Noisy wavs at 96kHz vs normal 44kHz / 48kHz wavs

AA2.0 can't record at 96kHz vs 44kHz / 48kHz

Note: After seting up a track for recording 9632, just putting the record track in play (w/o) recording causes severe jitter noise as I play guitar.


Tested Echo Mona with A) WDM  Purewave drivers.

Sound Card MONA / EchoAudio
7.2 driver / Records fine: 44.1kHz 32bit // 48 kHz 32bit // Can't record 96 kHz 32 bit

Cln sound when previewing 96kHz 32 bit trks w/playback device (multitrack) / inserting 9632 trks into multitrack. When I "arm for record" at 9632 resolution is where the pblms start.......

Edit / Audio hardware Setup / Echo PCI 9632 / 4096 samples / WDM driver
- setup Mona 3 (mono) as "in" // Mona 4 (mono) as "out"
- hit rec button (turns red)
- save session (name it)
- play this back - w/o rec
.:. recording / "arming a trk for record" at 96kHz 32 bit - trks sound slow / with distortion

Note: closing down / then reopening AA 2.0 allows me to preview / open / insert a 96kHz 32 bit trk (multitrack) w/cln sound.

Reference / Cln Sound: 44.1kHz 32bit // 48 kHz 32bit
importing 128kbps 44.1kHz 16 bit mP3 file into 44.1kHz 32bit (or) 48 kHz 32bit multitrack / setup recording Mona "analog 3 In" / "analog 3/4 out" / arm for rec /
.:. recorded 44.1kHz 32bit // 48 kHz 32bit trks have clean sound

Installed older 6.11 echoaudio driver / mona
- Can listen to 9632 trks in CEP 2.1 / AA 1.5
- Can preview only AA2.0. When I open ANY trk 48k/96k in AA 2.0 - playback doesn't move

Reinstalled recent echoaudio / mona 7.2 driver
- can open trks / rec 48kHz 32 bit

Need to fix: AA2.0 can't record at 96kHz vs 44kHz / 48kHz

Note: Setting up a track for recording at 9632 / then just putting the track into play - while playing guitar - not even recording sound - causes high noise. 

I need to know how to correct this defective software.

I need to know how to correct this defective software.

Since others don't have any trouble with this (although why on earth they bother I don't know), I'd say that it's either a card fault or a driver fault, but certainly not Audition's fault!

I need to know how to correct this defective software.

I think it's something else in your setup which is 'defective'.  Although I would never mormally use such a high sampling rate, I have just run a simple test - along exactly the same lines as you indicated, but using my LynxOne card - and it worked perfectly for me.

Since others don't have any trouble with this (although why on earth they bother I don't know),

Thanks Graeme / Steve G - I'd like all options open / working.

My card EchoAudio/Mona works fine at 9632 with: CEP 1.0 / CEP 2.1 / AA 1.0 / AA 1.5 - but not AA 2.0.

Regarding your quoted "others" - How many people here record at 9632?

Frank Fillipetti allegedly records at 88.2 at Right Track Studios NYNY (as told to me by his Asst). He's engineered many hit records. Won a Grammy as well.

Sampling rate and the need to have current AD/DA conversion filters is excellently discussed in a thread here.

I’ve recorded without problems at those specs on my 233Mhz AMD K6 machine, with 64 meg of (slow, old RAM), running Win95, using its original hard drives and an Audiophile 2496 card.

My card EchoAudio/Mona works fine at 9632 with: CEP 1.0 / CEP 2.1 / AA 1.0 / AA 1.5 - but not AA 2.0.

Well, there's only one difference between AA2.0 and all the others in your list - and that is that all the rest don't use ASIO - but AA2.0 has to. So this puts your problem fair and square into driver territory.

Regarding your quoted "others" - How many people here record at 9632?

I generally don't - because there's no point. But if I do, for test purposes, using the MOTU Traveler at 96k/32bit then I get an RMS average noise floor with the mic preamp gain set to minimum, of -106.26dB.

If I redo the test without altering anything about the Traveler except the sample rate to 44.1k, then I get an average RMS noise floor of -110.63dB

The reason that 44.1k has a lower noise floor than 96k is because the noise bandwidth of a 96k recording is 48kHz, and the noise bandwidth of a 44.1k one is 22.05kHz. The noise figure is effectively  a cumulative measure of all of the noise in the bandwidth, so if the bandwidth goes up, so does the noise figure. Nominally it would be 3dB worse if you doubled the bandwidth, but in this case I've done slightly more than that, so there is a 4dB difference instead of a 3dB one.

