Bouncing with hardware inserts... some observations

[James Steele]

Well I'd typed most of this lengthy post in Tapatalk, which then slowed to a crawl when I was about finished, froze up, and I lost it all. Sigh. Start over on my Mac and web browser.

SO...

I recently purchased a Warm Audio WA-2A. It wasn't packed very well for shipping so I figured I'd better test it as soon as I could. I decided to test it using DP's Hardware Insert plug. I connected Analog 3 OUT on my 828es to the INPUT on the WA-2A, and then connected the WA-2A's OUTPUT to Analog 3 IN on the 828es. After some struggling with Pro Audio Control, I got the routing set up and put an instance of the Hardware Insert on a mono vocal track and routed it out to, and back from, the WA-2A. I did the "Detect" test and it returned a latency of 41 samples. Okay I guess.

I then decided to try and bounce this vocal track with the WA-2A insert onto a new mono audio track. I tried 3 things:

1) Without changing anything, I created a new mono track, and then set the input source to Analog 3 IN... the return of the WA-2A. I put the track in record, then did the bounce. The resulting audio was significantly EARLIER than the original vocal track and I had to manually shift it later on the timeline to correct it.

2) I then figured I'd select the mono vocal track (with the hardware insert plug on it) and do a "Freeze." However, even though my vocal track was mono, DP's Freeze function created a new stereo track and proceeded to bounce to stereo. So I bailed out of that and decided to do a "manual" freeze. I routed the vocal track to a mono bus, then created a new mono audio track, set it's input to the same mono buss and bounced. The audio this time was later than the original by a bit. Not sure if tweaking with the samples setting in the Hardware Insert might have fixed this or not.

3) Lastly, I decided to do something very BASIC. I set the output of the vocal track to Analog 3 OUT (which fed directly into the WA-2A). I then made a new mono audio track and set it's input to Analog 3 IN (to which the WA-2A's output was directly connected), and then put that track in record and bounced. This time... the timing of the bounced audio was, for all practical purposes, spot on. There might have been some round trip latency of a few milliseconds (??) but if it was there it was negligible.

Anyway... the last method seems like sort of the dumbed down way to do it, but might work the best. When using this compressor I suspect I'm going to want to pretty much run the signal through it just like that to print it on a track. I may as well avoid the hardware insert. Feeding it a pre-recorded track will be the exception anyway. Most of the time I intend to track my vocals through it live, so it will be fed by a mic pre, and then be recorded directly into an input on the 828es.

[primeevolutionary]

James,

I always have to assign my mono audio tracks to a mono aux when using hardware insert. The output of the track affects the insert’s mono or stereo assignment.

[James Steele]

James,

I always have to assign my mono audio tracks to a mono aux when using hardware insert. The output of the track affects the insert’s mono or stereo assignment.

Ahhhh, right. So even though my vocal track was a mono track, because it's destination was the main *stereo* master... DP creates a stereo buss and freezes to a stereo track. Derp!! Got it... makes sense. Oddly enough though, I think I'm not going to freeze and just route tracks directly to the input of the WA-2A, and then record a mono track directly from it's output. That seemed to work the best.

I don't know if it's placebo effect, but it does sound as if processing it though the outboard WA-2A (which is an LA-2A clone) does seem to impart some warmth on the track since it is running through some vacuum tubes in the process.

[primeevolutionary]

It certainly does. I typically use four outboard compressors and the character I get from them seems more satisfying than plugins (even though the plugins do a great job). If I had to guess, even the non-lookahead compressors must use some kind of look ahead technology. The results just seem too predictable. The hardware only has the choice of reacting to the signal. It’s slightly too slow and irregular. It’s a new story every pass. I think that’s what gives the outboard compression its liveliness and excitement.

I only record tracks these days and only bounce if I’m converting.