If not TDM, then what?

[s2n]

If not TDM, then what?

Logic running MOTU hardware (with it's zero-latency monitoring and DSP-based mixer) and an MBox with PT LE for editing audio.

Or, ask Dave LeBolt nicely for Pro Tools LE to support CoreAudio and 3rd-party hardware.

[Drew Mazurek]

You're wrong.....(snip) but placing plugs will drive that level up, and quality plugs up even higher, usually to an unacceptable level.

which is it? Am I wrong or is it unacceptable?

[ianmiller]

I did not intend to start a flame war - as I said in my post, I love the TDM system I have and recognize its strengths. I am not recording live bands and the only artist I record is me.HD is worth it for those who are pro and have a need for large track counts or plentiful plug ins, That's not me - hence my request for advice.

Personally, I think Jon makes great points - and I wonder if Digi could do a cut down TDM system for that middle market. I can afford more than an MBox but don't to be held ransom to spend $4-5k ro stay current. A $2-3k all in 48 track TDM system including interface would win market share IMHO.

Anyway - interesting debate and thanks for the feedback.

[Chris Cavell]

Sorry, not meant to be a flame war at all. Just wanted to make sure that no one is thinking that you can't have a RTAS plug enabled on a record enabled track on an LE system. You can. In my LE experience, a buffer of 128 on an LE system isn't noticeable to somewhere around %75 of the clientel, and a buffer of 64 doesn't impart an unacceptable amount of latency for anyone I've met...perhaps there do exist people who find this minimal latency to be very discernable and annoying. The latency on an LE track is increased by any plugins enabled on the track, just as it is on a TDM system. I use the furman HDS-6 system wired throughout my studio and prefer to let the musicians set their desired mixes themselves (unfortunately no panning for the mono sources) using 6 hardware outs. I do this for both my LE and TDM systems. With a buffer of 64 in LE, I don't need to load plugs while tracking b/c the musician can tailor their mix to their own needs (I.E. I don't need to tailor my mix to thier needs by loading plugs). However, I often adjust their mixes if I notice that they need help with something and don't know how to get their headphones set in a manner that will be condusive to a good take (like a singer losing the melody often does better with a tad more piano). Adjusting for the latency of plugins on an LE system when mixing is another matter altogether. The delay display in the channel strip appears to function perfectly on TDM systems, but on LE systems seems to display obvious nonsense (zero delay for nearly all plugs on the market).

[Giles Reaves]

On my LE system, a zero on the delay display means just what it says...zero.
I just did a test to find out, just to be sure. If I bounce the same audio throught two different channels (one with no plugs, and one with 4 plugs) there is no time domain difference between the two resultant channels. The delay display says zero, the delay is zero. If I do the same thing while the first two tracks are in record, than I get 2x the buffer of delay. If the buffer is set to 128, I get 256 samples delay (the same with and without plugs). As far as I can remember, it's always been that way. One of the avantages of LE has been that you can insert a plug on snare (for example) and not mess with the overall drum sound.
HOWEVER, there are plugs (L1 and Maxim come to mind) that use "look-ahead" technology and you will see a difference with these plugs. The good news is that the delay display accuratly tells you this is happening, so it's not just random. In the maxims case, the delay is 1024 samples on the display, and 1024 samples measured after bouncing a track.

Chris, can you repeat my tests on your system and confirm?

(I'm using McDSP plugs for my test: Analog Channel, Compressor Bank, Filter Bank, and MC2000 in that order. I consider them to be extremely high quality as plugins go)

[Chris Cavell]

Cool, I thought for the longest time that LE had been using some sort of delay compensation on many plugs where there was an obvious delay on my TDM system, but after hearing so many people shout at me about it, I never tried it out. I did write digi to ask them about it since the RG clearly makes it appear that the DLA display shouldn't even exist on LE systems even though it's there on all of them, even on PTFree! The response from digi did mention the look ahead plugs definitely having delay just as you noticed.

