DP 9.1 Hardware inserts

[Robert Randolph]

So I got to trying this out today.

I'm somewhat baffled as to why it lists the bare audio I/O rather than using bundles. It makes it excessively confusing for someone that has a lot of I/O that's neatly arranged in named bundles.

Not only that, but the latency detect always gives me a different and wrong number. Even with a straight pass-through (output connected to input), it gives a poor reading.

Does anyone have this working in a sane manner? Is there some trick to it? The manual has basically nothing on it.

edit: I'm getting ridiculous values like 12000-13000 samples when I have a 64 sample buffer size selected. Not even close. It should be exactly 332 samples for this interface.

edit 2: Welp... changing buffer sizes with hardware insert open crashed DP.

[bayswater]

That's disappointing. I have been using hardware and external processors all along, setting them up manually with Aux tracks and manually pinging to set the delay required to offset latency. it would be convenient to have it all automated.

Typical round trip time on my setups is 200-300 ms, so 12,000 seems bizarre. Maybe it's not reporting samples? Do you hear a delay that long in real situations?

[Robert Randolph]

That's disappointing. I have been using hardware and external processors all along, setting them up manually with Aux tracks and manually pinging to set the delay required to offset latency. it would be convenient to have it all automated.

Typical round trip time on my setups is 200-300 ms, so 12,000 seems bizarre. Maybe it's not reporting samples? Do you hear a delay that long in real situations?

The delay definitely isn't that long. I checked it manually. It's exactly 390 or 405 samples (depending on certain settings that I'm familiar with).

I've had a couple crashes now too. Changing buffer sizes with it open seems to be rolling the dice.

[bayswater]

That's disappointing. I have been using hardware and external processors all along, setting them up manually with Aux tracks and manually pinging to set the delay required to offset latency. it would be convenient to have it all automated.

Typical round trip time on my setups is 200-300 ms, so 12,000 seems bizarre. Maybe it's not reporting samples? Do you hear a delay that long in real situations?

The delay definitely isn't that long. I checked it manually. It's exactly 390 or 405 samples (depending on certain settings that I'm familiar with).

I've had a couple crashes now too. Changing buffer sizes with it open seems to be rolling the dice.

Yes, I understand delay would not be that long. What I meant was, what happens when you play back through the internal effect? Does DP actually apply delay comp of 12,000 samples?

[Robert Randolph]

That's disappointing. I have been using hardware and external processors all along, setting them up manually with Aux tracks and manually pinging to set the delay required to offset latency. it would be convenient to have it all automated.

Typical round trip time on my setups is 200-300 ms, so 12,000 seems bizarre. Maybe it's not reporting samples? Do you hear a delay that long in real situations?

The delay definitely isn't that long. I checked it manually. It's exactly 390 or 405 samples (depending on certain settings that I'm familiar with).

I've had a couple crashes now too. Changing buffer sizes with it open seems to be rolling the dice.

Yes, I understand delay would not be that long. What I meant was, what happens when you play back through the internal effect? Does DP actually apply delay comp of 12,000 samples?

Yes.

I went to try it again about 30 minutes ago, and I can't even get the damn thing to pass audio now!

[Phil O]

Ouch! Thanks for the heads up.

[Robert Randolph]

Ouch! Thanks for the heads up.

Please try it yourself if you can! I can only assume that MOTU had this working on their end.

It must be something about my system that's messed up somewhow, right? Is anyone else seeing weird issues with it?

[David Polich]

Or try another piece of hardware gear. It could be the hardware unit itself...

I really doubt MOTU would not quality check something like this and release 9.1 knowing it didn't
actually work.

[Robert Randolph]

Or try another piece of hardware gear. It could be the hardware unit itself...

I really doubt MOTU would not quality check something like this and release 9.1 knowing it didn't
actually work.

I'm just doing a straight loopback while testing this. Out->In. No different than a piece of hardware with unbuffered bypass.

[corbo-billy]

I really doubt MOTU would not quality check something like this and release 9.1 knowing it didn't
actually work.

Have you try this new Plug-in ?

[bayswater]

I'm somewhat baffled as to why it lists the bare audio I/O rather than using bundles. It makes it excessively confusing for someone that has a lot of I/O that's neatly arranged in named bundles.

I noticed the manual says it uses the bundles. I invoked it and it didn't list my bundles.

Send: Choose the output (or output pair) on your audio interface that is connected the input on your outboard gear. The choices in this menu are set up in DP’s Bundles window Outputs tab.

[Robert Randolph]

I'm somewhat baffled as to why it lists the bare audio I/O rather than using bundles. It makes it excessively confusing for someone that has a lot of I/O that's neatly arranged in named bundles.

I noticed the manual says it uses the bundles. I invoked it and it didn't list my bundles.

Send: Choose the output (or output pair) on your audio interface that is connected the input on your outboard gear. The choices in this menu are set up in DP’s Bundles window Outputs tab.

Yeah, no bundles here. It's very confusing for me because with both my interfaces, I end up having to select oddly numbered things for the same physical I/O combination. Like send to in 15, return from 17... both for ADAT 5.

Edit: I submitted a techlink for this.

[Robert Randolph]

Regarding bundles:

I got this reply to my techlink...

Thanks for reaching out. Currently, the hardware insert plugin will not publish custom Bundle names, just the channel numbers. We plan to add this functionality in a future release. I recommend staying up to date with new DP releases.

[Maxxy]

I am also not able to get Hardware Insert working ...like your experience, I seem to be getting either zero samples detected or some very large and wrong reading.

I also did simple interface out and back in latency test... I can't get it anywhere near right....if the detected time is very large the recorded audio comes back in early ... with larger buffers ever earlier

It seems so obvious but it took me a while to realise that the plug-in *is* the send...you don't use any other send...I also discovered some routing issues I had in the 16A matrix...I was getting submarine type-pinging echoes that I chased down to one of the Monitor mixes is the 16A's web app

At the moment it is seems random to me...I did shut off all networking without effect...and am using the interface via USB on its own bus at the front...as I understand it there are 2 USB busses on this Mac Pro...one with 2 slots on the front end and the other with 3 slots at the rear...so I have iLok, tracking ball/keyboard through Cinema Display ports, and USB hub with MIDI in the rear slots...is that right?

Like you I'm stumped for now ... hopefully there'll be a fix at some stage

Thanks a lot

[bayswater]

I'm getting mixed results. I put 9.1 on a test partition with OS X 11.6 (previous El Cap versions turned my iMac into a brick. 11.6 is the first to install successfully, but that's another story)

I first tried a straight loopback. A ping shows 1226 samples using a buffer of 512. So that sort of makes sense. But ping again, and the result is zero. Ping several more times, and I usually get zero, but often 1226 and now and then something a little bigger. Playing back something gives incorrect timing. So either 1226 is wrong, or it's not compensating correctly.

Next I ran the loop through a digital tape deck on pause and recording armed. The ping gives 1273 or zero, like the above test, but again, if the result is not zero, it's usually 1273, and now and then 1650. I changed the buffer to 64, and the ping changed to 377 (should be 374, shouldn't it?). Playback sounds fine and in sync.

I tried dragging the insert to another channel strip and DP crashed.

So, better that what you guys are seeing, but still not right.