DIY automation Ider

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

pucho812

Well-known member
Joined
Oct 4, 2004
Messages
15,832
Location
third stone from the sun
o.k. how about this...

Console automation that reads and writes to protools but does not change level in pro tools only on the console faders IN other words console automation data is stored,recalled,palyed,etc,etc in pro tools but does not affect the faders on the slow tools mixer....hmmmmmm

Ider(in my best NYD accent) welcome back Dave we missed ya :thumb:
 
It's IDEUHr--or something similar. All those Rs that are left off other words have to end up somewhere!

That's how my Dad says it, anyway... Me? I don't have an accent.

I've never used Da Proto-Tools. Does it have a MIDI track? That could do it.
 
Yeah... i ran some things around in my head for a while regarding that. I thought about a max/msp RTAS plugin that sends the audio out to the converters at 0dB (bypassing the faders and muting the normal route to the converters in the protools mixer) and also generates continuous midi data relating to the value of the fader.

Then using a temperature compensated midi to CV converter to control a motorised fader or somethin.

i figured I’d only want it for channels that were going to be sent to an outboard compressor/saturation/distortion unit because all of the others could just stay in the box.

Paia do an 8channel midi to CV converter here:

http://www.paia.com/midi2cv.htm

It was all part of my very over ambitious summing amp automation daydream!
 
you could probably do it using the midibox idea.

I've thought about this before, but the design is way over my head.
I thought about using something like this:
http://www.ucapps.de/midibox_lc.html
but also have the facers control a VCA or something like that. I'm not even sure it is possible, but it was just an idea.

The real trick would be to reverse engineer Digi's protocol for ethernet controlers and use that to control your automation.

Lots o money....

peace
Gil
 
I've spent quite and embarrassing amount of time pipe dreaming and researching about this one - some background..

I'm a full time project studio owner running Pro Tools HD. I have a different client in almost every day of the week and albums can often be recorded over the space of a year or so. Total recall of all settings is now essential to what I do and I love it - a regular analog console is not an option for the way I or my clients work. I also depend heavily on the editing and processing in Pro Tools to get the job done but as many have protested working with a mouse and screen is unintuitive and limits gut and instinct reactions to the music. I use a Mackie Control heavily for rides etc but the banking on a session with 50 tracks of audio (don't ask why) is a pain in the ass and again gets in the way of mixing so I know I need real faders.

Like Pucho I've thought about the idea of building an analog fader system that is controlled by Pro Tools automation. Paul at Tonylux has done this but it is seriously spendy (price list on tonylux - rrp of $5995 for 16 faders without summing). His system works like this -

-audio is sent from a pre fade aux send fader set at 0db (max resolution from the D/A)
-The analog faders communicate with the DAW using the HUI protocol. If you've ever used a Hui or Mackie control you'll know how this goes.
-The faders are touch sensitive and respond to all automation modes in Pro Tools.
-The fader movements relate perfectly to those represented on screen and the audio is summed in the analog domain.

Apparently designing the analog solo function was a challenge as was getting the scale of the analog fader to correspond with the digital scale in Pro Tools. Several prototype sytems are in use and the response has been positive.

Getting back to my pipedream. I'd like to build a DIY mixer with moving faders controlled as above, mute switches, pans and an active discrete summing bus (solos in PT will do). This is a seriously long term plan as I have much to learn (thinking of doing a diploma in electronic design at night to get my chops up).

My first plan is to prototype a small 8 channel mixer with some used analog faders and get that working. This will be the test bed for the pan and summing circuits. Then somewhere around 2010 :grin: I will dismantle the mackie Baby Hui that I have secreted for this project and figure out whats happening in there - the Hui protocol is not available to the public though a UCapps member claims to have cracked it in recent months. Then I'll try and drop a real moving fader into my proto mixer and spend several months chasing noises/hums and buzzes caused by the digital control circuits. Finally when the whole world has moved onto to mixing in vitrual reality I will build my dream console with 32 faders and mix records in the retirement home I'll by then live in.

I love the idea of building things that I really need rather than just building circuits because they are available. Don't even ask me about the idea I had for the built in Calrec eqs with rotary switches..

It's taken me so long to type this that there will probably be a working schemo on the thread by the time I post :green:

Ruairi
 
Hmm...MIDI and continuous fades...brings back bad memories. MIDI CCs have only 8 bit resolution, yielding 128 values and they need to send three bytes down the wire to do this (one status byte, one byte for the CC number and one actual data byte). Most MIDI implimentations I've used only utilize these "low-resolution" controllers. I don't know how well PT supports them, but I would definately reccommend the high-resolution controllers (basically combining two CCs to make one, 16-bit CC - and 6 bytes down the wire) if you actually want good, clean, fader moves.