But AA2.0 has no problems with the MOTU ASIO driver. So if you are having a problem, I'd get in touch with Echo and ask them why - and you can tell them that other drivers like the MOTU one don't have any problem with AA2.0 at all. See what they say...

Frank Fillipetti allegedly records at 88.2 at Right Track Studios NYNY (as told to me by his Asst). He's engineered many hit records. Won a Grammy as well.

Are you trying to tell me that he won a Grammy because he records at 88.2k? Or that recording at 88.2k had anything to do with this at all? Since I'm pretty damn sure that there have been plenty of grammy award winners who've not even used digital kit to record on, I don't see the relevance of quoting names of people who allegedly do this. Am I supposed to be impressed?

Having a 'hit' record has far more to do with marketing than anything else. In the UK we've had 'hit' records made with low-rate consumer kit in people's bedrooms - and this is from people who have gone on to be established artistes. But what gave their material 'hit' potential as sure as heck had nothing  to do with the sample rates they used!

Glad to know Audiophile 2496 cards work at 9632 in AA 2.0. Mona Don't. More and more everyone says it's a driver pblm. I'll deal with it. Thanks.

Noise floor is interesting to know about. This is how I learn. Get Steve G all pumped up and the torrent of gleaming knowledge he throws out are diamonds to keep forever!!!!!
 
You're right. Springsten (USA) recorded Nebraska on a 4 trk reel to reel. But then it was analog reel to reel 15 IPS quality. Even if magnetized particles are not perfectly continuous, they are aligned longitudinally with the tape not samples bunched together. Analog tape is not digital.

My point mentioning the NY Producer / engineer above was he has credibility and works at high sample rates. If he's won a Grammy, he knows a thing or two how to put it all together. Or maybe he should read this forum and really learn (like I am) that it's the sample rate converter that's key.

Brewer

My point mentioning the NY Producer / engineer above was he has credibility and works at high sample rates. If he's won a Grammy, he knows a thing or two how to put it all together. Or maybe he should read this forum and really learn (like I am) that it's the sample rate converter that's key.

Actually, God himself would be having the same problem as you if he were trying the same thing with the same gear as you ..right? But I was taken aback with the comment that Steve made about "why should anybody want to record at 96" comment, and this was not really backed up .. yet, and I suspect that you were responding to that comment. So, Steve, why shouldn't anybody even bother recording at 96 .. heck, there's Frank Fillippetti over there in the far flung bloody colonies who has the audacity to do this, and he's even won awards! 

But I was taken aback with the comment that Steve made about "why should anybody want to record at 96" comment, and this was not really backed up .. yet

Actually it was - thoroughly. But not in this thread. You need to read this one  all the way through, where it was effectively torn to bits by no less than three (or was it four?) people - I was only one of them.

[not]

As for all of the arguments about Famous Producer x records with a 768k sample rate, or whatever, and that's the reason he's had hits - well, none of them stand up to even the slightest bit of scrutiny as an inherent reason, but there may be other considerations...

He may, for instance, have exactly the same problem that Brewer's got - he can actually hear a difference. Why? Well for a start, most Famous Producers are having to use ProFools - and this is widely acknowledged to be sonically flawed anyway. That's why there is a buoyant market in selling passive external summing mixers for it - Digidesign don't allow other hardware access to their system, and the only thing you can do to improve the situation is to do your mixes from the individual channel outs. It wouldn't surprise me in the slightest to find that it sounds even worse at lower sample rates.

But - and it is a very important but - this is emphatically not an inherent fault of digital systems. It is a fault in some systems in particular, and because the manufacturers insist it's not their kit causing the problem (despite the fact that it clearly is), people blame the next most convenient thing. And this has been the butt of a lot of criticism almost since the dawn of digital audio - the damned sample rate. And yet they have absolutely no justifiable grounds whatsoever for doing this. Yes, a lot of early systems sounded crap, and they still do - but this is because they are crap, not because of digital sample rate limitations.

You only need to listen to 44.1k audio that's been digitised and played by an Apogee converter to realise how bad some of these other systems really are. I suspect that if you do the same thing with Lavry converters, you get the same revelation, although I haven't tried it with theirs. But it's the very small number of converters around that do this that absolutely prove beyond a shadow of any doubt that it's not the sample rate that's an issue here.

And as I said before, I can play you a competently made recording here - and there is no way that you could tell me just from listening to it in isolation what the sample rate was. And this has been proven time and time again by loads of listening panels in ABX testing. And this fact alone demolishes all arguments about famous producer x and his sample rate - if it's undetectable, then it's not inherently a part of the winning formula. All it shows is that Famous Producer x realises that at lower rates, he's got crap equipment.

