sound card recommendations for classical piano

", I tracked down this: RME Babyface Pro"

With the VERY greatest respect, that ^ is one of the few things you have said that makes sense.

The overwhelming evidence is that >48kHz sampling makes no audible difference to recording except in certain rare circumstances but not, I would aver, in a single instrument like a piano that does not have strong ultrasonic harmonics.

You can hear the difference? Blind test? Sampling at 192kHz could well makes things sound differently from 44/48 but all the evidence again is that people are hearing the SHORTCOMINGS of the converts at these rates. But, if YOU like it fine.

"So why" you might ask "do they put a 192kHz capability in converters?" Several reasons spring to mind...

Marketing BS*. Probably the biggest reason and they HAVE to do because of people like you! (again, WTGR!)

AI get used for other purpose. Bats spring to mind but in fact most small mammals produce ultrasonics (I want a cheap, w'proof mic that goes to 50kH for hedgehogs!)
Engineering. Fracture energies probably go way past 100kHz? Creaks in structures.

*Couple more bits of shit, "leading people by the nose" marketing? "True" bypass electronics as manifested mostly in guitar pedals. Total BS.
"Class A, discrete electrical path" WTF would you make a pre amp OTHER that in class A? Op amps have WAY less distortion than discrete components unless you use a hell of a lot of them and THAT pushes the cost up (luvly jubbly!) and completes the circular argument.

"Things", active speakers, pre amps, power amps MUST have a response into the MW radio band. Yeah, great, then guys like me can try to keep the RF shit out!

Dave.
 
I don't want to sound like a CD, I want to have at least the quality of first cut vinal- can't anyone else hear the difference?

I think the primary differences lie above 20k with those media.
Vinyl bass response is compromised through the mechanical nature of the thing, though, so I guess, if anything, you'd want CD quality. ;)

but 48k sounds so flat- not good at all for my taste.

Honestly, if you can blind test 48k against anything higher I'd suggest there are more variables at play.
Perhaps this particular instrument suite recorded a completely different sample set for their higher resolutions and it sounds better for some other reason?

Maybe the digital effects play better at higher sample rates? I think that's possible.

Not trying to rain on your parade but generally when people can blind test above 48 they're super duper sure of themselves and the difference is night and day.
I don't know a huge amount about it but I'm I doubt anyone could blind test a straight up recording sampled at 48 and 96 or higher.
 
Call me a skeptic but if you look at the frequency response of even direct to vinyl you're not going to find any real content in frequency ranges that higher sample rates will help. But, it's a pointless discussion to have without having real content.

So, please, [MENTION=199150]sassysquatch[/MENTION], post a link to the 192kHz raw tracks, e.g., at Dropbox or downloadable SoundCloud, you've already recorded with your current setup that already sound better than 48kHz and I'll tuck them away and wait for some new ones of the same piece also done at 192kHz with the RME (which I'd love to hear BTW since I'll never be able to justify one for my use!).

And, even take notes of all FX settings and give us a 192kHz non-lossy bounce using the tracks from your current rig and then one with the RME captured tracks. I really, really do want to hear the difference! (I do realize the interface might make a difference on playback that I won't be able to hear because I don't have an RME, but I'm hoping it will be audible even on my old FW Focusrite.)
 
Anyway, I really don't know know what you guys are talking about, I wasn't asking about the merits of sound resolution. My 2 cents though: at 96k or 192k (I can't hear the difference over my speakers) both sound much, much better to me than 48k. I don't think it has anything to do with inaudible frequencies (which apparently only gives you in advantages in processing, which I have essentially no interest in, nor do I understand) but is related to the much higher time resolution and just the overall purity of the sound, hence the 'live' sound, like I used to get from my parents old LP's. I just think you should be able to put in an album and hear the music as if it were being played in the same room (like people had been able to do before cassette tapes spoiled peoples ears).
Anyway cheers!
ps. apparently the cheapest RME units are sold in Lithuania? Have you guys ever heard of chinese counterfeit high-end AI's? Does that exist?
 
