VCC Question

Ermz

¯\(°_o)/¯
Apr 5, 2002
20,370
32
38
37
Melbourne, Australia
www.myspace.com
Hey guys,

Since VCC doesn't have any 'bus' instances, what are you supposed to use to emulate the 'gluing' effect of routing multiple channels to a bus on the desk? Just another 'channel' instance?

I ask because the Mix Bus instance takes a ton of CPU power.

Cheers.
 
Hey guys,

Since VCC doesn't have any 'bus' instances, what are you supposed to use to emulate the 'gluing' effect of routing multiple channels to a bus on the desk? Just another 'channel' instance?

I ask because the Mix Bus instance takes a ton of CPU power.

Cheers.

That's what I do Ermz. Same with FX returns
 
What's your CPU? I'm running literally over 100 plugs, including VCC on every channel of like an at least 20 track mix, and not having CPU issues. The MixBuss is the bus instance.
 
Yep, around 140 at the moment, and might grow on other songs.

Thanks guys, I'll just use the channel instance on buses. I'm thinking in the future I'll print the individual channel instances directly to the audio tracks, much as is my workflow with Nebula. This way I can get the 8x oversampling in there, and also conserve much-needed CPU power for hungry shit like the half-dozen instances of EchoBoy etc.
 
Yep, around 140 at the moment, and might grow on other songs.

Thanks guys, I'll just use the channel instance on buses. I'm thinking in the future I'll print the individual channel instances directly to the audio tracks, much as is my workflow with Nebula. This way I can get the 8x oversampling in there, and also conserve much-needed CPU power for hungry shit like the half-dozen instances of EchoBoy etc.

8x oversampling in VCC is quite extreme and muddies things up, Theres a huge thread over on gearslutz on VCC. The last couple of pages slate discuss oversampling and how he himself uses Zero or very rarely 2x oversampling.
Ive dont shoot outs and have always preferred the sound of Oversampling off with VCC just my 2 cents.
 
I don't understand.. why would oversampling muddy things up? Are there errors in the algorithm? Does that mean if you mix at 192k and keep it at zero, it still sounds worse than mixing a 44k?

I just checked the thread and all I've read Steven write is that there are some 'very subtle things' about 4x that he prefers. He feels 8x is overkill. Haven't read anything about anything creating extra muddyness.
 
I don't understand.. why would oversampling muddy things up? Are there errors in the algorithm? Does that mean if you mix at 192k and keep it at zero, it still sounds worse than mixing a 44k?

I just checked the thread and all I've read Steven write is that there are some 'very subtle things' about 4x that he prefers. He feels 8x is overkill. Haven't read anything about anything creating extra muddyness.

I dont know how to describe it, But the high end changes to my ears and its not for the better i feel i'm losing clarity at 8x, not adding mudd. Could just be me describing it wrong once again, But all the times i've bounced a song and A/B'd the different Oversampling i've always preferred the no oversampling option.
I'd try it for yourself and see what you find i'm curious if your thoughts would differ.
 
Tried it out quickly just then.

Bounced an instance of my in-progress mix both with and without 2x oversampling on the mix bus instance (all my CPU can handle in this session). The no oversampling version has harsher, and somewhat more disconnected sounding highs. The oversampled one, for lack of better word, sounds more 'analogue' to me. This was most obvious on the crashes, which sound more piercing and less 'round' on the version lacking oversampling.

I think this is what I may have been missing. The whole time I've been using VCC on this project, I've just not felt 'it'. I think it could be the shitty handling of high frequencies at 44.1kHz. For the next project I'm thinking to print all the channel instances at 4x directly to the audio, and use the left over CPU power to run the mix bus instance at 4x, along with 4x channel instances on all buses.

To be honest, the 'mud' you might be hearing on the higher oversampled versions is what analogue gear sounds like to me. It all creates that veil, and surreal quality to the highs, which is a huge part of what constitutes a good sounding mix to me.

Can't wait to try a more extreme 4x o/s vs no o/s version!
 
Prepare your wallet for ivy bridge :D

Missed this one!

I'm going to give Ivy Bridge a miss, more or less, and likely invest in Socket 2011. That seems to be the one they're gearing at workstation use etc. Finding it odd that they're staggering development so that the consumer chipset (Ivy) gets the newer architecture. Don't understand what they're thinking.