Yes, even in the colonies they appear to have ears... 

What I said: 1. Famous producer X uses high sampling rates 2. Famous producer X has credibillity in the music business 3. Credibility was determined by Famous producer X winning a serious award.  Never said the reason Famous X had hits was due to high sampling rates, or Famous producer X won Awards due to using high sample rates. Next, I'm very much interested in using lower sample rates - if it's really true converters make such a difference - as pointed out here.

That said, I'd like to compare the Mona sound card with another at 48 kHz. Fair enough? Spoke with Marcel (Tech Support) at EchoAudio. He says audio recorded at 44.1kHz is similar to higher sample rates. He further said AA 2.0 having serious noise at 96kHz 32bit is an Audition 2.0 problem, not an ASIO driver problem. Finally, the reason Mona was discontinued was margin was too small, not that component parts were inferior quality. Email Marcel:  Marcel @ echoaudio . com

All it shows is that Famous Producer x realises that at lower rates, he's got crap equipment.

Forget Famous Producer X for a moment. Checking my studio: I capture sound with quality mics: SM57 / Beyer M88 / Neumann TLM103. I make sure sound does not clip before ADDA by having audio pass through DXP 160A limiter/compressor (gentle settings). I use Adobe Audition 1.5 / 2.0 / CEP 2.1 software to record sound. Harddrives are 7200 rpm EIDE Maxtor drives. Lots of space for recording. PC: Del PIII 1 GHZ / 512 rambus ram. Is this crap equipment?

He may, for instance, have exactly the same problem that Brewer's got - he can actually hear a difference.

Maybe people here have audio hearing loss? That happens.

Yes, even in the colonies they appear to have ears... 

But more to the point is how we use the space between them.

Brewer

[He further said AA 2.0 having serious noise at 96kHz 32bit is an Audition 2.0 problem, not an ASIO driver problem.]

That said, I'd like to compare the Mona sound card with another at 48 kHz. Fair enough? Spoke with Marcel (Tech Support) at EchoAudio. He says audio recorded at 44.1kHz is similar to higher sample rates. He further said AA 2.0 having serious noise at 96kHz 32bit is an Audition 2.0 problem, not an ASIO driver problem. Finally, the reason Mona was discontinued was margin was too small, not that component parts were inferior quality. Email Marcel:  Marcel @ echoaudio . com

Yeah, like he was going to say "Sorry Brewer - you've got us bang to rights - our card is old, and the driver has problems"... Get real. This guy works for Echo, for heaven's sake, and they give him a paycheck at the end of the month. What did you expect  him to say?

Try asking him a different question. Try, for instance "Audition 2.0 works fine at 96k with a MOTU Traveler - no excess noise at all. Why doesn't it work fine in the same way with an Echo Mona?" Let's see him crawl out of that  one quite so easily. If you are going to interview the guy, at least try to make him earn  that paycheck...

So let's get this quite clear - it works for me, and loads of other people using AA2.0 not using Echo Monas, but using ASIO. It's not Audition that's at fault - QED. If it was that screwed up, do you seriously think that the BBC would have bought x-thousand seats of it???

Yes, even in the colonies they appear to have ears... 

But more importantly is how we use the space between them.....!!!!!!!!!!

Yes.... 

48 kHz Recording


Check this out: Pushing the Limits of Native Processing / Toney Nunes / Mix July 2007
"To me, record production at 44.1/48 kHz is passé; at higher sample rates of 88.2 kHz and above, there is a definite sonic improvement. "

[never]

"To me, record production at 44.1/48 kHz is passé; at higher sample rates of 88.2 kHz and above, there is a definite sonic improvement. "

Either crap equipment again, or more likely absolute bollocks. He couldn't tell in a blind test either.

It's not worth quoting loads of people saying this - there has been a campaign by manufacturers for years to promote the idea - because it sells more kit. And that is all they are interested in at the end of the day. And they don't care how much mischievous nonsense they spread around to do it, either. It usually goes like this:

"What? You can't hear the difference? Then you must be deaf! Of course  it's clearer/more detailed/insert your own BS - it's going faster!"

Now, nobody will ever admit to being deaf in this context, so the manufacturers win all the way. And eventually all the people that they CON start to believe it. But, as has been pointed out in the other thread, in ABX tests this never stands up to any real scrutiny.

I'm not trying to sell you equipment - I have no axe to grind other than a desire for a little more logical behaviour from that bunch of sheep called equipment users, and the consequential evil wish to pull the wool from over their eyes/ears...