sampling rate explained? 48 vs 96 khz

C

CrayPhish

New member
My new question (from another thread that got lost/buried in the long line of replies):

Is there an audible difference in the digital sampling rate from 48 to 96 khz? I'm looking at the midiman Dio2448 and dio2496 models respectively and both have the features I need, just the price and sampling rates differ.

http://www.midiman.com/m-audio/product.htm

I searched the bbs but could not find any similar discussions previously, so if you know of some point them out to me.

Thanks!
 
If you can afford the 96k card... get it. Even if you don't use 96k right away. The sampling rate is how often the card samples the analog signal per second. The higher the sampling rate the better the resolution. 96k vs. 48k won't be as much of a difference as 24 bits vs. 16 bits, but it will be a difference. On the downside, a 96k track will take twice as much drive space as the same length track at 48k (twice as many samples per second), and DSP processing will take a performance hit as well. But computers are getting faster every day, and hard drives are getting bigger. Get the card that will serve you longer.
 
better for effects

From my understanding, higher sample rates realy improve the quality of digital effects.
 
CrayPhish,

The higher the KHz the better the sound quality, the resolution and sound after the mixdown will be better than sampling at 48KHz. I sample at 96KHz


Hope this helps you out.


Carter
 
I should have been more clear. I'm wondering if all i intend to do is export a stereo signal from my digital four track for the sole purpose of burning it to CD (which I understand CDs have limited rate of 44.1?) then why would I need the 96 khz? I will be mixing everything before it gets to my PC. Thanks again
 
The human ear cant even hear 44 khz....
Perhaps a bit of one upsmanship, but for what?
Your ears cant hear it. I dont think most microphones would even pick up those type of frequencies(?)...
Sure people have vague theories about sound resonating your skull and phantom headroom. Id say its not important really weather your soundcard is 48, or 96.
Its called marketing.
Dont beleive the hype.
 
Well- robert, what you meant to say was the human ear (usually) can't hear 22KHz, which is the ~the highest theoretical frequency that can be captured sampling at 44KHz. But the other comments about subsequent processing of the digital signal make a lot of sense. If you're just recording to digital and burning to CD: sample at 44.1 and cut out the middleman (downsampling).
 
interesting topic. a guy with a lot of time on his hands thought so too...

http://www.nanophon.com/audio/

it's a very technical article in pdf format titled '
The benefits of 96kHz sampling rate formats for those who cannot hear above 20kHz'. jump to page 8 for the conclusion.

micmac
 
Yeah, Dunn is an interesting character. These are good papers, and they have a lot of meaning to the high-end audiophile population. However, for us hobbyist type recordists, how much of that is really applicable?

I think we all have to try it out with our rigs, and see if the bang for the buck is there. I've already done this in my room with a very quality-concious audience, and concluded that for the type of music I've working with, and with the current run of commercially-available hardware I own, it is *not*. For me, the bus stops at 48kHz- even though all my gear can go to 96.

It seems that a goodly number of pro studio purchasing people also agree (not that that really matters here). In that previous thread, https://homerecording.com/bbs/showthread.php?threadid=21183 , Tom Cram of dbx commented that the professional market is being very slow to pick up on the 96kHz option in their high-end pro gear. The demand just isn't there. As pglewis points out, it may be some day- but it isn't, yet.

I've been recording with very good oversampling converters and using 48kHz. I personally *can* hear the difference between 44.1 and 48. I just flat cannot hear the difference between 48 and 96, and the additional storage costs double for the doubled sample rate. In my experience, the bit depth is more important than the sampling rate with currently available hardware- so I'll stay with 24bit/48kHz for my room, and downsample to 16/44.1 only at the last possible minute just before going to CD.

However, your mileage certainly may vary. As it turns out, all my gear uses the AK53xx-family oversampling converters, and they seem to do the job very well. If your gear uses different converters that have different implementations of their internal digital decimation filters, *you may very well* be able to hear enough of a difference going to 96 to make it worth your while- that's Dunn's assertion in his paper, as well. Hell, you might very well be able to listen on my rig and hear that much difference! Everybody knows that I'm old and going deaf...

