AHHHHHHH! I think I've reached level 5!!!!

Bob Katz may offer a clue. Here's what he says in "Mastering Audio":

...the so-called "dramatic" differences people hear between sample rate systems are not likely to be due to bandwidth, but probably to the filter design itself.

I also looked up what Nika Aldrich says about it in "Digital Audio Explained":

...there can be audible differences between lower sample rate and higher sample rate recordings...but only insomuch as the quality of the filters in the converters varies. The notion that there is an inherent difference is false, as with properly designed, audibly transparent equipment, no difference will be audible.

So it sounds like I accidentally hit the nail on the head myself when I said that "If you can hear a difference, it's due to inadequacies of your hardware." It may be that DVD players, being more expensive, just have higher-quality filters than your average CD player, and that's why they often sound better.

(Still doesn't answer your question, though, does it?)

ORIGINAL: bitflipper

Although I can objectively make the case that there is no difference, DVD audio often does sound better to me. I remember the first time I watched a DVD movie on a nice home theater setup - it was the movie Twister, which won an Oscar for sound effects. I went out and bought my first DVD player the next day, mainly for the sound.

So why does it sound better? Don't know. In fact, quite often it does NOT sound better, but I think that's due to a lack of audio standards for DVD (compared to theater presentations). When it sounds "better" is it an illusion, maybe due to hyped frequency response and/or high compression ratios? Don't know.

It could just be that the studios doing 96Khz mixes for DVD tend to have experienced people and have all the right gear and acoustics to do great mixes? Or maybe just more attention is being paid to the 96Khz mixes?

As for SACD, I think that's pure marketing hype. But that doesn't explain why higher sample rates are even used on DVDs, a practice that limits the number of audio channels you can include on the disk.

SACD came out more or less when the patent on CDs ran out. Coincendence? Yeah right...

UnderTow

I record at 88.2khz in the studio pretty much all the time now. The reason is not that the audio itself sounds any different, but that the plugins sound so much better at higher rates.

Sonitus EQ for instance gets a great top end at 88.2

Kind regards

Dave Rich

So it sounds like I accidentally hit the nail on the head myself when I said that "If you can hear a difference, it's due to inadequacies of your hardware." It may be that DVD players, being more expensive, just have higher-quality filters than your average CD player, and that's why they often sound better.

(Still doesn't answer your question, though, does it?)

Actually it does. That's exactly what I suspected the culprit might've been. Still, there should be no reason why we need a higher sampling rate format if all that is requiered is better converters in our CD players. Can someone say marketing?

I have recorded at 44.1, 48, 88 and 96. I can't tell any difference between 88 and 96 (so usually use 96kHz between the two, personal choice). Between 44.1 and 96, as much as I kept telling myself that theory says I shouldn't be able to tell them apart, I actually can on the drums, namely the cymbals. At 96kHz the cymbals seem to sparkle more.

On the gearslutz forum do a search for sample rate. There were many tests and some top engineering and gear designers have left comments. Basically with today's algorithms it seems that you shouldn't worry about 96kHz to 44.1 or 48 to 44.1, the conversion will be fine (i.e. no need to use 88.2 so that it's nicely divided etc). As far as using 96kHz, the aliasing curve is smoother than at 44.1 and this would be why 96 sounds better than 44.1.

I left a lot of details out so if you want more do the search as I mentioned above.

Cheers,

ORIGINAL: tazman

I have recorded at 44.1, 48, 88 and 96. I can't tell any difference between 88 and 96 (so usually use 96kHz between the two, personal choice). Between 44.1 and 96, as much as I kept telling myself that theory says I shouldn't be able to tell them apart, I actually can on the drums, namely the cymbals. At 96kHz the cymbals seem to sparkle more.

On the gearslutz forum do a search for sample rate. There were many tests and some top engineering and gear designers have left comments. Basically with today's algorithms it seems that you shouldn't worry about 96kHz to 44.1 or 48 to 44.1, the conversion will be fine (i.e. no need to use 88.2 so that it's nicely divided etc). As far as using 96kHz, the aliasing curve is smoother than at 44.1 and this would be why 96 sounds better than 44.1.

I left a lot of details out so if you want more do the search as I mentioned above.

Cheers,

I record at 88.2 merely because I can hear no difference between 88.2 and 96khz and I'd rather eek out the cpu a bit ;)

Kind regards

Dave Rich

Increasing sampling rates in converters has everything to do with marketing. But what Dave says is also true: Some plugins do sound smoother at higher sampling rates. I prefer to carefully choose my plugins rather than go to higher sampling rates for the entire project.

UnderTow

I record at 88.2khz in the studio pretty much all the time now. The reason is not that the audio itself sounds any different, but that the plugins sound so much better at higher rates.

Interesting. That would actually be a good reason to record at higher sampling rates. Also having a higher resolution format would be justified since it eliminates an extra step like SRC that would actually introduce artifacts to the audio (read aliasing, ringing, phase, etc).

Here are couple nice pictures showing the specal data for recordings using 96kHz:

E-MU 0404 USB @ 24/96:

http://img455.imageshack.us/img455/1327/auditionspectralchangelmt0.jpg


SB Audigy 2 @ 24/96:

http://img503.imageshack.us/img503/7579/sbaudigy22496recordingzn6.jpg


NOTE: I don't know the E-MU source but, the Audigy recording is from vinyl (Toto: Fahrenheit: 1st minute from track called "Without Your Love").

As showen, there are some life on Mars.


Junski

EDIT:

For comparison sake, here is the same sample audio as ripped from CD and as recording using 16/48 (Audigy 2):

CD - rip:

http://img263.imageshack.us/img263/843/spectrumcd1648es2.png

and

recorded from Vinyl:

http://img339.imageshack.us/img339/2391/spectrumvinyl1648rn5.png