|
|
Post by gravesnumber9 on Nov 27, 2023 17:02:54 GMT -6
How would someone do the equivalent test now? It's not really possible to track live instruments at 48khz and then again at 96khz and then compare them. Sure it is. It's simple. Split the mic pre signals, and go to one DAW at 48k, and a second DAW at 96k. Use the same converter for each. Then compare. That would work and could be fun. A bit difficult on my end since my inputs are coming from a console and then through D-Sub. I guess I could use sends but that introduces a variable. Maybe a simpler way to do it would be to play prerecorded material into the room and capture it on at both sample rates. Would obviously be a terrible recording but maybe enough to detect a difference. And easier than repatching. Maybe a M/S setup and then a room mic capturing program material out of a monitor. |
|
|
|
Post by the other mark williams on Nov 27, 2023 18:07:36 GMT -6
Not much unlike a formula 1 team trying to shave fractions of seconds off their lap times... those performance benefits are rarely one specific thing. I'll just say that to me, regardless of converters... regardless of the shop... and this is really super easy for anyone to do (or should be anyway...) Across the board the higher sample rates sound more like the input. If anyone's ever worked on 2" tape you know the difference between monitor input, playback off the record head and playback off the play head. They all sound a bit different. They all sound like music, but its kinda like looking at the same picture through different windows. The first time I heard it was producing a record at the Phish Barn. We tracked to 2" and wanted to dump to PT... Pete Carini said we need to run at 88.2 and we were all a bit skeptical. Drive space was expensive in those days. But he set up the blind A/B test... dumped one song at 44 and 88 and ya know what? Everyone in the room heard the difference. Full stop. No debate. Even here, now today... I can print a mix off the SSL and the 88.2 even through "lowly motu" converters is indistinguishable from the output of the desk. At 44.1 everything collapses inwards ever so slightly. I hear it. Most of the artists I work with hear it. And the mastering cats? They absolutely hear it too. We're chasing greatness yes? Most of this forum, especially all the naval gazing on "whuz the best ____" is about chasing greatness... seeking those small improvements... yes no?! How would someone do the equivalent test now? It's not really possible to track live instruments at 48khz and then again at 96khz and then compare them. I feel like the performance variables would be too great to really tell me anything. Nah, if the band is decently good, two different performances ain’t going to matter. You’ll be able to readily hear the sonic differences. |
|
|
|
Post by christopher on Nov 27, 2023 19:23:07 GMT -6
I think I've started to get an ok understanding what it sounds like. Its not that I'm hearing the filter mess with the FR much, or really aliasing so much.. its that I'm hearing some MIT chip designer's decision to force the manufacturer to use their "inaudible" on-chip questionable analog circuit, they put into the chain at lower rates. At higher rates more its bypassed so its not as much an issue. At lower rates its like a couple BOSS gate pedals set to always open, - sure its doing nothing.. but you know. I don't really know for sure if this is whats going on, but thats how it feels to me. Its not the end of the world at lower rates, I mostly don't care.. but that's what I'd try to listen for, a cleaner path type sound. And it is nicer, if I had my choice higher rates.
