I've been away from the music world since 2013, when my last band broke up. It's been a long hiatus, but I've been chatting with a few members of that old band, and we're going to form a new project this Spring/ Summer. Which means I need to buy and build a new recording setup.
Our old setup was analog into digital. Mics ran into preamps and then into a 50' stage snake, which went off to another room where it all plugged into a big, beastly Behringer Eurodesk MX9000 (I miss meter bridges, dang it!). We'd then run TRS from the subs to an audio interface, and pipe it all into Adobe Audition where we'd record at 24-bit/ 192 KHz.
We no longer have that gear, so I need to build a whole new setup from the ground up, from the console to the mic clips. I've been looking into going as digital as possible, but I may end up just sticking with analog-to-digital, because digital is friggin' confusing to me.
Here are a bunch of questions I can't seem to answer.
1. Why do all of these digital mixers output at only 24-bit/ 48 KHz? I can hear a VERY big, very obvious difference in sound quality between 48 and 96. I don't really hear the difference between 96 and 192 if I'm honest, but I record at 192 and convert down just so I'm editing the cleanest signal possible in post. So why are these top-shelf mixers, like the PreSonus StudioLive Series 3, only capable of 48 KHz by USB? Or does that only apply to the mixer recording to a thumb drive or something?
2. Do I even need a mixer? Can I just get a digital snake/ stage box and run a single cable from the live room to a computer in the control room, and then use a virtual console through software somehow? From my limited reading last night on this subject, it seems like Ethernet/ AES is only capable of transmitting at 48 KHz, too. What the heck? And I don't think a USB cable can even run 50+ feet long without latency issues... if I can even find a digital stage box capable of outputting 192 KHz via USB, which also seems impossible.
3. Why is the industry still stuck on Cat5e? My apartment runs Cat7. That's 10,000 mbps, as opposed to 1000 mbps max from a Cat5e cable. So if Cat5e is stuck at 48 KHz (which it seems to be, but again, I'm not sure), wouldn't a Cat7 be capable of 192 KHz? Or maybe even 384 KHz, which PreSonus Studio One is apparently capable of recording at?
So yeah, there's my quandary. Do I go with an analog-to-digital system consisting of an old-school snake, a mixer, and an audio interface like I had before? or is there a better modern solution?
My requirements: 24 to 32 XLR inputs (XLR+TRS is even better), preferably with decent-quality preamps/ phantom power, some returns (nice to have but not necessary), and signals going into a PC at 24-bit 192 KHz resolution (or better, if that's even remotely possible). My budget will be in the $3k to $4k area for the snake, the mixer, and the audio interface combined, including all the cables I need.
Should I make the leap to digital? Or am I better off sticking with analog-to-digital like before?
You could have a computer and interface setup in the recording room, with all the XLRs running right into it, and headphone mixes coming out. I'm not sure from your post if you still have the other room, but if you do, and you don't have someone to run things during a tracking session, your best bet is to have the whole setup in the recording room. If necessary, put it on a rolling computer cart and move it to the other room for mixing or whatever. Add an old fashioned snake and headphone extensions if you want to do the separate control room setup. There might be a digital snake that will do 192kHz, but I'd just go analog in this case.
I don't know if any USB interface can do 24 channels of 24 bit/96kHz audio. It may take Thunderbolt or a PCI host card system.
You're requirements are 24-32 channels @96k ...Just making sure...are you talking about recording 24 tracks at once at that SR?
We'll be using a live room, and then a separate "control room" with a video camera hooked up to a TV as our "window" lol. But yes, there will be two separate rooms. 50' of cable should cover it (though considering the wackiness of going around doorways, I might prefer 100' to be safe).
I was thinking I could maybe get 3 audio interfaces with 8 mic inputs each, but that's three USBs plugged into the desktop PC at once. I've never tried something like so I'm not sure if the three AI's will interfere with each other.
We'll be using the space to multitrack, but also to record live and maybe stream, which is where it all gets tricky. As it's looking right now there will be 18 to 20 mics all running at once, and naturally I want those extra four because I always seem to need extra inputs lol.
