hey your suggestion is really nice ..
The problem is ....
SL plays the Mp3 / Wave files follow the Analog Input data of Time Code Vinyl (or CD) ...
The only solution can be record on HD only the variation of velocity ...
So probably store the variation of Input Time Code ... you can save 50% of audio data written ...
But .... all depend on the internal audio engine.
I think taht this is a good idea ... but not easy to implement.
NOTE: nuwave_afro the output data goes into the HD .. so on USB side .. nothing changes about USB traffic ( only trap the fader automation ).

It would reduce the time needed to write down the mixtape.

nuwave, I think this sounds like an interesting idea, but as you said... this is the complicated problem: "how exactly to translate deck motion into editable automation i have no idea."

A Scratch Live plugin for Ableton?

My latest note was related only to the bandwidth on USB channel.

That was a generic comment from my side, praising the advantages ;)
My quote was not on point, true.

@gevola - exactly. record the direction and velocity input from control disc instead of the audio. if that data controls what the engine does live, i dont see any reason why it cant be saved and applied to the audio file later.
how does this NOT save on bandwidth? velocity and direction is one float and one binary per channel vs the current 5 aif files per mixtape .als file... this isnt being transferred over the usb connection?

@nik39 - i can live without the editable automation although that would be SICK. it will first require the same redesign of the ableton audio engine the that scratching clips will need. and this needs to happen in order for the bridge to be considered a success. scratching clips is clearly the #1 missing feature in the bridge. im worried that if serato/ableton doesnt do it, someone else will. have you looked into the ms. pinky plug-in at all?

I don't knwo which is the CPU present into SL1 / SL3 hardware.
Currently the SL1 / SL3 send a simple Digital data through the USB.
So I can't assume that you can implement the engine that analyze data and convert into speed variation.
Moreover .. this change need .. to change also the SL application that need to use this new data format.
Moreover I think that this algorithm can't assure a low latency.
So IMO i think that the HW should leave the actual HW implemnetatio, that assure a high performance..

i dont know either, but are you saying that NO audio is being sent from hardware to cpu over usb?
it looked like audio output was being recorded just like normal, but now including fader automation and +4 stereo channnels. this would seem to be the physical limit of how much a usb 1.1 connect can handle.

but if it is only a digital data signal being sent, then even better. that means my cpu (not the hardware's) is already processing this information, and could in theory, record that process. yes, hopefully the writing of this data would not affect latency.

i agree that performance level is still the main priority.

That is wrong.

The SL box is... dumb. Dumb as in it does not process audio (else than A/D/A).

BTW, I have discusses this with Ableton support and what you suggest has a major flaw:

What would you record? I assume you would record tempo changes, correct? This will work as long as the movement is always strictly positive. The moment you stop the record or move it backwards - how do you want to record this?

@nik39 - similar to the obstacle 'scratching clips' faces, yes?
i realize the master transport must remain constantly in forward motion, but the motions of the virtual decks, which is what we want to capture, do not.

what i am suggesting is not to have deck motions as editable automation in session/track view, but to have deck motions *applied* to the original audio file to generate the 'recording'. it would only potentially be editable in some sort of editor plug-in. or maybe incorporated as a envelope somehow.

but even without editable automation, it could still give you some cool midi benefits - substitute a different sound for a scratch, swap a track with another in a juggle (very interesting. i would tighten all my beatgrids down to the sample). and it would also cut down on usb traffic... or would it? if you could please clarify your first statement:


i'm still not sure whose model is right? once and for all, i'd like to know what exactly gets sent via usb in each direction?

i appreciate yout interest in this and that you talked with ableton heads about it. thanks for your patience in explaining some of the simple stuff. i'm trying to be as clear as possible, but im not a super-techie and its always hard to articulate abstract ideas.

Pardon?



SL1/3/57/68 -> control signal from vinyl (via USB) -> computer
computer -> music/generated audio (via USB) -> SL1/3/57/68


That's good and important.


Neither am I.


Absolutely! At least for me :)

'scratching clips' has been a feature request since early beta, but ableton isnt really designed for that. it seems to run into the same problem of how to interpret and display forward and reverse motion while preserving forward motion. seems to be the same issue


okay. even when recording? obviously, when using serato hardware as a soundcard, audio does go from box -(usb)-> computer. how does that work with the bridge (can you record live audio, record, etc in mixtape mode)?

Gotcha.

Nah, recording movement and making it accessible as mixtape are different issues I think.


Not sure what you mean.. sounds like you are overcomplicating things here ;) Things do not change when using The Bridge I assume. Why should they?

ok. the model u provided has control signal goin from serato hardware to cpu and audio going the other way. but audio ALSO goes from hardware to computer when recording in serato pre-bridge (example: recording vinyl thru usb). does the bridge record this way too?

maybe i'm missing something here. i dont have much experience recording thru serato prior to the bridge. but doesnt mix output audio go from hardware to cpu? you CAN record vinyl using the hardware as a soundcard, ive done that.

if the bridge records the audio on the computer without sending instead of the final mixer output then its a change.

again, hard to articulate. hope thats clear.. ?

It should in this case.


Correct.


Eh.. I don't understand this sentence. Can you explain?

ok, if the bridge records the same way that serato always has then its not a issue, but i can see why they would opt for a more efficient recording model, look:

normal record path:
control vinyl -> serato box ->(control signal via usb) -> computer -> (audio via usb) -> serato box -> (record output via usb) -> computer

different bridge record path:
control vinyl -> serato box ->(control signal via usb) -> computer -> (record audio to .als)

if the computer processes the audio as per the control signal data. whats the point of sending it back to the mixer to have it come back again to record. the bridge tracks mixer automation already so the computer already has everything it needs. this is where usb traffic could be seriously cut down. are we certain that the bridge does not do this already?

Not sure what u mean with cutting serious usb traffic. Of course u would not send the audio from the computer to the mixer and back to the computer - TB records pre fader audio (post movement) and that is exactly what SL already creates. This audio is being rendere/written as soon as u click on "save recording" if I am not mistaken.

Now when u record real vinyl then I must say that I havent checked it yet.

only IF the audio is being sent both ways, then this method would save. if its already handled entirely on the computer then it doesnt change traffic at all.

but thats why im asking. it helps to know how it works. it sounds like theres not much more that needs to be done - but i'd really like to see automation for effects, flexfx, loops and rolls if deck movement automation is too ambitious for now

It can't be sent both ways on the SL1 and TTM57 - not enough USB bandwidth.

ok, now that thats clear, the idea doesnt sound so huge anymore. i was really concerned about how much audio was trying to be squeezed thru a very small pipeline. i'm glad it was engineered with that limitation in mind. good. now..

one of my favorite features about ableton was how most of my projects would take very little space on my hard drive until final rendering. ableton accesses the files and handles them according to the projects instructions. similar concept as midi. i hope this is the direction that the bridge is heading. efficiency in engineering is everything. ableton always had that. now we have large uneditable aif files that is kind of a halfstep backwards in terms of mixtape construction in ableton.

undrerstand im VERY happy with the overall development of the bridge, great work everyone! but this concept on the mixtape side, and clip scratching (on the beat performance side) would be the 2 features that would go a long way to making it a truly unique production tool.

@nuwave_afro

Watchwww.youtube.com

Your request should be the Bridge but was more near to NI remix decks... It is by far doable but require imaginative people, not just technicians.

If you want more info about it (or even serato guys) feel free to contact me.

Regards.