96k or 192k?

  • Thread starter Thread starter Purple Pig
  • Start date Start date
P

Purple Pig

New member
Am wondering how big a difference these are. the +'s and -'s for them. does the higher # effect latency?

My cpu is a 3.0, w/1gig of ram. at the moment, I am about to upgrade it to a 4gig ram.

Am looking at the firestudio(96k), and the motu mk3(192k)
Thnx for your help.
 
Which MOTU? They've got like four different devices with a mk3 version
 
I can't tell the difference between 192 and 96, except that 192 is m-m-m-massive on the hard drive.

I actually think 44.1 is probably fine for what I do, but I use 96 anyway for two less-than-completely-compelling reasons:

a) I did an experiment with my electric guitar plugged in using modeling software on the computer and strummed a chord and let it fade out wearing headphones. I could definitely tell the difference between 44.1 and 96 as the chord was about to fade away completely (the 44.1 seemed to deteriorate into bitcrusher-like distortion). Since then, I've upgraded my equipment a little, and haven't re-executed the test. I've read that it was more likely low quality gear than the sampling rate that caused that problem, and

b) I run a couple of converters together a lot with an external clock, and for some reason they only like to play together at 96
 
MOFO Pro said:
Which MOTU? They've got like four different devices with a mk3 version
Click to expand...

I am looking at the MOTU UltraLite mk3, vs. the firestudio project. I already had the firestudio here a few days ago, and aparently it was doa, as the first channel would not show any signal, and that shouldnt have had anything to do w/the pci card, as it is supposed to be able to stand alone. They are sending another here, and if it has the same issue, I am looking at the motu I mentioned above.
It is very confusing, as most any interface I have researched I have found issues. I wish there was just a simple way to go w/out the problems. :eek:
 
Daveg62 said:
Honestly, I'm more than happy with my lowly Lucid converters @ 44.1.

Here's some good reading :)
(cant post outside url's till I post 500 times!!)

thanx for that link. It was very helpful.
Click to expand...
 
Upon further investigation, even with the obscenely high sample rates available and incredibly cheap digital storage, you'll still find that around 70-80% of full-time industry professionals record at the target rate -- 44.1k for audio, 48k for video.

With good converters, if you can't make an absolutely stellar recording at 44.1kHz, upping your sample rate isn't going to help - I guarantee it.
 
You might want to try 88.2khz IF you are doing sound with extremely wide dynamic contenct - very soft jazz or quiet orchestral or choral pieces.

Otherwise, you're just chasing number for bragging rights and loading down your computer for nothing.....
 
Purple Pig said:
(cant post outside url's till I post 500 times!!)

thanx for that link. It was very helpful.
Click to expand...

(I think it's only 5 posts)

Dan Lavry knows his stuff; anything he says is gospel. I've got the black series. (Yup, I'm, bragging a little) :D

Massive Master said:
Upon further investigation, even with the obscenely high sample rates available and incredibly cheap digital storage, you'll still find that around 70-80% of full-time industry professionals record at the target rate -- 44.1k for audio, 48k for video.

With good converters, if you can't make an absolutely stellar recording at 44.1kHz, upping your sample rate isn't going to help - I guarantee it.
Click to expand...

TimOBrien said:
You might want to try 88.2khz IF you are doing sound with extremely wide dynamic contenct - very soft jazz or quiet orchestral or choral pieces.

Otherwise, you're just chasing number for bragging rights and loading down your computer for nothing.....
Click to expand...

Man, now you guys have me rethinking my approach. I've always recorded at 24/96 because I thought ME's liked it that way. But if I can save space and cpu cycles, I'd love to cut that down to 24/44.1.

Peace.
 
Last edited:
I agree about being able to make a great recording at 44.1K, but it's not always going to be the target rate-someday we'll look back at it and laught (many are already doing so).
 
if ya read lavreys paper you'll find he thinks the "perfect" rate is 60k...

and since most all of the current crop of convertors are the super-fast one bit type that then use reconstruction filters the point is pretty moot...
 
craigmorris74 said:
I agree about being able to make a great recording at 44.1K, but it's not always going to be the target rate-someday we'll look back at it and laught (many are already doing so).
Click to expand...

yea, well, until another physical format becomes widely accepted in place of CD's, or until digital downloads are sold almost exclusively in 24-bit formats, the target rate in most instances will continue to be 16/44.1...
 
