Calling Digital Experts! What Is 96khz Sampling Frequency, Really?

Barefoot,

Please DO go for it. All it would take is the cooperation of a moderator in each forum topic. You could even start one of these threads without the moderator's help, but then you wouldn't be able to lock it and stick it to the top of the forum. What kills these type of "library" threads IME is that if they are not kept locked, they get all clogged up with replies, and then you can't find anything. Perhaps Dragon or the Mod in question could give temporary mod powers to anyone creating one of these threads?

Trebles,

I hear what you're saying, but at Talkbass at least, measures that drastic have not been needed so far. Whenever someone posts some kind of "repeat post" there, I usually just point them to the links thread and ask that they read all existing material about the subject before posting a new thread. Usually when I do this, the poster in question will either reply to an existing thread, or create a new one asking different questions than have been asked so far. It's not a perfect system, but like I said, it's cleared out much of the annoying repeat postings since we started implementing it.
 
The participation of some of the more knowledgably members like sonusman have even dropped off significantly because of their frustration around this issue.
Click to expand...


sonusman i believe has departed already sad but true im sure he too was just plain fedup anyway check out

https://homerecording.com and you will see that all his work he had there is all gone now from our main page
 
something I thought you guys should know

I recently Read at tom'shardware.com or audiophile (or some other website I was reading at 4AM) a thorough review of the SoundBlaster Audigy Vs some 24b/96k sound card that I can't remember offhand. The most interesting (and somewhat alarming) thing that they said was that THE SOUNDBLASTER AUDIGY DOWNSAMPLES INPUT FROM 24 BIT TO 16 BIT, THEN RESAMPLES BACK UP TO 24 BIT. I know from personal experience that the audigy sounds great, but wouldn't this lead to more error when applying effects/ eq'ing? anybody's thoughts on this?

UPDATE: I believe I remember the card is was being compared to, it was the Terratec EWX24/96. the Audigy was deemed the winner greatly on features, but the Terratec had a more accurate representation of all frequencies, and a lower S/N ratio
 
something I thought you guys should know

I recently Read at tom'shardware.com or audiophile (or some other website I was reading at 4AM) a thorough review of the SoundBlaster Audigy Vs some 24b/96k sound card that I can't remember offhand. The most interesting (and somewhat alarming) thing that they said was that THE SOUNDBLASTER AUDIGY DOWNSAMPLES INPUT FROM 24 BIT TO 16 BIT, THEN RESAMPLES BACK UP TO 24 BIT. I know from personal experience that the audigy sounds great, but wouldn't this lead to more error when applying effects/ eq'ing? anybody's thoughts on this?
 
Hmmm.

I'm no genius on this one, but I'll go out on a limb, and say you're correct. I'll even go out on another limb here and say that anything from Soundblaster should come with a warning label: WARNING: This card is for gaming and self-amuzement, and should in no way be confused with a real sound card.
 
Downsampling

Actually, I guess the more correct term for going from 24-bit to 16-bit would be "truncating" (or, if it's more fancy than that, "dithering"). To go from 16-bit to 24-bit, it would be "padding."

In general terms (this isn't correct in all details), what you do when you truncate is treat a value that's been written in 24-bit format as, say, .239787, and turn it into .2398 in 16-bit format. So ... what do you do to change a 26-bit value into a 24-bit value? Not much you can do besides make it .239800 or, if you prefer, .239855.

You can pick up something of value doing this, though it isn't obvious. If you have a bunch of data of relatively lower precision that you want to manipulate (like in a digital mixer or effects box), you can do so more accurately if you use more precise values in the course of doing the calculations, even if you truncate back to the lower-precision format at the end. In fact (I think, I might be wrong about this), if you lose precision in the course of doing your data manipulation, any extra precision you have in the input data would be wasted anyway.
 
What I've noticed about the subject at hand (all the noise aside) is that if your target format is 128Kbps mp3 and you're not applying a serious shitload of software effects to your .wav file after the tracking is over then 96 KHz is a waste of time and in some cases, money: you do the math!

Just my opinion. And I record at 24/96 for the same reason a dog licks his own balls! The extra burden is no big deal on a decent P4 platform. Whatever organizational skills that apply to extending your HD capacity under 16/44 also apply to 24/96.
You have to store the product of your labor sooner or later, wipe the drive and move on. It's happened on every system I've ever owned.
 
You must log in or register to reply here.
Back
Top