Bottom line: All that matters is what *your* ears hear on your rig in your room, and the buzzwords just don't freakin' matter- especially if, as a hobbyist, you aren't using the buzzwords to inflate the numbers on your ratecard (;-).

Give me Dunn over Marsh and Jung any day. At least Dunn does decent research, and doesn't just fall back on the "I can hear it, and I don't have to measure it!" arguments... But don't get me started on audiophiles: you know how I get. You can do the search on "skippy" and "audiophile" if you want to see that stuff (;-).
 
Okay.
I do understand that there IS *some* benefit. This would be that just because your ear cant hear certain frequencies doesnt mean the vibrations in those frequencies dont effect the vibrations you do hear(my pseudo explanation).

Still. I forward that for vast majority of users in this forum those *ultra subtle* benefits wont reveal themselves for many reasons(room noise, cheap moniters, cheap mics, cheap pre amps, cheap mixers). Lets not forget also that were still in the cd age. Now dithering down from 88khz does have advantages. I dont think most of us have the type of equipment to really see the benefits from 96 khz;these advantages existance is somewhat debatable in real world application. Add to this room noise from recording in your basement, the fact that dvd's are not an audio standard(yet)...

Okay, so if the benefits are subtle, on genelecs, what about
some guys living room speakers? There in that environment its not a noticable difference.The difference between 24 bits and 16 IS quite noticable on any old stereo, 46 khz to 96khz? I dont think so. Also, when you have set reference points you can listen for there may be a noticable difference if for nothing else the plecebo effect. But, if your just listening and dont have any reference such as that then 96khz with or without is a non issue.

This is my inderstanding. Now if I was listening to a dvd and I had a $3,000 surround dvd audio system and I was in the most quiet room imaginable where we were conciously a/b ing different parts the difference woul be very subtle, really VERY subtle. Maybe an engineer with GREAT ears could tell the difference right away. If it was a bling folded "pepsi challenge" its not a given that they could.

So, to me 96khz is NOT a selling point. In essence we cant hear the difference.Maybe some people can in a top flight recording studio if they hold their breath and close their eyes. It seems to me a waste of cpu to work with that.
 
Robery Jaybird wrote, "just because your ear cant hear certain frequencies doesnt mean the vibrations in those frequencies dont effect the vibrations you do hear(my pseudo explanation)."

I remember sonusman wrote about the same thing several months ago and I posted a reply asking for more details behind the mechanics of this phenomenon or a pointer to a reference to no avail. If someone can explain how higher frequencies outside of my hearing range can manifest themselves on the lower ones that I can hear, please, *PLEASE* do--I'm all ears. (Pun intended.) Likewise, pointers to any articles or essays that explain the same are welcome. Thanks!
 
OK, I'll be real specific this time (third time's a charm, but it may not be the last)

I have a fostex FD4 digital four track that samples up to 44.1 khz

My intent is to do all my mixing via the fostex and outboard effects

I need a method to burn CDs from this medium and I have a PC with a kick a$$ cdrw drive

Why would I ever need a sound card that sampled at 96 khz? The one I'm looking at will 'clock down' to the digital signal I'm sending it which will max at 44.1. What inputs could possibly synch to 96 khz these days?

Can anyone give me a compelling reason to spend the extra $110?

(I've done a bit of research since originally posting to gather these facts, so it wasn't an intentional withholding of info, I just want to make sure I'm not making a mistake here)
 
There isnt much reason to....unless you r looking ahead way in the future to burn DVD's (24 bit/96hz).......I'd go with the dio 2448.... it will serve u well......
 
Here's a good reference on bit depth issues, dither, and why you might possibly want to record and mix at 20-24 bits, and dither down to 16 only at the last possible minute before buring a CD.

http://www.digido.com/ditheressay.html

I've said my piece (and embarassed the hell out of myself) on sample rates already, and I leave it as an exercise for the interested student to decide to agree or disagree with my personal tradeoff of bit depth versus sample rate. Yours may vary, of course. I'll shut up now...
 
You must log in or register to reply here.
Back
Top