I'd prefer to record at 192 KHz, but I'll settle for 96 KHz. I can't hear the difference between 192 and 96 personally, but I feel like it's wiser to record at 192 just to future-proof the tracks and to make sure my editing and processing is using the best raw audio I can get.
Someone will probably come along with more specific info, but there may be USB interfaces you can use in multiples. The thing is that they have to be clocked together and the driver has to be designed for multiples.
Yikes. That's the benefit of a mixer really. You can just route your subs out via TRS and plug into a single 4-port or 8-port AI. I get the strange feeling the people who design DAW equipment don't appreciate the fact that some of us play old-timey wooden instruments and like to mic as many of them as possible, lol.
If you can live with eight inputs at once you have lots of USB options, complete with mic preamps. You just might have to re-patch things now and then.
I'm always afraid I'll say "hey, go unplug that XLR on channel 2" and someone will somehow break a pin or something, lol. More heartbreaking things have happened in studios sadly. Once you watch a keyboardist dump a super-sized Coke on a 32x8 Mackie because he thought your meter bridge was a good place to put his drink down, it makes you obsessively nervous lol.
Then get an 8-channel snake and let it take all the wear and tear.
I’d be going a digital interface with dB-25 I/O to breakout cables. That will sort out your SR and distance issues.
You can do 192KHz on things like the MOTU B16's and quite a few others, in batches of 16 inputs. Cat 7 wiring is backwards compatible - so if a new build is being wired, for more money on cable infrastructure, anything can be squirted down them, almost. CAT5E is quite happy with digital audio data rates, 6 and 7 have spare capacity.
In nearly 40 years I have NEVER broken a pin pin an XLR. I have, when things bash into a patch panel, snapped a few push buttons off sockets, but plugging in and out is risk free.
I'm a bit at a loss to understand why you feel like you do about digital - I liked analogue, until I tried digits and I've never looked back. Even the cheapest digital desks offer more inputs than most people need. I've sadly become very detached with the upping the sampling rate. I sat in my studio and tried all the different rates I could access, with my usual sound sources. Even 96, on a blind test left me on the fence. I might have heard a difference, but I might not - I don't think so. Going up again while I tried it, was pointless.
I recording the jangling keys - and that did make a very small difference between 44.1 and 96, and probably a bit between 48 and 96, but 192? I couldn't tell. I'm NOT saying other people are wrong, and can hear it, but my ears run out of steam at 13K now, I don't think when CDs first came out I could hear more than 18K in my 20s. I know we can record at 192, and processor speed can keep up - but I ask myself why I personally would want to? I don't think there's a point.
If you want a 192K stage box that you can plumb straight into the computer - there are quite a few out there. The mixer manufacturers are not quite so convinced. The ones out there are VERY expensive, and I guess they're just not quite convinced yet? However, as most recording setups don't need a hardware mixer, leaving those for the live audio people, there's no problem - as all the main players are happy with all kinds of audio input specs.
Or you could get the Orion32 alone with a good analog mixer..
four grand is a lot of money, isn't it?
Maybe a bit late to point this out, but if you're comparing sample rates, you're probably just comparing the performance of one converter design at different rates. A given converter may just sound better at one rate than the others, and it's not always the highest rate it's capable of.
There are good reasons to stay at lower sample rates. There's something of a consensus among the "smart people" in the field that the optimum sample rate is around 50-70kHz, and that 88.2 is the best overall option.
I also wonder if re-coding between sections or devices is actually as transparent as we think. Lots of kit now tracks about their preamps sampling at 192KHz, but the transfers are at a lower rate - 48 or 96, and then the DAW works again at it's favourite rate - that's a lot of fudging numbers around.
bouldersound, post: 455614, member: 38959 wrote: There are good reasons to stay at lower sample rates. There's something of a consensus among the "smart people" in the field that the optimum sample rate is around 50-70kHz, and that 88.2 is the best overall option.
