 fingered simian) would only lead to a PIC latency of 12ms.

This said, transmitting all that data via cumbersome old MIDI would be completed in just under 64ms (actually, more like 61.44ms, but who's counting)

As TK also said:

Basically analog inputs are converted faster than events can be transmitted via MIDI.

Edit: In fact, I've just realised, if all the 64 knobs were assigned to ordinary CC numbers 0-127 (or rather, 0-95), the transmission would be done in less than 41ms (40.96ms), as these instructions are only 2 bytes long.

Link to comment
Share on other sites

Can you suggest me the best linear 10K faders? I can spend 4-5 euros each one.

... you're not really finding a good fader for 4-5€.

http://www.produktinfo.conrad.com/datenblaetter/425000-449999/442844-da-01-en-STUDIO-FADER_10K_LIN.pdf

This one is about €30... if you want to talk about good. ;)

Greets, Roger

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...