Studio One with Outboard Summing Boxes?

[jstudio]

Hi, I'd like to find out how to correctly set the latency compensation when using summing boxes?

So that's 16 outputs (sends), with only a stereo return.

But Pipeline XT only has a stereo send & stereo return.

I wanted to use Pipeline XT for the Summing channel's stereo return signal.

To calculate the correct latency I would add one of the Summing sends (so only 2 out of the 16 summing possibilities - just to calculate the correct latency).

Then I wanted to remove the Pipeline XT send and have "None" there - only having the summing return signal actually going into the channel.

However, when using Pipeline XT, after adding the send (so e.g. going from "None" to "Main - 1 + 2"), I am unable to get back to "None" = no output/send.

I want Pipeline XT plugin to just return a signal onto a channel, with latency compensation.

If this is not possible with Pipeline XT, how do I add the correct latency compensation to the summing stereo return channel?

 

[dean.winter]

Hi jstudio,
I do something similar with 16 channels of summing and can try to help out. As far as I can tell you can't do what you're trying to do with Pipeline. It seems Pipeline is designed to function as a combined send/return, and once the output or input is set you cannot shift it back to "none" (or at least I couldn't find a way). I'm not sure how to go about making automatic delay compensation work for the use case of a summing box.

That said, my solution is to just not worry about delay compensation. Delay compensation only matters when some signals are delayed and others are not. If everything is delayed by the same value then there's no need to compensate for anything. So for my use case everything just goes out to channels on my SSL and then a I have a single stereo track whose input is the output of the SSL. So all my tracks go to busses, the busses go to individual channels of the SSL, they get summed, EQ'd, and compressed, and then go to a print track. Because everything is making the same round trip journey I don't have to do any latency compensation.

I do use pipeline when I want to send to some outboard gear as an insert on a channel or something, just not for summing.

So my console might look something like this:

Note the "SSL x/y" outputs in the dropdown and the red "SSL MIx Out" track on the right with monitoring enabled. That "SSL Mix Out" track is the only thing going to the main outputs, and when I'm ready to mixdown I just record arm that track and record a pass. If I want a new mix I duplicate the "SSL Mix Out" track and record a new pass and so on.

Anywho, I also realize that you might have a different use case....maybe for you there are some tracks going to the summing box and some that aren't so you need latency compensation? In any case, if you can describe your use case I can try to help.

Dean

 

[lokeyfly]

Hi jstudio,
....... Anywho, I also realize that you might have a different use case....maybe for you there are some tracks going to the summing box and some that aren't so you need latency compensation? In any case, if you can describe your use case I can try to help.

Dean

Exactly my immediate reaction prompted the question, why jstudio [OP] cares about latency when summing in the 1st place.

At least for most summing needs (and we may be missing something), summing from some external device or hardware shouldn't require tracking in real time. So why the latency concern at all?

Even from a live or Studio scenario, simply capture and print the summed outputs or mix. Only then afterwards, proceed to work within Studio One with latency being little concern.

Where is the need for combining multiple tracks into a main mix here? Would be great to know more about the process to provide the best answer. Appreciated.

Thanks!

 

[SwitchBack]

Mixing in analog hardware can add character to a mix (saturation, slew rate limitation, bus power clipping, noise floor, ...) compared to digital mixing which is basically an arithmatic sum. It would be interesting to compare the result of the analog mix with what can be accomplished with some of the Mix FX models available in/for Studio One. If the same or even a better result can be achieved with Mix FX, then some of the problems mentioned in this thread won't be.

 

[lokeyfly]

Mixing in analog hardware can add character to a mix (saturation, slew rate limitation, bus power clipping, noise floor, ...) compared to digital mixing which is basically an arithmatic sum. It would be interesting to compare the result of the analog mix with what can be accomplished with some of the Mix FX models available in/for Studio One. If the same or even a better result can be achieved with Mix FX, then some of the problems mentioned in this thread won't be.