This is a debate that's been going on for 20 years. I'm kinda doing the ostrich with this for many years.
All I know is that when tried to record at 88.2 vs 96khz, 96khz sounded better to me.
I have to say that when I export a song, there's always a dithering done, so maybe this is where the difference goes in my twisted brain.. ;)
Seriously, It would need someone with 2 daws to test this or an external recorder and a DAW.
To acheive a true null test, we'd need to record the same sources twice, at 88.2 and 96. Then export to 44khz and make a null test.
Altought the DAW used and converters could screwup the results.. but it's worth a shot.
Anyone up to it ?? Sadly I only have 1 daw and no external recorder...
Anyone already did it ?
I only ever recorded at 192 because I believed (perhaps foolishly) that even though I couldn't hear it, the signal was in its purest form, so when I'm adding things in post like compression, reverb, what have you, the processes are higher quality. It made sense in my head at the time, but yeah, if you can't actually hear the benefits they might not really be benefits, lol.
I did find this gem on Sweetwater, the Roland Studio Capture. It has 12 XLR/ TRS preamps and 4 TRS inputs and monitor out. But here's the cool bit: you can turn its 16 channels into 32 by daisy-chaining a second unit via coax. I read through the manual and found that while the marketing claims it can do 192 KHz, it's actually limited to 96 KHz when you're using all of its channels. But like I said, I can't hear the difference between 96 and 192, and reading everything here I'm starting to feel a bit silly for thinking editing at 192 was better even though I couldn't hear the benefits.
I can get two of those and combine/ synch them for $1600, which is cheaper than some of the analog mixers I was looking into. I'm still lost on choosing between a digital snake versus an old-school analog snake, because it appears digital snakes are all limited to 48 KHz, but the analog snake is a lot cheaper of course.
Edit: I've never used DB25 or any of that. How does that all work? Does anyone know offhand what the max SR is?
DB25 is a multipin connector used for different things. It's one way to get a lot of connectivity without using a lot of panel space. Tascam uses it for their digital multichannel format, and it is also used for analog multichannel audio.
Can DB25 handle 96 KHz and/ or 192 KHz? I've read that Ethernet-based AES can't work over 48 KHz, and I can't find anything talking about resolution when looking through DB25 info (at your link or pretty much anywhere). It's surprising how inaccessible information regarding digital audio there is online, with regards to cable formats anyway.
DB25 has no SR limits. It’s a pure analog connector. Limitations exist only in the AD/DA converter interface. I used my Orion32 Antelope with them, it does 192k for example.
Very cool and good to know, thank you!
As an aside...
Have you ever recorded that many tracks at once into your current computer at those higher SR's to make sure the CPU can handle that load and the storage? Or are you also looking at an upgrade to your system as well? The reason I mentioned this is that you might have to add the cost of a new computer into your production budget as well...
I've never been able to discern the sonic differences between 96 and 192, and honestly I've not really been able to hear anything other than very subtle differences between 48 and 96, and even then, I'm not sure that my perception on that difference wasn't perhaps psychological; I'm not sure I could pick one over the other in a blind test... Then again, I'm in my 50's now, so I'm sure I'm not hearing details and nuances like I once was able to.
If you can record at 96 I think you should - if you can hear a difference then that's what matters most...second to that would be simply because of the math involved and the overall higher resolution; but you will need a system beefy enough in CPU and storage to do that efficiently.
I know nothing about Roland Preamps or interfaces. They may be fine, though I don't believe that pro rooms are using Roland for anything other than in-house keys/synths.
You'll also need a cue system that will support individual HP mixes for each player, something like an Aviom system would do nicely.
If it wasn't for your 96k criteria, I'd probably suggest just using a Presonus SL, which uses FW as it's connection to whatever capture DAW you use...but it tops out at 48k ...or at least I'm pretty sure it does. Dave Hawk (dvdhawk) is RO's resident Presonus guy, he could check me on those specs...