From what I hear from some tech/review contacts, the hexacores on the 2011 platform are quite something. I think it's also set to support 8 core Xeons.

The 3960k they've just released on it at launch is quite underwhelming. Performs only a touch better than the 980X. Hoping as it develops we get some better options. In the mean time I might see if I can grab a 2nd hand 980X, since the enthusiasts will be dropping them.
 
Ermz dude i think it'd be a good idea if you could put up examples of what you've mentioned with the no /os vs o/s at different stages x4 / x8 etc.
Would be very interested to hear what it is you're hearing on your side.
 
If you're using 2x oversampling, then with the project at 44.1khz, internally the plugin will run at 88.2khz and then downsample to 44.1.khz at the output.

So realistically, extra oversampling will not add mud. In fact it would result in clearing up some of the low-end, because there will be much less aliasing (if it's coded right!) and less Nyquist wrap-around.

Running at 4x oversampling is enough for most things to be honest. It's all a question of taste though. With the DCAM Synths we make, I tend to prefer running then at 2x oversampling, because anything higher makes the filter much less saturated - and I liekz ma saturationz.
 
+1 for oversampling.

Mixing at x1 and then hitting off-line render at x8; the results always put a smile on my face. Almost like someone else changed your mix to the better.
That said, VCC is not the only plug that upsamples in my projects.

As for workflow, I am leaning towards the multi bus/console, RC-tube on the master approach ATM.
 
Good to see this is being talked about since a few of us are unsure. For me, I run the channel strip on each bus and the mixbus on the master bus (sometimes an additional mixbus for the sub-bus of an instrument ie:drums, etc).

I'm curious on how the oversampling works with the rending options. I usually mix at 1x and choose the 4x or 8x for offline rending. Is this applied when you "bounce" in PT?
 
Missed this one!

I'm going to give Ivy Bridge a miss, more or less, and likely invest in Socket 2011. That seems to be the one they're gearing at workstation use etc. Finding it odd that they're staggering development so that the consumer chipset (Ivy) gets the newer architecture. Don't understand what they're thinking.

From what I hear from some tech/review contacts, the hexacores on the 2011 platform are quite something. I think it's also set to support 8 core Xeons.

The 3960k they've just released on it at launch is quite underwhelming. Performs only a touch better than the 980X. Hoping as it develops we get some better options. In the mean time I might see if I can grab a 2nd hand 980X, since the enthusiasts will be dropping them.

Yeah not a bad shout that. I think I will be staying on 1366 until after the next tick or tock thats for sure as this machine is doing the business for the moment. That said, once the hype kicks off good and proper I might have problems resisiting as usual :lol: Im still yet to buy into VCC or Alex B's nebula banks. Can you see VCC becoming a replacement for Nebula? I remember Alex B insinuating that the response of VCC was based on his original broken code. I dont know how much of that was him being pissed at slate for becoming a competitor though , but I never heard him try and take it any further so can only assume he was bitter.

P.S

Let me know if you are thinking about a hex core to tide you through- I've a spare x5650 that might pique your interest.
 
Just saw this on gearslutz and thought it was good info straight from Fabrice

"I have a question for Fabrice/the Slate team about the use of VCC on subgroups:

In a session I have two subgroups (drums, other instruments) going into my mixbus. Until now I used another instance of the Mixbus on each of those groups. (So each track goes through 2 mixbus-plugs in total.) Is this the intended use or would you rather put a Channel on those groups and the Mixbus only on the Output-channel?

Thanks!

Personally I would use the Mix Buss on those groups, because you will have more of the summing effect when using [Group VCC Channels] => [Group VCC Mix Buss] => [Final Stereo VCC Mix Buss].

I have found the VCC Mix Buss to be absolutely amazing to stack up over several mix groups, you can really control how you tighten up all individual groups, and then the final mix.



Fabrice"
 
Tried it out quickly just then.

To be honest, the 'mud' you might be hearing on the higher oversampled versions is what analogue gear sounds like to me. It all creates that veil, and surreal quality to the highs, which is a huge part of what constitutes a good sounding mix to me.

Can't wait to try a more extreme 4x o/s vs no o/s version!

I think you nailed what i'm hearing, VCC isn't the only saturation plug-in i'm using so the extra added analog-ness in oversampling might just be pushing it over the too much ledge. Ill have to play with this more though to be honest. Loving the RC Tube on mix bus and guitars.