Here's that e-mail:

Hello,

Thank you for contacting Digidesign Technical Support.

In the future please use the link below to access the tech support request
form on our web site. The information requested in the form will help us
give you more specific information. I took a look at this and it is actual
functional. It reports 0 delay for everything I tried except for Maxim, it
reports a delay of 1024 samples for Maxim. I tested this with the same
material on two tracks, the one with Maxim was delayed. This makes sense
because Maxim is a look-ahead limiter so it needs to analyze information
and make calculations then process it. I would imagine other look-ahead
limiters may also exhibit delay. Since there is no time adjuster you would
need to nudge the regions in the track back in time by the reported number
of samples. I will for this information to our tech writers. Thanks for
the good question.


Best regards,
Digidesign Technical Support

===========================================

All replies to this message must be sent through:
http://www.digidesign.com/tsr

Other helpful links:
http://answerbase.digidesign.com
http://www.digidesign.com/compato
http://www.digidesign.com/support
http://duc.digidesign.com

===========================================



-------------------- original message:

> Dear Digidesign, Page 578 of the manual states the channel delay
>indicator shows delay incurred by TDM plugins... The delay indicator
>exists on LE systems as well where it is impossible to load TDM plugins.
>What does the delay indicator display on an LE system? Is this a
>misprint or is this a feature that is not truly functional in LE that
>just appears to be? There have been a few questions on the DUC about
>this and I'd like to be able to answer them correctly. Thanks, Chris

I'll check too, but I think this digi rep confirms your observations without my having to as well.

[hurdy gurdy]

Dmazurek wrote:
"it's been my experience that singers hear/sense the delay and it's unnerving for them".

This is absolutely true and totally unacceptable.
Also, just curious, but what is the delay/latency of a drummer wacking his snare that is 20 feet away from me (the guitar player/singer). This is a question for someone smarter than myself, or quick with math, but there are milleseconds of latency in this situation however minimal.
Correct me if I'm wrong (with a firm hand please).
Enjoy!
L.A.Branville

[Chris Cavell]

less than 2 milliseconds (around 1.8 I think).

Dmazurek wrote:
"it's been my experience that singers hear/sense the delay and it's unnerving for them".

This is absolutely true and totally unacceptable.

I agree to this, when the performer notices it. The best scenario is to run it through an outboard mixer...period...for LE as a result of the HW buffer delay and any induced host based dsp delay and routing delay...for tdm as a result of dsp induced delay (rtas or TDM) and routing.

It's called appendix A.

[brokemusician]

I tried recording on a 001 a couple of years ago and found the delay unbearable, but that was on a 400 mghz G4. My point is that the delay in native systems is going to go away soon if it hasn't already with the whole computer industry trying to get faster by the minute.

[Philthy]

less than 2 milliseconds (around 1.8 I think).

I agree to this, when the performer notices it. The best scenario is to run it through an outboard mixer...period...for LE as a result of the HW buffer delay and any induced host based dsp delay and routing delay...for tdm as a result of dsp induced delay (rtas or TDM) and routing.

It's called appendix A.

You don't address the previous post where someone mentioned monitoring through plug-ins, which I myself also do extensively. The TDM plugs have far less latency and you can actually monitor through most of them without any noticeable delay. This is a matter of routine for me.

Setting up monitor mixes through ANOTHER mixer is a serious pain in the #$$, and you can't monitor through plug-ins at that point. I don't want to keep two mixing boards going during a session; one is enough. That keeps me from getting more involved with native systems. In PT, the 32-track limit doesn't help things either.

As far as debates about native growing more powerful- hey, I just upgraded my G4 processor- now I have no-latency monitoring through the PT mixer, 64 tracks of audio, AND more RTAS plugs than I could ever need.

The solution to the original post: don't fix what isn't broken. Upgrade your processor if you want more horsepower. The Mix systems should make it to 10.3- Digi said as much.