Finally.... Have you given any thought to simply taking your budget, or part of it, and booking time at a pro studio that already has everything you need in place? Things like high SR tracking, nice mics, preamps, great conversion, cue systems, good sounding spaces....along with a good engineer who knows the gear and the room - somewhere you and your band could record where all you would need to be concerned with would be the music? You don't have to mix the project there; just get the engineer to export the raw wav files to an external HDD or flash drive for you, and you can mix yourself ifyou feel you have a good sounding room and monitors that will support you mixing it.
I get the DIY thing, I really do. But there's a lot involved in the "guerilla" type of recording you want to do...plan on spending quite a bit of time and energy running lines, setting up the systems, mics, acoustic treatment...and plan on having to work out a lot of bugs along the way, too. DIY sounds great on the surface,, and it sounds cheaper, too ... But it's never as easy - or as cheap - as it seems to be in the initial planning stages, as it really turns out to be when you're actually trying to do it.
Lol...ask me how I know. ;)
Anyway, it wasn't my intention to hijack your thread. I'm just kinda thinking out loud...maybe giving you another angle or two to look at this scenario from.
Sounds like you need to get your old gear back! Your requirements are actually quite high end.
The advices about differences between 96khz and higher at 192khz are well informed.
It is really superfluous to be recording an electric band at above the 88,2 or 96khz dynamic bandwidth.
An acoustic , jazz or orchestral ensemble, it can be beneficial given highest quality Mics, Preamps
with lots of headroom and converters that over sample in the DXD realm; but all futile in an ordinary less than ideal space like a smallish square room untreated acoustically.
There are plenty of excellent interfaces with Adat capability, but this will keep you in the 48k realm.
The MotuB16 would be a way to go with included DSP to attain your desired sample rate with low latency and long cable runs.
If overdub latency is not required some of the older 196khz converters might be the more affordable ticket if your computer is up to spec.
Keep up the research and good luck with it.
Bartie Boy, post: 455674, member: 51087 wrote: There are plenty of excellent interfaces with Adat capability, but this will keep you in the 48k realm.
Adat can run at 96khz but you cut the channel count in half...
On my Cubase system, some of my old favourite reverbs and VSTis are 32 bit ones, and now Cubase doesn't support them, so I run them via a bridge - they work in 32 bit, but Cubase doesn't know! I've got loads of old projects at 44.1K, most at 48K, and a few from outside I work on at 96K. Some of these projects contain 32oKbs MP3s. Some contain material from Mini Disc recorders. They all sound very similar - clearly they are VERY different, but they really are not dreadful vs wonderful, and the content on some of the mp3s sounds great, while some of the 96K content sounds quite dull and not as good. I fear that many people assume that the higher the sampling rate, the higher the quality - but we forget that often the sounds we record have been produced inside another machine. I don't know the sampling rate my Korg uses, or the other electronic sound sources I have. I suspect most are 16 bit at 32 or 48KHz sampling - so using a higher rate seems a bit pointless. Are my favourite processors and pedals working at 96K or above? I doubt it. Does my SM57 produce anything at all that the 96 or 192KHz sampling machines can capture? The filters fitted in most gear slams shut at 20K, so what is there left that will benefit?
Just because we can do something doesn't mean we should.
Boswell wrote: Although the range of my hearing as measured by a clinical hearing test extends only to about 15KHz these days, I have no problem in distinguishing a 7KHz sinewave from a 7KHz squarewave (adjusted to have the same medium to high level fundamental amplitudes), where the only differences between the two start at 21KHz. At lower levels, I can't easily distinguish between them. This squarewave test demonstrates that the upper end of the human hearing range is not a sudden cut-off but a sloping curve with a shape that changes with received level.
When it comes to transient sounds, there is a big difference for me between a 44.1KHz and a 96KHz recording of things like tingsha bells, which have significant energy in the 20-30KHz region. Incidentally, to make this comparison, you have to use a microphone and other equipment capable of operating to 30KHz or higher. It's very likely that to process transients the ear uses additional mechanisms that not only extend to a higher frequency than the steady-state sound detection mechanisms but do not degrade as much with age.