Oh, there are definately benefits to the hardware analog summing mix, or as you say hearing those differences between the analog mix and Studio One's summing/effects as a kind of shootout. I just dont see why latency is of any concern (at this juncture) unless there's some detail that hasn't yet been explained. One needn't be tracking to some summed ideal sounding mix. Where latency is more of a concern.
Hoping we get the latency reasoning from jstudio.
By all means, the summing quality comparison can be followed up and sounds interesting.

 

[jstudio]

Thanks everyone - I think you're correct. I should be able to work with the stereo signal without having to worry about the latency, now that I changed my summing box to Dangerous Bus+.
I had in my mind about the Neve 5059, which has 2 sets of stereo outputs - where I was going to do a rear bus technique. I would have needed those both to be in sync, where there was different external processing/routing going on.
But I have changed this now anyway, and didn't think about it all properly. Should be fine now actually.

 

[dean.winter]

Thanks everyone - I think you're correct. I should be able to work with the stereo signal without having to worry about the latency, now that I changed my summing box to Dangerous Bus+.
I had in my mind about the Neve 5059, which has 2 sets of stereo outputs - where I was going to do a rear bus technique. I would have needed those both to be in sync, where there was different external processing/routing going on.
But I have changed this now anyway, and didn't think about it all properly. Should be fine now actually.

Hey jstudio - glad to hear you got it sorted. If I can help out with anything feel free and hit me up.

Dean

 

[dean.winter]

Oh, there are definately benefits to the hardware analog summing mix, or as you say hearing those differences between the analog mix and Studio One's summing/effects as a kind of shootout. I just dont see why latency is of any concern (at this juncture) unless there's some detail that hasn't yet been explained. One needn't be tracking to some summed ideal sounding mix. Where latency is more of a concern.
Hoping we get the latency reasoning from jstudio.
By all means, the summing quality comparison can be followed up and sounds interesting.

Hey folks - since there seemed to be some interest I whipped up a quick A/B to demonstrate what the SSL does here. Note - I'm playing all the instruments here and they're all pretty much just one quick take, so please be kind and don't expect any particular performance virtuosity. I also didn't do much level matching...it's pretty close, but adjust as you see fit. Oh, and I flipped phase on one of the files for easy null testing.

ITB (no summing) - https://drive.google.com/file/d/1YDjXEdHuWjARSsNObbkpE36c-IsivOnE/view?usp=sharing
OTB (SSL summing) - https://drive.google.com/file/d/1nOZSvAk7LdVYACdS1MrMk1gbdaS-8Dlo/view?usp=sharing

I'd suggest downloading each and dropping them into S1 for easy A/B'ing.

Some notes - the project has 11 channels of drums, 1 of bass, 2 of guitar, and one virtual instrument for the e. piano. Drums were tracked through the 1073s and the ISAs, bass was DI through a Sebatron tube pre with a bit of grit, guitar was a 57 and 121 through Great River pres. There's a bit of EQ on the overheads, and compressor on the bass, but nothing else. Tracks are routed through 14 channels of the SSL, then through the SSL Bus Comp and SSL stereo EQ. SSL Comp was engaged but the threshold was all the way up so no compression was happening and the SSL EQ was flat (so signal traversed them, but they weren't doing anything). I would have pulled them out but I didn't want to re-patch.

To my ears the way I'm driving the SSL inputs is making the sides come up a bit and is generally rounding off transients, particularly on the snare. Threre's also some really nice saturation on the low frequencies. That said, YMMV. We can all like different things so if you prefer the ITB version then more power to you

Dean

 

[SwitchBack]

Hi Dean,

The fun test would be to see if you can make Studio One sound like the SSL console/mix bus by using the Brit Console Mix FX plugin. That plugin is said to be modeled after the SSL 4000E/4000G/9000J consoles. I certainly don't want to spoil the fun of using proper analog hardware. Just see it as another (more portable) tool in the box.