In addition, MIDI, being the height of 1983 technology, has a throughput of 31,250 bps. Do the math, and even assuming perfect performance, you're going to hit a wall pretty soon, especially when using multiple channels - not to mention that as a serial connection, no two CC messages can actually be sent at once, so no two messages are ever truly sync'd. You could theoretically do 16 channels with 2.5ms moves, but that's assuming an awful lot of things go right.

I don't mean to disuade you, but from my experience, anything real-time with MIDI pretty much, well, blows - no nice way to put it. If you don't need gentle fades and/or aren't using many channels of automation at once, it's probably doable, and I'd be happy to help any way I can if you decide to go forward with it.

I wonder if there isn't a better way. Weren't there some old automation systems that saved automation data multiplexed on an extra track or tracks as audio on the tape machine? They were ultimately dumped and replaced by computer-based systems because of the hassle of syncing with a tape machine, because you couldn't easily edit automation data, because it was rather inconvenient to use up valuable tape tracks, etc. Using an extra PT track is a lot easier, though, can be edited, is much more reliable, etc. Would it make any sense as a poor man's automation system?

Example idea (I'm sure someone can do better): Choose 16 (or whatever) frequencies over the audio spectrum and assign one channel to each. This audio signal gets routed to the automation system, which filters out 16 seperate signals corresponding to those frequencies. Each fader then responds to the amplitude of that signal. To write automation, all you need is 16 oscillators and something to sum them. Offline editing could be a hassle, but in a rush, you could still tweak things with an EQ over the automation data (Or, better yet, create a custom plugin, one that could even be automated from within the DAW). Need more channels? Just add another audio stream. Could even set up to automate pan, mutes, solos or whatever else you feel you need. Lots of possiblities with custom plugin programming, too.

Yeah, it's ugly. Goes to show you how much I really hate MIDI.

Anyway, just some thoughts.

- E
 
IIRC the hui protocol is 10 bit so that gives us 1024 steps - not too shabby. I use the Mackie Control all the time and while it's not super reponsive it's pretty good. SSL use Hui for their DAW control in the AWS900 and it seemed to work really well and fast at the AES demo I saw - didn't get my own paws on it though. The Tonelux system is out there and has received good feedback so far.

As for timing there do seem to be a lot of people out there with multiple Mackie Control setups or that new Tascam unit - granted many are home users that may not be the most discerning. HUI uses a different midi channel and port for each 8 channels so that may help us on the bandwidth front. I'm guessing that a lot of guys would trade off a little response time for the sheer joy of real analog faders, summing and the ability to patch in analog outboard before the fader (which is a PITA with Folcrom style solutions unless you insert it in an DA/AD loop).

As I said in my post I definitely have the need for the unit but I certainly don't yet have the skills to pull it together so take my input with a grain of salt,

Ruairi
 
[quote author="NewYorkDave"]It's IDEUHr--or something similar. All those Rs that are left off other words have to end up somewhere!

That's how my Dad says it, anyway... Me? I don't have an accent.

I've never used Da Proto-Tools. Does it have a MIDI track? That could do it.[/quote]

thanks for the pronuciation(sp?)Dave. IMO you have a slight one or so I remember but then again there were lots of beers in me at the AES gathering so maybe I imagined it. uh yes it does have MIDI and midi tracks. It's getting better but not up to par with other software. Havn't sat with the AWS yet. But at AES was said to be doing hui protocol. I really would like it to work via ethernet dam you digi won't release that info. afraid? I wonder. I have used hui emulation b4 and such and it was alright but not as good as their ethernet protocol. but I think it one could do eithernet with the slowtools as aboved mentioned would make $$...
 
Some very good points about existing HUI solutions. I've used the SSL AWS myself and certainly never had problems with the DAW control. Splitting the datastream over multiple MIDI connections obviously helps, too - as I recall, the AWS uses three connections for 24 faders (plus one more for total recal data), making eight channels per connection. You also bring up a good point - no need to stick with standard MIDI CCs. They just jumped to mind because it seemed like the easiest way to DIY something.

