Anyone see this yet?
Black Lion Audio | Focusrite Liquid 56

I'd be interested to know how much better the conversion is. Also if the Liquid pres are affected (improved as well?)

I probably won't be doing a mod on my LS56 anytime soon, but it sure would be nice to know that there is a cheap upgrade path for the future.

What they claim to do with it sounds awfully nice, but vague, and not very extensive...which now that i think about it could reflect good things. I also don't know that i agree with their sentiment that stacking tracks demands corrective eq. I've done 5 months of work in since getting my '56, and i cant say that i've had problems stacking guitar tracks & vocals right on up. Though at $5xx bucks, and the ebay price of the '56 @ $720, a total of $1200 and change for this rocking piece of gear (plus [potential] improvement in sound over stock) is really not bad. I can honestly say i love my '56 and haven't found a thing i dislike about it yet.

-chris

You and me both Chris. I love mine as well. I have seen threads talk about the converters not being as strong as some other gear. And some talk about the preamps also not being. Amazing, but at the price and the features I think it's outstanding. This addition in quality on the conversion will be a nice addition for the future. As for the pres I've already bought some dedicated units.
I'd love to hear what people have to say about the upgrade. :(

My next purchase...

I'm thinking of getting it done too. I love the connectivity of the Liquid 56, but the pres are just sort of meh.

I'm a little skeptical about some of these BLA mods. They sound good on paper. But are there actual improvements in measurable real-world specs? If I'm plunking down several hundred dollars, I might want that...

I don't think I'd have mine modded - I don't need to and it would involve all the Customs hassle of shipping it to/from the US. However, when I looked at the BLA page describing the mod the other day, I was a bit puzzled. According to the BLA site, the LS56 has AKM converters, but I've read in various other places that the converters are Cirrus Logic. Looking at some of the waveforms over in the big AD-DA test thread in the Gear tests sub-forum, the LS56 sawtooth response also looks more like that from a Cirrus converter.

All of which leaves me wondering - did BLA just lift the text on their website from the description of one of their other mods but forget to change the converter details? Or are the various other places/people wrong when they say the LS56 converters are Cirrus? Or something else altogether?

Anyone know for sure which converter chips/chipsets the LS56 uses...?

"Measureable" with "specs" is not a good approach to audio gear. Scientific method wants things as simple and clean as possible, to eliminate variables. But music is inherently messy and variable. Look at THD, everyone's favorite spec when they want to pretend they know what they're talking about. It's measured with a 1khz sine wave. That's like measuring how well a car drives 60mph. MUSIC is made up of ten octaves of frequencies and a 20-60db dynamic range! Measuring "total harmonic distortion" in a way that is meaningful to actual music is nonsense.

A/B testing is a little better, but still not in any way perfect.

Now here's the question... do you think there's an audible difference between budget gear and pro gear? If so, then why? That's probably due to a: circuit differences, and b: parts quality. The mods are addressing both. So it makes sense than they'd show some improvement. Whether the improvement is worth the money is a judgment call, not something that can be measured.

Me too Chris, i read the spec and did question the stacking tracks thing. I use 16 tracks through mine, about to add another 8 and have been really happy with how they stack up

In fact i did a mix sending two stereo tracks summed from 8 monos and the two stereo tracks summed perfectly

I in know way doubt black lions expertise but this hasnt been my experience

Kludge - not sure it would improve the preamps per say, though if the conversion from AD & DA gets improved, I would assume it would capture the preamps in a more accurate way, which may seem like it's improving their function.

Adrianww - seems odd if they are in fact incorrect about the converters being AKM if they actually are Cirrus Logic? At this point we don't know if they are AKM or not right? (anyone care to clarify this) But lets say for some reason they made a mistake in the name. I would be more concerned with how the mod actually affects the quality of the sound. These guys are know by some of the top engineers for a reason. Their modds have been used to make high end stuff, so with that reputation, I'd bet their mod to the Liquid Saffire 56 would be an improvement on the stock unit. How much improvement? That I'm not sure.