Anyway, I really don't know know what you guys are talking about, I wasn't asking about the merits of sound resolution. My 2 cents though: at 96k or 192k (I can't hear the difference over my speakers) both sound much, much better to me than 48k. I don't think it has anything to do with inaudible frequencies (which apparently only gives you in advantages in processing, which I have essentially no interest in, nor do I understand) but is related to the much higher time resolution and just the overall purity of the sound, hence the 'live' sound, like I used to get from my parents old LP's. I just think you should be able to put in an album and hear the music as if it were being played in the same room (like people had been able to do before cassette tapes spoiled peoples ears).
Anyway cheers!
ps. apparently the cheapest RME units are sold in Lithuania? Have you guys ever heard of chinese counterfeit high-end AI's? Does that exist?

Well then it's going to be a real letdown when you listen to your stuff at CD quality 44.1kHz, or even worse, compressed MP3 quality....so I'm not sure how you are going to get around that. :)

If those samples were done at 192kHz...and they don't include 48kHz versions...it may then that your DAW is resampling them and changing what you hear....but again, like it or not, your will end up resampling them at some point to get them to CD or MP3 format.

I have plenty of piano samples done at 48KHz that sound "near perfect (or better) approximation".
I wouldn't waste my time or my system processing performance and HD space with 192kHz...*unless every single piece of audio gear in my studio signal chain was at that same 192kHz quality level from start to finish, for both the samples and any live recorded tracks* ...and I doubt your signal chains are either.
 
Anyway, I really don't know know what you guys are talking about, I wasn't asking about the merits of sound resolution. My 2 cents though: at 96k or 192k (I can't hear the difference over my speakers) both sound much, much better to me than 48k. I don't think it has anything to do with inaudible frequencies (which apparently only gives you in advantages in processing, which I have essentially no interest in, nor do I understand) but is related to the much higher time resolution and just the overall purity of the sound, hence the 'live' sound, like I used to get from my parents old LP's. I just think you should be able to put in an album and hear the music as if it were being played in the same room (like people had been able to do before cassette tapes spoiled peoples ears).
Anyway cheers!
ps. apparently the cheapest RME units are sold in Lithuania? Have you guys ever heard of chinese counterfeit high-end AI's? Does that exist?

So, you admit "...don't know what you guys are talking about" but then go off on some technical guff about "time resolution"! The 'Resolution' of a digital recording is not improved by a higher sampling rate, just the high frequency limit to just under half the sampling rate.

Then "...Hear the music as if it were being played in the same room" Nostalgic twaddle. In the average UK living room of say 120cu mtr a concert grand would sound bloody awful!

Those old LPs were BALANCED by very skilled recording engineers many of whom were top class musicians, to MAKE the RAH and the BBC phil SOUND good in a small space.

The cassette thing. Yes, many of the commercial releases were poor (I understand, never bought hardly any) but a dupe of an LP (or a BBC prom recording on good FM kit) using a top grade machine such as my Sony Dolby S job or the Dennon I had with 'computer' tape optimization and HX were 'as the record' using TDK SA tape or equivalent. AND! Yes the cassette machine were capable of excellent results but the copying was made easier because of the technical shortcomings of vinyl! Copying a well balanced CD DID show the limitations of cassette (good enough for the car tho'but!)

Dave.
 
This is completely unproductive- you guys clearly have a bug up your butt that has nothing to do with recording, so I think I'm giving up on this particular thread! Cheers, and thanks for the advice!
 
This is completely unproductive- you guys clearly have a bug up your butt that has nothing to do with recording, so I think I'm giving up on this particular thread! Cheers, and thanks for the advice!

Fine, I am wrong and everyone at HR is wrong and all the experts I have ever read in Sound on Sound and the people on their forum are wrong.

I wish you well.

Dave.
 
Hi Steenamaroon,
no problem, I totally appreciate the help, it's only that after the very helpful first page of comments, which was awesome, the conversation seems to have steered toward the merits of high res recording. I'm really not interested in discussing that. And though these folks are free to talk all they want, I'm a world class pianist and I have extremely keen ears, especially for the piano, and no amount of insisting is going to convince me that what I hear is incorrect.
Dave. Although I said that I can't hear the dif between on 96k and 192k on my home speakers, there is a difference when you're at 48k. So yes, everyone must be wrong, although I doubt "everyone" agrees with you.
I just need something that will support live hi-res sound because this is a bit of an experimental venture and the classical music crowd won't accept a clean but synthetic sounding instrument in any way.
 