 

[Craig Anderton]

I was curious about the MixFX, so I created a test setup to isolate only what the MixFX added to the sound (i.e., summing two channels, one dry and out of phase compared to one with the MixFX). The various MixFX offer a wide variety of sounds. The serial number is even an Easter Egg variable for obtaining variations. One "non-main bus" application I like is using MixFX to add crosstalk with drum buses.

I don't have any of the consoles on which the MixFX were modeled, so I can't vouch for their accuracy. However, one CTC-1 model did recall some of the mixers I used back in the 60s. I wrote up the test results for the PreSonus blog, but I think it was lost in the Blog Post Extinction Event.

Transformers seemed to me like they were a big part of the "analog" sound. So I was happy to see that when looking on Scope to see what the MixFX did to square waves at low frequencies, the waveform had the overshoot and ringing you'd expect from an input transformer.

 

[dean.winter]

Hi Dean,

The fun test would be to see if you can make Studio One sound like the SSL console/mix bus by using the Brit Console Mix FX plugin. That plugin is said to be modeled after the SSL 4000E/4000G/9000J consoles. I certainly don't want to spoil the fun of using proper analog hardware. Just see it as another (more portable) tool in the box.

Hey SwitchBack - let me see if I can get a few minutes to try out the Brit Console. Might need a while though, I have a head cold and have entered the phase of "ears super plugged up, everything sounds funny and my stereo imaging seems way out of whack", so my ability to listen and make decisions might be real bad just now.

Alternately, if you want to try it out I can send you the project/tracks.

Dean

 

[dean.winter]

Hi folks (and particularly @SwitchBack )
OK, so I'm a giant nerd and did some work with the Brit console emulation to see if I could get it to sound like my SSL X-Rack.

Short answer - No. Maybe sort of, but still no. The console emulation wasn't bad, but it wasn't the same.

Longer answer -

First I moved through the 3 emulations on the Brit Console to find the one that I thought was closest and settled on the "Modern" emulation. Then I level matched as best as I could. Below is the loudness graph over time, blue is ITB with console emulation, Orange is OTB through the SSL. Note the for the majority of the track the loudness levels match up tightly, but at the beginning and end (where the levels are peaking louder) the blue line gets a bit higher. I *think* this is due to the way I'm driving the hardware. The hardware is limiting the peaks a bit. I tried to replicate that in the console emulation by pushing the drive, but couldn't quite get there. Also, pushing the drive started causing other differences.

I also couldn't replicate the stereo imaging that was coming out of the hardware, no matter what I did with the crosstalk. Here's a graphic showing the differences (again, Orange is SSL, Blue is ITB with console). Note how the Orange lines are wider, particularly through the mids and upper mids:

And next, frequency distribution averaged over the length of the track. Again, note the differences:

Maybe most interestingly - when comparing dynamic range the Hardware generally had greater dynamic range than the console emulation. That kinda surprised me, particularly given the loudness graph above. Note - the lower the line is the less dynamic range is present. So if I were to slap a limiter on something and crush the crap out of it, the line would be very low in the graph.

Finally, just listening to it the results were still quite audibly different.

Anywho, hope someone finds that interesting. It was a fantastic way for me to avoid doing any actual studio work on a Saturday morning, so.....mission accomplished??

Dean

 

[SwitchBack]

Great info! I was curious especially about the listening experience (I hope your beat the cold), but the technical stuff is very insightful too. Still a bit to do for PreSonus

NB. Looks like something is triggering round about 17.5dB in the ITB dynamics graph. Strange.

 

[Yluko]

Hiya, just joined a few minutes ago, first post I saw. Interesting thing is I just added analog summing about three weeks ago to my setup. I just send everything I make automatically to the last stereo channel to avoid the delay from latency. Works fine,as far as the sound goes, omg, no plugs do what this thing does, I know most of it as the stereo transformer at the end but wow, if I’d known the big difference it would make I would have gotten one years ago, it just feels closer to a finished record and makes me smile. I got a phoenix audio nicerizer , allot more colored then a d box or ssl. Clients are blown away when they came over and I showed em the difference. Loving it.