Dubmunkey - I haven't stacked many tracks myself, but would be interested in hearing others that do and see if there are a few who have units that do have problems stacking tracks. Do you send out some of your tracks to outboard processing units and then back into your daw. (ex. send tracks to compressor & effect units then back ITB) This process of capturing an analogue signal (AD conversion) then sending it to an outboard compressor for example (DA conversion) and then receiving that compressed signal back (AD conversion again) might degrade the signal a bit more. And when stacking many tracks in this manner, or sending them out to be process in other ways might reflect what BLA is taking about. Just a thought?

Yeah could be.. I've hooked it up to allow me to send tracks out of my daw, into my mixer and outboard and back in but havent used it in anger yet

In terms of just sending in i've found it great although im neither a pro or an audiophile

When i get a mo i could loop some audio a few times and post the result and you can see how it degrades on each pass through the ad/da

If you want to see what the LS56 does to audio looped back through its own DA-AD stages a few times, take a look-see at the aforementioned AD-DA shootout thread in the gear tests sub-forum. I provided LS56 test runs for that thread, as did one or two other folks.

On the one hand, it doesn't look pretty. There are screenshots of a 60Hz(?) sawtooth wave that's part of the test signal in there and it gets mangled to hell.

On the other hand, that still doesn't mean to say it sounds bad 'cos that test is based on looping out and back ten times and then comparing the original to the loopback using Audio DiffMaker. Interesting stuff, but it's not going to make me throw the Saffire away - I don't think that the test is flawed per se, but there isn't a specific point where you can say "This sounds great and that sounds really bad". It's not clear just how poor the DiffMaker results have to be (and in what way) before something sounds obviously "bad". All it really tells us is whether the conversion is transparent (i.e. slavishly accurate) or not. Something could be quite inaccurate, but still sound musical to human ears.

Agreed. I'd like to see how that 60Hz sawtooth wave holds up after the same process into a tape recorder.

I would guess that the actual AD/DA stuff is just fine, and most of the shortcomings are in the analog parts - mic preamps and line-level outs. I know that my FMR RNP easily smokes the mic pres on the 56, especially with low-sensitivity mics like dynamics and ribbons (and keep in mind that, technically, the RNP's output is still going through the 56 pres, just at the lower line-level gain). And I use a vintage Audio Alchemy DDE DAC from the 1990s for my monitors, which sounds better than the 56 outputs despite its clearly dated D/A chip - the win is all in the analog stages, I'm sure.

So at some point, I'll roll the dice on the BLA upgrade. If it can get the mic pres up to the RNP's level for $600, it's a bargain for that alone. That's the cost of two or three channels of good budget mic pres.

Back in college in the early 90s, my recording teacher showed us exactly that test. He fired up a square wave generator, put it on an oscilloscope to show us the accuracy of the generator (which was imperfect itself), then ran it to tape (a beautiful Otari mastering deck), and two different digital systems, and played the results back through the scope. The tape showed a sharp transient at the leading edge of the square wave, a basically flat top, and a softly curved slope at the trailing edge - distorted, yes, but just losing some harmonics. The digital systems looked AWFUL. The leading and trailing edges were perfect, but the supposedly flat top of the wave, which represents the infinite harmonic series of a square wave, was deeply rippled, showing significant changes in harmonic balance. And playing them back through the monitors (given the limits of the monitors at reproducing square waves!), the tape sounded WAY better than the digital.

I'm personally not experienced or technical enough to give it much mind and will probably loop back once per track and only if I need to send it to something outside, I'd welcome any musical imperfections the multiple ad/da does but doubt I'd hear it as I'll be changing the sound radically anyway

for me the 56 is just best bang for buck, I had a mild tizzy when I realised it's FW support wasn't great but the latest release does seem to start addressing that - but in terms of ins/outs, mix control, the liquid pres and the stock pres it's hard to be for the price - having not tested any other interface btw

only problem is I'm about to run out of inputs

this is interesting, how do you feel about the liquid pre's? I may well get a RNP just for an alternative and cos I love FMR

I actually don't use them that much. They have "color" that can sometimes be useful, but they also have a little less gain than the regular pres, and I usually want something clean.