I'm sure it's frustrating. You're set on your point and everyone else is set on theirs, but keep it in context.
Your thread is asking for hardware recommendations because A: You need? to record at >48k and B: You're having performance issues.

This is a conscientious forum. People won't just blindly recommend interfaces to satisfy your criteria if they think there's another/better solution.

I mean, they might now because your position's been made clear but even then, what is your final delivery format?
If it's not 96k digital then, with respect, the debate becomes irrelevant.

If you're adamant 96k sounds better to you and that's your final delivery format then fine...We'll keep it on-track. Right fellas?
 
I'm sure it's frustrating. You're set on your point and everyone else is set on theirs, but keep it in context.
Your thread is asking for hardware recommendations because A: You need? to record at >48k and B: You're having performance issues.

This is a conscientious forum. People won't just blindly recommend interfaces to satisfy your criteria if they think there's another/better solution.

I mean, they might now because your position's been made clear but even then, what is your final delivery format?
If it's not 96k digital then, with respect, the debate becomes irrelevant.

If you're adamant 96k sounds better to you and that's your final delivery format then fine...We'll keep it on-track. Right fellas?

Yes, I can live with that. Not sure how to repro' at 96k? Straight out of a D/A converter to digital monitors? But I thought the thread was also going with "if 96kHz is better than 48, 192kHz MUST be even better" . Pretty sure there is no evidence for that. Quite the contrary I understand?


Just thought. Re a totally digital repro chain? Dangerous, I would definitely want a big analogue knob in there? ESPECIALLY for live performance!
Dave.
 
The OP just has his opinion that higher sample rates sound better. I'm cool with this, but don't waste my time with it. My concert pianist collaborator has a very, very keen ear, and we've had to set up a metronome with a slight delay on his own system because the Cubase click is a tiny bit early, resulting in him being consistently about 128th/bar ahead of the absolute time. He cannot cope with this and spends a huge amount of time trimming teeny bits off the start of some notes - because he can hear it. I cannot. Yet he cannot hear any difference between the samples we have on his website - 320K/mp3 and the 44.1KHz CDs. I tried 48 and 96 recently - he states they sound identical, and I agree. I cannot hear any difference in a blind test. When I know I'm listening to 96K, I can convince myself I can hear it - but when I accidentally mixed some recordings up - the only way I could identify which was which without looking at the file data was by the time I'd written on the card!

Some people are certain they can hear the difference, and maybe they can - but this doesn't mean everyone else is wrong for not agreeing.

What exactly is in the extra data? If you edit out a glitch in the digital waveform, and remove a sample or two, or draw in a new curve with the pen - the click vanishes, and the data that you generated is transparent. All the stuff about LPs and other media really boils down to noise - hiss is bad, crackles are bad - they annoy and are obvious. Once the hiss is gone and there are no crackles - what exactly is the 'quality' descriptor? I personally think it's simply accuracy and realism and tone. A Yamaha and Steinway piano pair sound different. Not better. Some people like one, some the other because of choice and I think this is nothing to do with quality. The top notes of a piano are not that high, and the harmonics don't extend very far before they've vanished because they are very low level. When somebody wants me to do quality things I simply don't believe are necessary, I charge more, and many of them pay. If it takes me longer to copy files, and backup and edit, that isn't my fault, and it's much simpler to say yes, do it, and not argue, because they KNOW what they ask for is better. If you sit with them, it's obvious it's in their head. Seeing numbers, or HD or 4K in the video edit make them happy. Happy clients are worth having, even if they have really weird things in their head.
 
So, while, like I said, I'm a skeptic, [MENTION=178786]rob aylestone[/MENTION], what is the equipment chain you use if you are getting paid to do 96k, including the computer specs? OP wants equipment recommendations, so that might be useful.
 