But why record at less than optimum fidelity because people choose to listen to recordings in less than optimum fidelity?

Especially when recording important projects, consider when remastering takes place 20 years from now. I think artists/engineers will be glad they chose higher depth/sample rates.
 
TimOBrien said:
You might want to try 88.2khz IF you are doing sound with extremely wide dynamic contenct - very soft jazz or quiet orchestral or choral pieces.

Otherwise, you're just chasing number for bragging rights and loading down your computer for nothing.....
Click to expand...

Increasing sample rate will not improve dynamic range. Only bit depth (or word length) determines how accurately the amplitude is recorded. So an increase from 16 bit to 24 bit can make dynamics more accurate.

craigmorris74 said:
But why record at less than optimum fidelity because people choose to listen to recordings in less than optimum fidelity?

Especially when recording important projects, consider when remastering takes place 20 years from now. I think artists/engineers will be glad they chose higher depth/sample rates.
Click to expand...

You must not have read the paper on Sampling Theory, (the link above.) Recording at 44.1KHz IS the optimum sample rate. The paper tries to show that recording at higher rates tends to introduce unwanted distortion due to the nature of hardware processing.
 
Buy a "decent" set of converters and record @ 24 bit 44.1. Your cpu will work better, you can pile on the plug-ins and you'll have plenty of hard drive space.
There's no advantage to buying cheap conversion that does 96 or 192 when the better converters sound much nicer @ 44.1. I've already been down that road and I've done direct comparisons. I was blown away when I finally heard a good converter. Not even close.
 
Daveg62 said:
Buy a "decent" set of converters and record @ 24 bit 44.1. Your cpu will work better, you can pile on the plug-ins and you'll have plenty of hard drive space.
There's no advantage to buying cheap conversion that does 96 or 192 when the better converters sound much nicer @ 44.1. I've already been down that road and I've done direct comparisons. I was blown away when I finally heard a good converter. Not even close.
Click to expand...

What are you calling decent/good?

I use RME and I think it sounds good.
 
RawDepth said:
What are you calling decent/good?

I use RME and I think it sounds good.
Click to expand...

Well I think it's pretty much relative to what you can afford. I used a Multiface for quite a few years and was happy with it. I then bought a fireface to replace it and thought it sounded just as good, and a few weeks later bought a Lucid 8824 to add 8 more channels to it. Comparing the Lucid to the fireface was what blew me away. It was like taking a blanket off the speakers. I quickly sold the fireface, bought an adi-8 and an RME adat card to run the Lucid and the adi-8. The adi-8 is much better than the fireface, but still not as good as the Lucid. If I could afford the Lavry's, Myteks, Cranesong's , or Apogees maybe they would probably be even better.
 
Just because one paper claims that 44.1 is the optimum sample rate doesn't make it a fact. I've read a couple of published pieces that measured potential for greater error at 192k. Some converters DO perform worse at 96K (ones that use a mutiplier by 44.1 to achieve 96k). My converters do not.

Again, my contention is this-and it is supported by the Lavry paper if you read it closely-there is a better sample rate than 44.1. Someday listeners will be able to take advantage of this. I can't make myself record at less than optimal rates.
 
craigmorris74 said:
Just because one paper claims that 44.1 is the optimum sample rate doesn't make it a fact. I've read a couple of published pieces that measured potential for greater error at 192k. Some converters DO perform worse at 96K (ones that use a mutiplier by 44.1 to achieve 96k). My converters do not.

Again, my contention is this-and it is supported by the Lavry paper if you read it closely-there is a better sample rate than 44.1. Someday listeners will be able to take advantage of this. I can't make myself record at less than optimal rates.
Click to expand...

I agree, but that optimal sampling rate is probably 88.2 or 96. It ain't 192kHz. Actually I like 64kHz, but that would just confuse a lot of gear out there.
 
mshilarious said:
I agree, but that optimal sampling rate is probably 88.2 or 96. It ain't 192kHz. Actually I like 64kHz, but that would just confuse a lot of gear out there.
Click to expand...


I agree 100%, and I think that was the point Mr. Lavry is trying to make-192kHz is pointless, not that 44.1kHz is best. That's why he only makes converters that sample up to 96kHz.
 
You must log in or register to reply here.
Back
Top