But the liquid pres aren't the reason I got it. I got it for the i/o convenience - no combo jacks! I can keep both XLR and TRS snakes wired up.

I should add that the REAL loser compared to the RNP isn't the mic pres as much as the instrument ins. The RNP as a DI utterly blows the 56 instrument inputs away! I record a lot of guitar and bass direct in, and the difference is not subtle.

ok cool but you also think that if you run a RNP into one of the line you'll retain most of the RNP goodness?

Cool, I still wonder the difference after 10 in/out procedures. I actually love the sound of analogue in many ways and the imperfections it imparts on sound.
As much as I like the pres in the LS56 and the liquid channels, I still bought a Sebatron tube preamp. There's just something about tubes that I like.

I use my Focusrite ISA 428 pre's to the line-in's on the '56 when i need to and the mojo is definitely still there after the trip in. It is well documented that the line-in's bypass the mic pre's, so for all intents and purposes, the line in shouldn't really change the sound too terribly much.

My understanding is that the line ins are still going through the preamp, but padded to a much lower gain (and certainly the gain knobs still work on the line ins). And with the '56 pres, the "stiff" sound starts to set in with the gain around halfway up and the noise starts to set in when the gain is around 3/4 up. At lower gains (like a hot mic or a hot source), they're great. So it's not surprising that there's little Focusrite-side coloration on the line ins - they're not doing what it isn't so good at, which is high gain.

(To be fair, the RNP gets noisy if you push it hard enough too, but it needs to be pushed a LOT harder first)

I was under the impression that you DO BYPASS the Preamps on the LS56 by using the 1/4" line inputs. As per this thread I read before?

By Passing Mic Pres - Liquid Saffire 56

Well, I turn the gain knob when it's on line input, and I get more or less gain. So SOMETHING isn't being bypassed.

Kludge,
I think they are calling it the "Leveling Amp". So there is something controlling the line level, but it's not going though the normal pre-amp which would have some sort of color to it? Not sure if that makes sense of it that's just their way of saying their "preamp" is so clean it won't affect your outboard preamps tone? In any case, it works and sounds good.
I get a nice sound out of my Sebatron in to line inputs on my LS56.

Actually, if you look at the gain range on that levelling amp, it's basically the same as the gain range on the pres, but padded down by around 20-odd dB. I'm prepared to bet that it is the exact same variable gain stage that is used for the mic pre and that the two share the same circuit (which makes sense from a design/build cost point of view). It's also pretty much got to be an active circuit, since it can provide up to 36dB of positive gain, so it's almost certain to have some effect on the sound passing through it.

Having said that, I think some folks here on GS get far too hung up on the whole thing of wanting to bypass pres/levelling amps. So long as they can be set to unity gain (0dB) - which just about all of them can regardless of which interface you're talking about - just set them up, connect your external pre and off you go. The unity point on my LS56 tends to be around 6 or 7 (or thereabouts) on the gain controls so if I don't want it to mess with an incoming line signal too much, I just make sure I run it at around that point. Works just fine. If I ever decide that I really, really have to have fixed line inputs for something, I'll just buy a good external converter with fixed inputs and an ADAT output and run it into the LS56 that way.

Yeah the only gain issue i have is with my sm57 dynamic...otherwise seem to remember the gain on all line ins is at 0

The 0 setting on the LS56 gain controls actually gives you about 10dB of cut (attenuation) on line inputs. The line input variable gain range is -10dB to +36dB. (The mic input variable gain range is around +13dB to +60dB.)

In fact, the one thing that I wish Focusrite had done differently with the LS56 was to mark the gain controls in dB rather than just a linear scale. Even just putting some kind of mark at the approximate unity gain/0dB point (for the line inputs) would have been nice. Of course, if they did that, it would show up just how non-linear the gain controls are, but that's no big deal - most gain controls are pretty non-linear on most interfaces (certainly in the low-end and mid-range market).

My other interface (a Yamaha n12) has the gain knobs marked up in dB (although using a slightly odd reverse scale based on input levels I think) and it can be useful.

Id have to double check as i may be quoting my levels before i sorted my gain staging but would make sense as i dont want 0db on my outboard to be 0dbfs inside my daw