Even summing with a mid-range mixer beats ITB

[davetron5000]

I see the summing discussion come up a lot, and dedicated summing mixers are quite pricey. I have an Allen & Heath MixWizard I used for preamps and EQ during tracking and decided to see what external summing was like, compared to ITB with ProTools. I also was able to compare the 002R's D/A converters to the Behringer Ultragain Pro-8's.

The difference between ITB summing and OTB summing is amazing; I can't see summing inside Pro Tools LE ever again. The difference between the 002R's D/A converters and Behringer's is not as drastic (though to my ears, the 002R's sound better, which is what you'd expect, compared to the B-word)

Take a listen:

http://www.naildrivin5.com/protools/summing

[Slim Shady]

hard to tell as the mixes aren't equal. Drum OH's are much louder in your "OTB" mix, as are the rhythm guitars and most people equate louder=better. Most likely despite your faders being at zero there are some gain mismatches between your A&H channels which caused the volume difference.

There's also a small amount of phasing in your OTB mix (the Drum OHs again) when monitored in mono which the ITB doesn't have.

[Slim Shady]

Also what was the reverb situation? Is that on a track in PT or was it an aux send from your A&H board? Sounds like there's significantly more of it on the OTB mix.

[davetron5000]

Yeah, I noticed some weird phasing on the ITB mix. The faders in PT were not touched (they were used to control levels for the mix, but were the same for all three mixes); I used sends to route the signals to the D/A converters, two sends per channel (one to the 002R's one to the Behringer's). These were all at unity. The A&H mixer channels were all at unity as were the mic pre trim pots (at least as close as I could manually make them).

The final mixes were dithered to 16bit/44.1 using POWr dither type 2 after having been very slightly limited by the Massey limiting plug (set to do around 1-3db of GR and slightly-below-0dbfs for output).

When I get home, I'll check into the ITB mix, but I wasn't totally sure how to make the test a fair comparison (this is why I routed the ITB out through 2 channels of the A&H, into the RNC, into the 1272, so the only difference would be the actually summing of the stems)

[davetron5000]

Oh, the reverb: there is a mono digiverb send that the snare and guitar solo use; this is routed to the drum buss. Since the guitars are using Amplitube, the amp models probably have some reverb on them as well as some slight ambient room sound. I don't know that I really like how the guitars sound, but the ITB vs. OTB seems to indicate there is more clarity overall and better defined bass in the OTB summed mixes (as well as better stereo imaging, though as you point out, there might be some weird phasing problem that the OTB sum avoided).

[Severe]

Ahhh...yes, davetron...

You've come to the other side.

I've been using a Soundtracs board for my mixing for some time now. The differences are huge. However, you're not keeping your levels at unity in PT? If not, try actually mixing at the board.. not just summing through it. I use a hybrid approach. I feel that a big part of my end result couldn't have been possible w/o a few of the plugs I use within PT and a number of outboard gear I'm using through the board. Best of both worlds to me.

Setting this up could get a bit expensive, with all the cabling, patchbays, etc. But I guess it's up to the user to decide if it's worth it to them.

I agree, I could never go back to mixing only ITB... for rock and the like, anyway.

[Slim Shady]

I don't know that I really like how the guitars sound, but the ITB vs. OTB seems to indicate there is more clarity overall and better defined bass in the OTB summed mixes (as well as better stereo imaging, though as you point out, there might be some weird phasing problem that the OTB sum avoided).

really? wow, I disagree. I find the bass to be much muddier in the OTB mix (I found myself wanting to EQ it badly while listening), but the harmonic distortion of the high end is very pleasant vs. the ITB mix. The OTB mix does seem livelier and wider, but again muddier in the low to low-mid area, and overall louder so it's really hard to compare exactly.

I generally work with film music so I avoid analog mixers like the plague with all my clients do to the amount of noise floor they add when working with low-volume samples, but it's interesting to hear what a relatively inexpensive mixer sounds like when running loud material through it where the noise floor isn't an issue. If I get a chance though, next time I'm at one of my big film clients studios I'll do a quick bounce of a Gigastudio going through a Mackie vs. the same thing going digitally vs. ADAT and post it, the amount of noise you're adding to your signal by going through a mid-level mixer might make you cry.

[Naagzh]

The difference between ITB summing and OTB summing is amazing; I can't see summing inside Pro Tools LE ever again.

I noticed that the mixes were hardly different when compared to each other in mono, yet the stereo differences are drastic.

What's the pan law on the Mix Wizard? Is it the same as PT's (-3)? The guitars are so much louder on the sides in both your OTB mixes that something seems awry with the setup. If the pan law on the Mix Wizard is different, then that would explain A LOT. Or if the inputs on the A&H for the guitar are a bit hotter than the others. It might be good to switch up the inputs on the A&H and see if you get the same mix.

[Severe]

Not to disregard your guys observations, but to me it's not even a question of the guitars or bass being muddier or louder. The OTB mix, to me, has qualities that can't be achieved by simply turning up the volume on a particular instrument.

Even with subtle differences in panning and/or amplitude, the effects of the OTB mix are obvious. I would think that one could hear its imprint even with some matching discrepancies from the ITB mix. The first time I tested this, I tried to match an ITB mix of the same material without using any outboard gear or board EQ... just running it through and attempting to match pan and volume. The differences where that big, before any other outboard processing. The 2buss alone was an instant improvement in depth, width, tone, etc.

[davetron5000]

Yes, I hadn't thought of that. The 002R outputs were going to channels 1-8 and the Behringer's going to 9-16. After reading this article

http://www.harmony-central.com/artic.../panning_laws/

I had really no idea that panning worked the way it does. The pan law of the MixWizard could indeed be the cause of the drastic difference.

As for the bass, I'm listening right now using (B&O) headphones; my monitors (cheap Fostex), I believe, inflate the bass and do not accurately reproduce low frequencies, as there's a general lacking of low end and punch on my headphones as compared to what I heard over my monitors. There, the kick drum felt more lost on the ITB mixes and that, plus the way my monitors reproduce low end, it felt better defined to me.

Maybe I need to work on my mixing and room treatment before worrying about summing....