The clip above comes from this thread. Might be worth a read!
DonnyThompson, post: 455638, member: 46114 wrote: If it wasn't for your 96k criteria, I'd probably suggest just using a Presonus SL, which uses FW as it's connection to whatever capture DAW you use...but it tops out at 48k ...or at least I'm pretty sure it does. Dave Hawk (dvdhawk) is RO's resident Presonus guy, he could check me on those specs...
The PreSonus AI series consoles and AI rackmount (iPad / Surface controlled) digital mixers, are all capable of 44.1, 48, 88.2, and 96kHz. The newer Series III (consoles and rackmounts alike) are currently 48kHz only. They plan to offer 44.1kHz in a future firmware update, no news yet of whether they'll be able to do higher sample-rate on the Series III via firmware.
Kind of responding all over the place, near with me lol.
The main reason I'm DIY is because I really nerd out over stuff like mic placement. I'm that guy who can debate overhead placement for five hours lol. I'll spend whole weekends just recording my drums or messing around with guitars. Also -- and nobody lose it on me -- I despise click tracks. Recording without them is harder, but I find it more fun, and I think there are benefits that few others agree with me about lol. It doesnt work on tracks that open with another instrument, though.
My other pastime outside of music is video games, so (without meaning to brag, of course) I'd reckon my PC is considerably more powerful than a typical DAW machine. I want to build a new DAW-only machine though; Ryzen 6-core, 16 GB, etc. It should make short work of 192, though I think I'm leaning more toward 88 0r 96 after these discussions.
I hadn't really considered mic quality in my 48 vs 96 vs 192 tests those years back. We did them all on an AT4040 and an e609 on snare and vocals. That 2020 was running through an Art Pro MPA II, so it wasn't really a very good test. I wish I had tried with something like an SM57.
I really want the Presonus 24R to handle 88 or 96 and output to a PC in another room. A digital stage box would replace the need for an analog snake, and also it'd be cheaper than two daisy-chained Roland Studio Captures by $500 or $600. That's a whole lot I could be spending on overheads lol.
I understand being obsessed with getting great sound. I think everyone here does, but your obsession might end up being your downfall if you're not careful.
One of the things I think you are obsessing over are these ultra high sampling rates...
There's no doubt that conversion quality matters. I don't think anyone here would argue that ...but that's more to do with the quality of the converters than it is the actual SR that a converter is capable of delivering.
96k in itself will not be an "instant great sound". There are different levels of conversion quality, and in some converters, particularly cheaper budget models, the quality Isn't as good as others...put it this way...I'd rather record at 48k (or even 44 for that matter) through something like an Antelope, Apogee, Lynx or RME, than I would record at 96k through an M-Audio or Tascam. Why? Because those budget level devices cut a lot of corners - on preamps, converters, circuitry, to make them affordable to the home recording hobbyist masses. There are reasons why a Grace or Millennia single channel mic pre is $800, and a reason why a Tascam 8 channel preamp/interface is $300. Corners are being cut to make them affordable to hobbyists, many of whom wouldn't be able to hear the difference between a Neve 1073 mic pre and an M-Audio mic pre, anyway.
But, if your listening skills are as honed and as sensitive as you say they are ( and I have no reason to doubt you, after all, how would I ever know what you can hear, right?) then you will also be able to instantly hear the differences in conversion quality as well. It's not true that "a converter is a converter".
We all know that's not true. But you'll hear a much greater difference in better converters - along with mics and preamps - than you will the sampling rate itself.
I'm suggesting that you may be putting too much reliance on the SR alone, and, honestly, I think you are also expecting too much for the budget you mentioned having.
In terms of pro level recording gear, $3-4000 is really not much in the grand scope of things. You could easily spend that much money alonenon 3-4 good mics, or 3-4 good mic pres, or even on a high quality standalone conversion system.
The environment you are recording in also matters a great deal.
For what you've described wanting to expect and accomplish, and at the caliber of what you are hoping for, is unrealistic, I think...