One big difference with HUI, however, is that we're talking about going in the other direction from the way protocol is generally used. When you write automation data into a DAW from a MIDI controller, the DAW can glide between steps, interpolating, on playback. Going the other way, into a physical fader or VCA - and in real time - it might be a bit more tricky to get that smooth movement.

Again, no reason it would be impossible, and I don't mean to dissuade anyone from trying. I'd love to see this kind of thing work.

- E
 
Why not use existing cheap solutions such as Behringer cybermix or mackie ultramix... Design a fader with a DBx202 VCA controller or another VCA.... bypass the VCA on the cybermix or ultramix and use the ones in the fader.... You'll have the software and all. You'll have to run it on a windows 98 PC but maybe you can interface it through midi with protools. I know some german guy did this for his studerdesk (the faders allready have VCA, only 128 values per fader though).... I don't know wether he controlled it through protools however, but with the ultramix you can make subgroups and mute automations at the same time. These boxes sell for around 120 USD a piece on evilbay and including the remote control and they enable you to control 32 channels..... You can also stack several ones if you need to cover more channels.

Geert
 
Personally I'm not really intersted in using VCAs in this instance if I can avoid them. I'm sure with the right implementation they could be pretty neutral but The idea of a proper analog fader appeals to me greatly. I really do want to have my cake and eat it - I could just use a regular analog board or outboard automation system and lose the ability to recall but that would not be compatible with the way I work.

I'm inspired by the fact that someone (with a considerable track record) has done this with apparent success and also by the fact the the concept is not protected by patents in any way that I know of. What it really involves is the combination of several known technologies, analog touch sensitive moving faders, the (semi protected) HUI protocol and small format console design.

As an aside I've spent quite a bit of time watching the midi info fly back and forth between my Mackie Control and my PT rig using a freeware application called midi monitor - it's over my head though. There is also the possibility of using the JL Cooper standard to implement this project but it is not as complete an implementation as the HUI.

I guess in that classic newbie fashion I'm being ambitious but I don't for a second believe that this will be an easy or quick project. I want to build projects that are driven by a real need rather than just build things I might use because there is a pcb available - I'm hoping this real need will help get me over the many hurdles I will encounter. I also don't mind spending money on getting the thing right - I'd rather not mix the output of a $70,000 investment in my studio through a $100 VCA box - but that's just me. Others will have wildly different needs, expectations and budgets. If this thing ever begins to take shape I can see many threads with things like "surely it wouldn't be hard to add a API/Neve/SSL summing buss, phase meter, Sontec eq on every channel, huge master section, 10 aux sends, G1176 in every channel etc etc.

Final aside - as far as I know Tonylux are using an OEM JL Cooper module for the HUI interface. Can't find much info on this on the JL Cooper website about this. And btw in case the Wolfe comes across this thread I have no interest in or plans to reverse his system. Ideas for a system like this have been around for a while and the need is as great as ever. The automation in Pro Tools is superb and the combination of fader rides and offline graphical tweaks would be very powerful.

Now I'll you back to folks with their feet on the ground.

Ruairi
 
there is already an implementation of something similar at ucapps.de using TI's PGA2310 device (±15V) for controlling attentuation in 1dB steps.

Cheers

R
 
We did a solution for the uCapps project, based on the THAT's "Digital Gain Control With Analog VCAs".

It's the same way like in big high-end consoles. You can control audio via your midi-application.

But you wrote that you're not interested in a VCA solution :-(
 
Tobwen,

this is not my thread by any stretch of the imagination. Pucho threw out a great idea and lots of people responded. I happen to have given a fair bit of time to "my" version of the idea which is based real moving faders ala Tonylux so I threw in a couple of rambling posts come pipedreams. Please please do not let that in any way stop you from posting any ideas, links or anything that might be of relevance to this thread.

I haven't ever worked on a VCA based console and have been a digital kid almost my entire 10 year career in the studio. I don't have anywhere near enough experience with VCAs to discount them as a viable option. The purist in me seeks a cleaner solution but I'm sure reality will knock that out of me in time,

cheers,
Ruairi
 
Then just go for the uCapps solution. Trust me, the solution is VERY good.
Your main problems will be the faders.
I worked with many cheap faders already and the only good ones come from P+G.
For homestudio, the ALPS might be okay. They've got sensitive touch knobs.
 
Thanks Tobwen, I'll definitely take a serious look at that. I've spent quite a bit of time lurking on Ucapps but I find the site to be heavy going. I've got to admit that I understand little of what goes on there,

Thanks,
Ruairi
 
Back
Top