Tascam 1641 - and I'll look at the computer tom arrow when we're in the studio. Not quite sure what getting paid has to do with it? My first outing with digital audio on computers was with a Soundscape SSHDR1 system with two cards - 48KHz and one of the worst user interfaces I've ever used - but it sounded good. I'm not sure audio quality has actually changed much since then really. There's a clear quality increase as sampling rate and bit depth goes up = but it suddenly runs out of steam. I'm not even sold on 44.1 to 48KHz. The old Panasonic DATs we were using could work at both sample rates, but one machine had copy protection that wouldn't turn off - and was 48KHz only. I don't think we ever really noticed. That was with 16 bit sampling, and I suspect pretty crude filtering but they sounded better than our reel to reels in so many ways. The Tascam interface is pretty silent, even with high monitor levels, and the tiny adjustments we make to the pianos is easy to hear, but switching to higher sampling rates seems invisible, with no quality jump we can determine. I'll add the spec asap.
 
Tascam 1641 - and I'll look at the computer tom arrow when we're in the studio. Not quite sure what getting paid has to do with it? My first outing with digital audio on computers was with a Soundscape SSHDR1 system with two cards - 48KHz and one of the worst user interfaces I've ever used - but it sounded good. I'm not sure audio quality has actually changed much since then really. There's a clear quality increase as sampling rate and bit depth goes up = but it suddenly runs out of steam. I'm not even sold on 44.1 to 48KHz. ...
Well, it seemed like your point was you'd only use higher rates if you were getting paid [for the higher sample rate]. I didn't know if that meant you used any different equipment.

48kHz is primarily for video compatibilty, from everything I've read. No special ultrasonic content enhancements claimed to my knowledge.

Thanks.
 
isn't that the point, Keith - if the 44.1 to 48K increase provides no claimed ultrasonic increase in the accepted hearing range - then any move above it MUST be a smaller one, moving into territory our hearing isn't designed for. If nobody went wow - when switching to 48K, then who are the people saying how wonderful the doubling and quadrupling is? Pseudo-science at it's most gullible. If I read something by any of the respected people that I pay attention to who are very interested in audio recording and replay quality, then I would listen to them - but they don't.

What I actually meant and didn't explain, was that I rarely detail specific technical stuff in client quotes. Video is the better example - lots of repeat regular clients, and one uses lots of archive material - back from when we shot 4:3 on SD. We shoot now on 1080, and could shoot on 4K, but don't mainly because the end media doesn't make it cost effective in terms of computer time - which is far worse than audio. Pressing render and seeing the clock start taking hours instead of minutes has a cost implication. Not once has any of our regular clients ever asked what format we are going to shoot in - it's changed over the years and they simply base conclusions on the end product. Lots of our projects will be edited in 720p - we can go up from SD, or down from 1080 and they live happily.

Sound wise, the projects sometimes start in the wrong format, by accident. If we've been working on a 44.1 project, then quite often, a new project inherits this, and my colleague started one in 96 by accident because I'd been fiddling - it makes NO difference we can hear.

We've just started a new story-telling project and distribution is going to be by USB sticks. The non-technical client has picked some very strange designed sticks that are part of the package and non-negotiable in terms of changing them. Delivery was planned to be 320K mp3. Now, it's becoming clear the length won't let the complete project fit, so we've just sent some samples to her at 256K mp3. Her comments have all come back about content. At the end of section 1, can that little pause be a smidge longer, can the trumpet in section 2 be lower, like a trombone? She was meant to be commenting on quality, so we can assess if dropping the mp3 quality one notch is acceptable. Clearly, she's not heard anything quality wise, and is concentrating hard on the content. A real end user, uninterested in quality by technical terms, simply the performance. Sending the example tracks for comment has generated far more work in content, and none in the fidelity area.

A friend of mine is an engineer, and it drives him crazy if screw heads are not all aligned the same. He cannot cope with ugly engineering. He over torques non-structural fixings to look pretty, and slams others for not doing this. In audio, double blind testing is very rare. Until people do some, we are resting on opinions, not facts.

EDIT
i7-4790-3.6GHz, 16Gb RAM, Windows 10
 
Last edited:
Back
Top