How to set up automatic compensation of HW insert?

[meech]

I recently got a hardware compressor that I want to use as an insert. I have it set up and analyzed waveforms to determine that the latency that occurs from going in and out of the computer/converters. How can I tell Pro Tools to automatically compensate for this? It's adding 57 samples of latency at 96k. In the I/O Setup I can specify delay in milliseconds, but 57 is less than a millisecond. Is there a way for me to set this up, or am I going to have to tell Pro Tools to ignore its predicted latency (84 samples, too much) and then manually shift tracks? Thank you for the help

[Stig Eliassen]

Set it up under the Insert tab in the I/O Setup. Use equal numbered ins and outs. If you use output #7, the signal must return at input #7. Enable the Compensate for input delays after record pass under the Input tab. Turn on ADC - done.

[meech]

Thank you for the reply. This doesn't seem to change anything. Pro Tools still reports my HW insert as 84 samples of latency, but when bounced while compensation is off, I'm measuring 57 samples exactly. Bouncing with ADC on results in the bounce being shifted too early.
I can compensate after bouncing manually, but I really shouldn't have to, I should be able to tell Pro Tools, "this is causing 57 samples of latency." When I bounce a track that has several plugins and this HW insert, I have to take note of the latency with the HW insert inactive, command-control click to ignore latency, activate the HW insert, bounce the track, and shift the bounce to compensate for the plugins+57 samples. I hope this can be sorted out.

Thank you for the help

[Stig Eliassen]

Analog outboard set up as inserts in Pro Tools doesn't need any manual manipulation to work. Furthermore, to test latency for a digital outboard device you need to have ADC on. PT takes care of the round trip, and you can then measure the latency for the device itself.

[meech]

Thank you again for the reply. I'm sending tracks to an analog compressor through Apogee converters, and Pro Tools is not calculating the round trip latency correctly.

You can see in these pictures that Pro Tools is reporting 84 samples of latency, but once bouncing the track, the timing is off. I'm getting the exact same results whether "Compensate for input delays after record pass" is checked or not.







thank you again

[Schwabbl]

Pro Tools will automatically compensate for any of the digi hardware (192 I/O). But I also use the AES/EBU on my 192 to go to a Mytek 8x192 converters. This has a little more latency (something like 86 samples at 48kHz). In the H/W Insert Delay menu in the I/O dialog box you can set delay comps in 100ths of milliseconds. So for my 86 samples at 48kHz it's around 1.79ms. Matches up perfectly for me. Each sample rate is a little different, but it's easy to setup I/O for all the sample rates. I name them MIXING 48kHZ and such. I'm not super creative at naming.

Best of luck and enjoy your Apogees.

Schwab

[meech]

Originally Posted by Schwabbl

Pro Tools will automatically compensate for any of the digi hardware (192 I/O). But I also use the AES/EBU on my 192 to go to a Mytek 8x192 converters. This has a little more latency (something like 86 samples at 48kHz). In the H/W Insert Delay menu in the I/O dialog box you can set delay comps in 100ths of milliseconds. So for my 86 samples at 48kHz it's around 1.79ms. Matches up perfectly for me. Each sample rate is a little different, but it's easy to setup I/O for all the sample rates. I name them MIXING 48kHZ and such. I'm not super creative at naming.

Best of luck and enjoy your Apogees.

Schwab

I didn't realize I could enter fractions of a millisecond in this dialog, it almost worked! Unfortunately, it doesn't replace the calculated latency of 84 samples, it adds on to it. The latency is really 57 samples, but I can only add on to the 84 pro tools think there is. By entering a value of .01 milliseconds, it then says there is 85 samples of latency overall... but thank you for the suggestion

[j200]

I hit this same issue with my Apogee converters (AD16x/DA16x) when running into my Digi 002 via optical (ADAT) input. The Apogee converters are faster (wrt round trip DA/AD latency) than the converters in the 002. With PT9 this setup does not work properly with ADC on a H/W insert. The result is that the H/W insert on the Apogee converters ends up too early by 10 samples.

It appears that if the converters you use to feed the digital inputs on the Digi interface are faster than the base interface you are out of luck for a hassle free PT9 ADC setup. If the converters are slower then you are fine; as you can *add* latency figures into the H/W Insert Delay setup, but you can't *subtract* latency! I've tried to put a negative value in the H/W Insert Delay and I get a popup stating I've entered a invalid value. A H/W insert on the 002 lines up properly, but not one on the Apogee's fed into the 002 via optical.

My workaround is to add a Time Adjuster plugin to the track that I'm placing a H/W insert on (when feeding the Apogee's). I add 10 samples with the Time Adjuster and things line up all proper. I'd like to see this fixed, but honestly my 002 is not long for this world anyway. However, does this problem exist with HD, and interfaces like the HD I/O or is this a PT9 non HD only issue?!?

Meech: are you using PT9 HD and hitting this issue? What hardware are you running?

Lastly if I use PT9 Native with my Apogee converters (with an X-HD card), or an HD PCIe core card will I enjoy hassle free ADC; or will I be subject to this same latency mismatch?

Thinking about this more; is this just a flat out bug? I would think that the latency figures you put in the H/W Insert Delay setup should be absolute (the latency for my box is X). Those figures should not be a delta to be added or subtracted from any base. The manual on page 962 seems to imply as much.