 

[Jemusic]

A few years back I saw an article that compared digital summing internally to passive summing externally. They showed these very complex spectrum waveforms etc and they were both identical. Interesting. Its more than likely the transformer output stage that is adding the sound you like. Great sounding mixes can be done all internally but that depends on the mix engineer. And with the plethora of plugins now that can emulate consoles and transformer stages, valves etc it is more likely now you can get a great sounding mix without any external summing.

It is also interesting that doing some things externally to digital signals can also improve the sound. If I send a bunch of stems from Studio One to my Yamaha digital mixer digitally and mix the stems there it always sounds wider to me for some reason. Compared to the all internal Studio One mix. A recent article I found by an engineer who recorded a late James Taylor album experienced exactly the same thing. He recorded it all digitally but preferred the mix of the stems done on an external digital mixer for the final result.

 

[dean.winter]

Great info! I was curious especially about the listening experience (I hope your beat the cold), but the technical stuff is very insightful too. Still a bit to do for PreSonus

NB. Looks like something is triggering round about 17.5dB in the ITB dynamics graph. Strange.

Oh, yeah...I shoulda posted a link to the mix with the console emulation.

FWIW - I wasn't making any particular effort at setting levels on this track and as a result the bass git (and low freq's in general) are a bit too hot. I think the Brit Console emulation has exacerbated this problem a bit, but YMMV.

Anywho - here it is: https://drive.google.com/file/d/1PMux0IY9ZIgkeYpfkgaeJWrD08qBny3u/view?usp=sharing

Dean

 

[dean.winter]

A few years back I saw an article that compared digital summing internally to passive summing externally. They showed these very complex spectrum waveforms etc and they were both identical. Interesting. Its more than likely the transformer output stage that is adding the sound you like. Great sounding mixes can be done all internally but that depends on the mix engineer. And with the plethora of plugins now that can emulate consoles and transformer stages, valves etc it is more likely now you can get a great sounding mix without any external summing.

It is also interesting that doing some things externally to digital signals can also improve the sound. If I send a bunch of stems from Studio One to my Yamaha digital mixer digitally and mix the stems there it always sounds wider to me for some reason. Compared to the all internal Studio One mix. A recent article I found by an engineer who recorded a late James Taylor album experienced exactly the same thing. He recorded it all digitally but preferred the mix of the stems done on an external digital mixer for the final result.

Hey Jemusic
I agree 100%. If you can't get a good mix ITB it's probably something lacking with the engineer, not the gear. And if I really wanted to, with a bunch of trial and error I could probably get pretty close to what my SSL is doing with plugins. That said, I know what my SSL does and I can get the results I want pretty fast so there hasn't been much motivation for me to undertake that.

Also....I still really like my hardware

 

[Yluko]

I’ve been doing mixes ITB for over 20 years, clearly can get great mixes ITB, it’s just a different sound. I think one issue is allot of these tests are done with passive summing verse something like an api or the phoenix with allot more going on, op amps on each channel ect. I can clearly hear the difference between it all going to a single stem through the transformer or split out. Though the transformers makes a clear difference. Stuff like ssl is made to sound clean so I think it will have less of an effect. I had a client over who’s newer to audio and set it up where he could switch between all synths summed to one channel or a few spread apart and asked him to tell me the difference and what he could hear, I was curious if he could tell. He could, he could even describe exactly the difference, the interesting thing was in this particular case he preferred the one where they all summed together, he felt like it was more forward and glued and the other way separated it more and he wanted a more glued stack of synths. He freaking loves the transformer sound though. The monitoring is on the ex Machina pulsars mkii which have great detail. To my ears saturation and compression is the one thing plugins when pushed hard fail to do as good as hardware yet, it’s the saturation driving into those transformers I quite like. Not needed just n icing on top I like.