Anyone using Focusrite LiquidMix yet?

[kritikon]

I was going through some spam today, and decided to read one from my nearest pro-dealer, all about the Liquid Mix.

And you know...it doesn't half look quite useable. I've still not bought my UAD card yet (now I've got a massive Visa bill from my trip back to Blighty to pay off, so it'll have to wait... ). But I thought I'd d/l the Liquid Mix manual to see what it does, and it looks like it could be one to buy instead of the UAD card. It's around about the same price...looks like it might possibly have a tad more CPU grunt under the lid, and has some impulse of some right tasty gear. It has definite advantages in that you can edit it from the h/w rather than diddle about with a mouse all the time (I really like that idea...I hate mice), and looking at the interface - it looks pretty well laid out.

From the list of impulses, it does most of what you can get on the UAD plugins - LiquidMix does Pultec Eqs, LA2As (in fact, it does the LA3A and LA4 as well), Fairchild (both UA and original Urei). The only thing the UAD does that I really really want is the EMT plate reverb.

The full list of impulses on the Liquid Mix is:

Compressors

API 2500
Avalon VT737SP
BSS DPR402
Chiswick Reach
Drawmer DL221X
Dbx160S
Dbx165
Empirical Labs EL8 Distressor
Fairchild 670
Focusrite ISA 115
Focusrite Green Channel Strip
Focusrite Red7
Manley Slam
Manley Variable Mu
Joe Meek SC2
Millenia STT1
Neve 2254/A
Neve 33609/B
Neve VR
Pye 84 4060/01
Smart Research C2
Solid State Logic FXG384
Solid State Logic SL4000G+
Solid State Logic SL510
Summit DCL200
Summit TLA100A
Teletronix LA2A
TL Audio C1
TubeTech CL1B
TubeTech LCA2B
Universal Audio 1176LN (re-issue)
Urei 1176LN (original)
Urei/Teletronix LA3A
Urei LA4

Eqs

Neve 1073
API 550b
API 559
Pultec EQP1
Pultec MEQ5
Focusrite VoiceMaster
Focusrite ISA 115
Neve 1058
API 550A
Avalon Vt747
Pultec EQH2
Ear 822Q
Solid State Logic E-series
Solid State Logic G-series
Manley Massive Passive
AMEC Anglela
Avalon Vt737
Avalon 2055
Focusrite Bass Factory
Chandler EMI Passive TG channel Abbey Rd


Not a bad list, eh?

But of course, they might not sound as lively as the UAD plugins, as they're all dynamic convolution rather than the usual algorithms. But they might. I'm a little hesitant to wonder how a compressor can perform via dynamic convolution (although I've got a few Eq impulses that work very well indeed). So I'm sold on the Eq side of things, but I'll need to actually hear one in action before buying, to hear how the comps perform. It has 32 channels, which seems more than enough, you can custom make your own 7band Eqs with mixes from various bands of desks or Eq units (sounds like a great idea). It loses some grunt when you use high samplerates (but I can't see myself using 192KHz anyway...my PC won't perform at that level, so it's a moot point to me...). It's quite rare I'd even go upto 88.2, which still lets you use 8 (if I remember correctly) channels.

Another worry is the latency thing...it says it's only Firewire, and has 2-3mS added latency due to that. Does that mean latency on top of the host's PDC, or can the host compensate for it?

According to the blurb from my mail...it's still only Mac, no PC version until October (or is that out of date now...is the PC one out yet?) So that gives me time to save, and time to consider and gather some opinions before I go out and demo it. But it could be a purchase instead of the UAD. (Just thought of another problem...don't think my PC has Firewire... Need to check). I'm not in a hurry to upgrade or buy another PC yet.

Anyone using/used it, then? How does it compare directly with the UAD plugins (i.e. the LA2A, Fairchild, Pultech Eqs). Close, close enough not to bother, or completely different. Does it actually bring the old h/w flavour to clean digital recordings, or does it just sound clean? And does it really do the h/w behaviour thing, where louder inputs bring out better character (usually...).


Hmmm....I wasn't in the market for a big purchase yet, but it really is tempting on 2nd look.

[kritikon]

Oh yeah...I just noticed there's an upgrade card available, which appears to double the CPU available. So you get 4 channels at 192KHz - which presumably means 64 at 44.1....way more than I'd ever likely need (and there's always that useful little "render" button on my host anyway.)

But looking again through the list of plugins available. I think I'd most likely use it and render for individual channels, then use it again live for the mixing. Especially as there are so many options - lots of master comp/limiters, masterEqs, and some that seem to have been specifically made for buss compression.


Noticed another drawback though...no sidechaining as yet. Dunno if that will be introduced through a simple OS upgrade to d/l or not. Also...doesn't look like there's any option for splitting a signal like with the UAD - the Cambridge Eq has a good set of selectable filters, so you can split a signal and send on to different busses to effectively do multiband compression. Although, technically you should be able to do the same trick with the Liquid Mix as well (but you'd need to know what type of filters the Eq units used to get the right effect without too much phase nonsense etc...)