What mics do you have? Any standalone hi quality Preamps for things like lead vocals? How does the room sound? Are all your cables in good shape? These are things that count...much more than 48k vs 96k.
Just sayin'. ;)
All valid points, but truly, on an AT4040 and an Art Pro MPA II, I could definitely hear a stark and noticeable difference between 48 and 96, and that was back in 2010 or 2011. If I remember correctly, we found out by accident, actually. We were using the 192 setting, and someone accidentally changed the sampling rate to 48. Suddenly the vocals sounded like cardboard, and it took me a good 20 minutes to figure out what the heck had happened, lol. That's when we tested all the different sample rates, and the general consensus was that 96 sounded best (one said she couldn't hear a difference between 48 and 96, another guy argued emphatically that 192 sounded best, but the rest of us agreed on 96). Still, we recorded at 192, because hey, why not?
I haven't bought the new mics yet (that's coming soon... I'm still arguing with myself about some of them) but as it stands, I'm getting an AT4040 for vocals (running it through an ART Voice Channel), a pair of e609's for guitars, an e602 II for the bass amp (I don't usually direct line), a pair of e614's or maybe SM81's for overheads going through an ART Pro MPA II, an e904 in the kick, ATM230's on the toms, and SM57's on the snare and hats respectively. And I'll naturally pick up a handful of 57's just to have lying around. That's the best thing about rebuilding everything from the ground up.
The budget I mentioned earlier was just for a mixer, AI, and snake. I'm probably dropping $2500 to $3000 on mics, $1000 (or less... probably less) on preamps, and then... well... I don't really know now. It's looking like those two Roland Studio Captures will replace the need for a mixer, so that's about $1600, which would leave around $1400 for a snake. I could buy a digital snake, but that might gum up the works for wanting to record at 96 KHz. Or I could just do what others have suggested in this thread and use analog snakes, and pump those saved funds into better preamps or channel strips. I'm a ways off from owning a 6176, but every bit counts right?
$2500 is about the total budget of the old gear we had, and we could hear the difference between 48 and 96 on that stuff. Really! We had a 4040, a pair of AT Pro 37's, an e609 we used on guitar/ bass cabs and the snare, a Pro MPA II (seriously an underrated preamp), running through decent Mogamis into a Behringer Eurodesk MX9000, which ran into a $200 AI. Even on that very cheap setup, you could clearly hear the contrast between 48 and 96. My theory going into all of this was that you'd hear an even bigger difference with better gear... maybe even hear the difference between 96 and 192, though I'm seriously doubting that no after everything in this thread and I'm probably sticking to 96 (again, though, if I can record at 192 in the end there's no reason not to really, is there?).
Mattrock607, post: 455718, member: 51138 wrote: if I can record at 192 in the end there's no reason not to really, is there?).
Except what people have told you about actual quality/performance scaling downward on some gear as the SR gets higher..... I've got an M-audio interface and I trust what Donny said about sticking to it's sweet spot in terms of SR/Quality.... there's a relationship there for sure..... The cheaper the gear and the more features it touts, the shittier the performance is across the array of features... the adage of doing one thing really really well, vs all things marginally well.
We were using the 192 setting, and someone accidentally changed the sampling rate to 48. Suddenly the vocals sounded like cardboard
This is the kind of thing that I just don't get. Lopping off the top makes this kind of difference? Really? I can identify quality fine, I can even change my mind (when one of the notables here suggested I needed to try different preamps to hear differences - I did, and changed my view. The trouble is we are talking very small differences, the kind you need to A/B a few times to appreciate what you hear, but the idea that 48K sounded like cardboard and the 192K didn't I just cannot accept. If the vocals sounded bad at 48K, then something else was wrong. The common sense and the physics just shouts too loudly for me to accept such radical statements.