|
|
|
|
Post by jmoose on Nov 28, 2023 20:40:49 GMT -6
How would someone do the equivalent test now? It's not really possible to track live instruments at 48khz and then again at 96khz and then compare them. Sure it is. It's simple. Split the mic pre signals, and go to one DAW at 48k, and a second DAW at 96k. Use the same converter for each. Then compare. Pretty much something like that yeah kinda! And maybe I should've put a disclaimer that its "easy" if you've got the right gear... but we'd like to bring the post converter A/D - D/A round trip audio back into a monitor controller where you can switch between the "direct mic pre" sound and the "post converter" sound in real time. People who are using an interface for everything..? Its the converters, the preamps, the monitor control etc this is WAY harder because you never really get to separate all the pieces out and hear exactly what each bit sounds like... what its doing to the sound... find where the bottleneck is. Having something like a monitor controller or some kinda analog mini mixer, even a smackie 1202 so you can A/B/C in real time the original source (like your 2 mix off the desk) to what the sound is post converter? Super clutch. In the olden dinosaur days, or when we do the whole civil war re-enactment bit and work analog tape? We'd select the 2-track return on the console center section and listen "through" the machine... listen through to hear how its changing the tone as sound passes through it. These days its not a tape machine its converters! More or less - If you can patch your direct, pure "mic/source" tone to the "post capture" tone - level matched and in real time? That's what 'ya wanna do. Like printing a mix off the mini-me SSL I can real time switch between the "source/desk" and the 2-track return - which is a motu m4 on a separate laptop from the main DAW. Then I can switch sample rates on the M4 and hear what the round trip change is vs what the desk is ACTUALLY putting out. I've done this with all kinds of converters at all kinds of sample rates over the years and the results tend to be fairly consistent. The one BIG OLD got'cha ! In this whole thing?! Gotta make sure all the converters are level matched BEFORE you start running audio. Run a 1kHz tone and confirm/deny calibration/levels. In a perfect world aiming for no more/less then .1dB difference between sources. Anything that's louder is always better. Fools 'ya every time. Even 1dB difference is too much. And I'd agree that a band, all miked up & levels set off an analog desk? If they can really play then running one take at 44 and another at 96 then doing a quick rough mix of each... with the same settings for each mix/sample rate? Should be close enough for rock & roll. Either you hear a difference, or 'ya don't. Or maybe 'ya hear one and then don't even care anymore. And that's fine too! |
|
|
|
Post by matt on Nov 30, 2023 20:02:28 GMT -6
I notice I'm attracted to mixes that have less instruments/production...like say Stapleton or Gregory Porter There's a lot to like in the methods of the guys who record Stapleton- Dave Cobb and Vance Powell. Of course, emulating them is practically impossible, but I think we can cop some of their approach. Vance's videos on Puremix have cost me a small fortune, but sometimes you just gotta say WTF and go for it. Maybe you should try a Burl B32 and any available converter for D/A just to see. Since you are in Nashville, you might be able to rent or even borrow the gear and not have a huge outlay of cash until you're sure.
|
|
|
|
Post by chessparov on Nov 30, 2023 20:18:14 GMT -6
John K. does have an exceptional voice. Maybe try some different approaches with it as a litmus? Then add a few instruments at a time around it? IMHO the human ear is most sensitive to the singing/speaking voice. (Especially if one is married  ) Chris |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Nov 30, 2023 22:30:42 GMT -6
Sorry, late to the party here. This was something I racked my brains about as well for quite a while when going ITB only, no I'm not blaming software (because it was the UA CH curve bender that fixed it) it's just the results were always strong yet collapsed (as in merged and center focused). I've tried all sorts of widening (with some woeful results), delay or verb tricks, different methodologies of panning etc. I even changed Pan law interaction to mimic an SSL desk (yes I know it's just a volume thing but I thought it might have been a psychoacoustic anomaly as in my brain was being tricked).
With HW this was quite easy, on the Empress I'd enage shift boost or cut and then use the HW mastering limiter M/S compressor. We've all being doing this a while so I once again scratched my head why this wasn't happening with Ozone (or even Logic's EQ) when I did M/S processing. Anyway, listen to the samples of the UA CH curve bender and the width is there, it's easy to mimic as well. However this doesn't sort out the mid side individual layer separation piece. I watched an in depth well known producer video and I did some work with a local engineer who does mainly pop productions (I always did metal before of course). The amount of plugins they use on a channel is perplexing, said plugins don't do much individually but they do stack, distortion, transient shapers, comps, limiters you name it. There's also some pretty wild EQ decisions compared to what I'm used to, boost 10dB, cut etc. doesn't matter. You do what's needed to get everything out of the way and the main focus is around bass, kick & vocals.
Finally I'm glad I decided to swap genre's at the same time I bought the Dyn 59's, yeah I know I'm like a broken record at this point but I'd be utterly lost without a monitoring system that sounds eerie in terms of seperation when you get it right or a mush of crap when you don't. It does easily translate to crapper mediums as well, however on those mediums it's really hard to tell what the problems are IME. You just sorta know it's wrong..
|
|
)