[Kingston]

I've used its predecessor and the big brother, the sintefex 8000 (same models, samples, algos, and tech)

It's a rather good deal, coming with the controller and all, and certainly a nice piece to load off tasks from the main CPU. But just one thing has to be remembered: none of the pieces of that list sound quite like the originals. It's not comparable to UAD-1 as such, as UAD generally nails whatever it emulates. The sintefex process has a very distinct sound of it's own, that seeps through from all the models. It's not a bad sound at all actually, and very high grade, usable anywhere, but just use those preset names as guidelines. The EQs can be a bit so and so, depending on the model.

Basically the 1176 sounds like 1176 up to a point, a big part of it the distinct sintefex sound, same with the LA2A. In a way you'll find them all too similar to justify the model names.

Anyway, it's still a good deal.

[jupiter8]

http://gearslutz.com/board/showthread.php?t=77611

Although i don't think you get 64 ch. with the expander.
As i understood it,it is only used on higher sampling rates for some reason.So you get double plugin count on 96 and 192 but not for 44.1 for some reason.Could be that the firewire can't take that many channels.

[kritikon]

Hmmmm....that's what worries me. The UAD appeals specifically because it does give you the character of the originals, warts and all (and it's the warts that sometimes set them apart).

I think I'll need to set up a direct A/B comparison when I go in and listen to it. Maybe it's just the convolution aspect that makes it sound a bit samey (or flat might be the word...certainly reverb impulses, although better than most s/w natives, are still a bit lifeless)?

[Kingston]

Hmmmm....that's what worries me. The UAD appeals specifically because it does give you the character of the originals, warts and all (and it's the warts that sometimes set them apart).

indeed. hence the thing is still on my list as well. this focusrite is of less interest to me personally.

I think I'll need to set up a direct A/B comparison when I go in and listen to it. Maybe it's just the convolution aspect that makes it sound a bit samey (or flat might be the word...certainly reverb impulses, although better than most s/w natives, are still a bit lifeless)?

Well it's *dynamic* convolution. And sintefex is the only one allowed to do that due to patent slyness. And it does sound great with the best of the models. Problem is the compressor algo. Only the transfer curve and the hardware go-through impulses are dynamic convolution, the rest is done using the same exact algorithm for everything (attack and release shapes, peak/RMS, program dependency, all the really important stuff). The EQs are fully dynamic convolution. They have a tendency to sound a bit less accurate (in a digital, not analog way) than they should, and can be a bit on the harsh side.

[kritikon]

Although i don't think you get 64 ch. with the expander.
As i understood it,it is only used on higher sampling rates for some reason

Hmmm...that seems a bit of a daft idea to me.
TBH, at 192KHz, I would imagine it's be used mostly as a fancy master comp/limiter/Eq. Then to offer an upgrade DSP card that doubles only the high samplerate number of plugins...I just can't see why anyone would need extra high quality busses, but only two. Surely it would be one buss for mastering or 8 busses (or maybe 4 at a pinch) for final mixing. Whereas the extra grunt would most likely be more useful to most users at the channel stage.


Maybe Focusrite just haven't got their heads around the in-the-box DAW thing yet...or let me guess...the Saffire interface has 4 outs maybe?

Initially...there seem to be not enough pros for Liquid Mix, whereas I can already see some cons. The proof is in the hearing, I suppose.


Anyway...thanks for the replies so far...keep 'em coming...I need all the justification I can get to stop me from splaashing out on a new toy that I can ill afford.

[kritikon]

Only the transfer curve and the hardware go-through impulses are dynamic convolution, the rest is done using the same exact algorithm for everything (attack and release shapes, peak/RMS, program dependency, all the really important stuff)

Booo...


When you put it like that...yes, it does seem a weird idea to have convolution (even dynamic) to emulate something as quirky as a tube compressor.

[Kingston]

as quirky as a tube compressor.

well tube sound it kind of manages, but not the eccentirities of vari-mu or opto sidechains that generally make up the tube compressor sound. VCA, being linear and all, it manages just fine. Problem is really in the attack and release shapes and speeds and program dependency, which is why none of the models quite deliver.

But there's still no reason to get upset over that. What sintefex does within those limitations still sounds very good and usable, most of the time on par or superior to our best VST plugins. I think you'll be pleasantly surprised upon hearing it the first time.

[bduffy]

So are you guys saying the UAD-1 is the only good emulator of the 'classics'?

[Kingston]

So are you guys saying the UAD-1 is the only good emulator of the 'classics'?

I don't know of the others here, but I think so, yes.

[bduffy]

So are you guys saying the UAD-1 is the only good emulator of the 'classics'?

I don't know of the others here, but I think so, yes.

The degree to which I should get one of those is just silly...ROLAND CHORUSSSS....

[DevonB]

Another worry is the latency thing...it says it's only Firewire, and has 2-3mS added latency due to that. Does that mean latency on top of the host's PDC, or can the host compensate for it?

Which is still better than the double latency you get from the UAD-1 (ie if your host is 6ms, with the UAD-1, its' 12ms.)

Devon

[Stupid American Pig]

So I take it that the powercore is just no good then for fx. The powercore models many of the same efx as the UAD card, how does *that* compare?

[bduffy]

Yeah, I'd like to know about the Powercore too. TC make some killer effects, but I suspect they're more of the pioneer-type designers, not the 'vintage' type. Love their TC Native Bundle, too bad they dropped VST.