I will have to remain unconvinced - I'm ready for another round of negotiation with a client on reverb today. I'm replacing unrealistic reverb on some tracks with better ones, and he cannot hear the difference, but I can. I can't convince him there is a difference, I can hear it, he cannot. Maybe this is similar? ~Could it be that what he hears sounds fine, because he is not listening to what I am hearing? Could I be missing the 96K/192K benefits? Maybe I can detect the difference but need training of my ears?
My client wants the tracks uploaded for download as soon as possible - I want more time to go through all the tracks, and there are lots, to replace the annoying reverb. Probably, as he's happy - I should just make them live and move on................?
paulears, post: 455843, member: 47782 wrote: This is the kind of thing that I just don't get. Lopping off the top makes this kind of difference? Really? I can identify quality fine, I can even change my mind (when one of the notables here suggested I needed to try different preamps to hear differences - I did, and changed my view. The trouble is we are talking very small differences, the kind you need to A/B a few times to appreciate what you hear, but the idea that 48K sounded like cardboard and the 192K didn't I just cannot accept. If the vocals sounded bad at 48K, then something else was wrong. The common sense and the physics just shouts too loudly for me to accept such radical statements.
Another thought on this... I'm trying to figure out how one can even "accidentally" change the sample rate in a project by " flipping a switch"...
Every DAW I've ever worked with would send up a warning notice if the sample rate was changed, accidentally or otherwise, prompting for a conversion of all the files in the project.
If they were using a DAW that was the exception to this rule, ( I don't know which one that would be), then what they were likely hearing was the result of a downsampling in SR for playback of tracks that were originally tracked at a higher rate and then converted down to try to play through a change in the DAW setting.
As far as A/B'ing a vocal that was tracked properly ( not recorded at one rate and then downsampled after wards) at both rates, I'm dubious that anyone could discern between the two in a blind test.
Mattrock607, post: 455718, member: 51138 wrote: We were using the 192 setting, and someone accidentally changed the sampling rate to 48. Suddenly the vocals sounded like cardboard,
paulears, post: 455843, member: 47782 wrote: This is the kind of thing that I just don't get. Lopping off the top makes this kind of difference?
We've been discussing that topic a year or two ago. To what I understood, this is very dependant of the audio interface and the converters themself.
On some converters, 48khz will sound better and on some 96 or more will sound better. It's kind of a maker's choice or maker's accident.
When I was using a focusrite liquid saffire 56, I did the test for myself and I could definitly hear the difference in favor of the 96khz.
I would guess that even two identical units may differ in sound if the tolerance of quality is not so good. You know those friday afternoon units...
Today with a mix of converters in my setup, 96khz is still my choice (FF800, 4-710 and AD96)
The best thing we could do is make the test ourselves with the gear we have and go foward making music.
In that I would advise the OP to make that test again when he switch to new converters.
Regarding the sound of different sample rates: double blind A/B/X testing or it didn't happen.
I think Mattrock has been listening to sales people too much.
Fi, cat7 doesn't exist yet and it isn't 10 Gbps either. It's just 1 Gbps, just like cat6, but it's destined at hostile environments, it's shielded. Cat6 is unshielded.
I can demo a lot of audio interfaces so they will produce an audible difference between 48 kHz and 96 kHz sample rate. But if the comparison is set up with good gear, level matched and doesn't include plugins or dsp's in the chain, I've never found anyone capable of being able to reliably recognize the difference in a blind test.
That's why live gear, like the X32 or the Presonus Live stops at 48 kHz.
What you hear IF you can tell the two apart, is bad filtering, or the absence of any frequencies above 20 kHz. Some interfaces pretend to do 192 kHz, but their filters don't let it pass. I know that for a fact cause I use audio interfaces for measurements and I've also been known to record bats and mice. Mice can go up to 100 kHz.
Of course, good gear goes up there too. But I've never understood the use of recording at 96, or even 192 kHz. Some people tell me some plugins benefit from it. I'm willing to accept that. So, that could be a viable reason to record at 192 kHz.
Another possible advantage is that latency halves (on any decent hardware) if SR doubles. But DSP capacity quadruples. That's why it's too expensive for live consoles, I think.