What sample rate do you record at?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Freddy G

Well-known member
Joined
Jun 4, 2004
Messages
463
Location
Canada
I'm just getting ready to record a band next week and I've been thinking about what sample rate to record at. I'd love to take a poll here, what do you record at?!
In the past I've always recorded at 44.1k and 24 bits. What do you think about 96k? Does anyone know if the higher resolution is not worth the conversion back down to 44.1k for CD production? I'm thinking that 88.2 makes sense because the conversion algorythm should be very simple right? Just remove every other sample.....or is it not that simple? :-\
Freddy
 
"I'm thinking that 88.2 makes sense because the conversion algorhythym should be very simple right? Just remove every other sample.....or is it not that simple?"

doesnt matter until we can throw enough cpu power at it, chekc this  interview btw:
http://www.gearslutz.com/board/music-computers/93177-daniel-weiss-interview-weiss-07-11-2006-a.html
 
I noticed a big improvement when I went up from 44.1 to 96. I'm sure 192 blows it out of the water but 96 is fantastic. I think 96 was the one that broke the "tape barrier".
 
abby normal said:
I noticed a big improvement when I went up from 44.1 to 96. I'm sure 192 blows it out of the water but 96 is fantastic. I think 96 was the one that broke the "tape barrier".

But how about diminishing returns ?

The 'Katz book' (=Mastering Audio) has a chapter on sample rates; IIRC it argues that it's not the
sample rate itself but the associated filter anomalities that make people draw conslusions w.r.t. sample rates.

From a certain point on (let's say 96) these 'issues' become largely irrelevant, but hey, manufacturers want to sell you double-capacity gear each few years... so don't be a s*cker and upgrade your present and perfectly fine gear... 
(I hope I summarize it correctly enough here)

In other words (mine this time): it might be best for our climate if the act of doubling sample rates each few years becomes unlawful ASAP, otherwise we keep running electronics harder & keep spinning harddrives much more than they actually need to.

Regards,

  Peter
 
clintrubber said:
The 'Katz book' (=Mastering Audio) has a chapter on sample rates; IIRC it argues that it's not the
sample rate itself but the associated filter anomalities that make people draw conslusions w.r.t. sample rates.

From a certain point on (let's say 96) these 'issues' become largely irrelevant, but hey, manufacturers want to sell you double-capacity gear each few years... so don't be a s*cker and upgrade your present and perfectly fine gear... 
(I hope I summarize it correctly enough here)

That makes a lot of sense. I'm not sure if there really is a need for me to upgrade to 192. The 96k boxes are stellar sounding. I believe you are right about them just wanting to sell you the latest and greatest. If everyone sat on what they had they would go out of business. I'm going to to some sitting for a while though.  ;)
 
I try and use tape as often as possible which has no sample rate. When I use the box I usually do 88.2 as when it down samples to 44.1 the remove an even number of samples and the conversion does very little to the sonics of the original mix
 
Thanks for the 88.2 tips guys. I've never heard that before. I just assumed standard dithering would compensate.
 
When Sample rate converting a file, there is no difference in the process when going from 96Khz to 44.1 or 88.2 to 44.1
At first sight you may think that 88.2 to 44.1 is just descarting very 2 samples, but in reality is not done that way. First the sound is filtered to aoid aliasing, and then It always resample to a very high arbitrary samplerate (even at Mhz. range), then the sound is downsampled to the desired sample rate.

I preffer using 44.1 when the final medium is 44.1 and 48 when is 48, what I`ve found is that the resample artifacts are worse than the diference between 44.1 and 48, so I skip resampling in those cases, but recording at 96 or 88.2 is a step ahead even with resample (but use a good SRC algorithm!)

:)Synthi
 
before you make that "SRC is EVIL!!!" assumption I'd like to remind everyone that the Benchmark DAC-1 (and their adc as well) , what is numerous times better converter than what 95% of the guys use/have here, has one SRC on its input, furthermore its _asynchronous_, doesnt care about the sample rate you give it AT ALL. It has one very nice attribute as well: it cancels digital phase noise aka. jitter in favor of a nice local clock...  Talkin about whether its not bit-perfect or what that means, I read that they made a listening test involving a group of people and the same ASRC process one after another, fifteen times , and it went unnoticed. Whoah. Pick your poison , what is the bigger culprit ? Digital interfacing jitter, or resampling artifacts ? What else has it? Lets see.. Cranesong rings a bell? Humm....
 
one easy solution:
mixing / summing analogue and you don't need sample rate conversion ;D
you can even use all your cool DIY gear :eek:
and you can go AD with whatever you will need, even at the same time ;)
 
Hi,

  Just started using 96kHz, depending on the project and expected track-count. Even with Protools HD3 Accell is seriously limited by double sample rate! If not, 48kHz, NEVER 44.1kHz if I can help it. 48kHz is SOOO much better. I always Master at 96Khz via Mini-Me on my laptop, even on a 44.1kHz session, since i mix on an analog desk, never in the box.

  I don't bother with 88.2kHz. All my work is mastered analogue, so no point in dropping from 96kHz.


    Kindest regards,


    ANdyP


     
 
If I remember correctly, all sample rates are up-converted to a higher rate, much higher than 192 then divided down again.
And you should consider the equipment as well.  44.1 on Apogee is of higher quality that 192 on MOTU or Mackie.  If you have the power, try 88.2.  While you are getting sounds, you can try and compare the higher rates.
Mike
 
Dan Lavry's $.02:

http://www.lavryengineering.com/documents/Sampling_Theory.pdf

As for me I use a DAW and mix through an analog desk - I use 88.2 - partially because of belief in the every other sample idea and partially because of a minor bit of drive space savings.  A noticeable (favorable) difference in sound between 44.1k and 88.2k when comparing multitrack playbacks.  On a theoretical level, I like to have as much information to work with as far down the chain as possible.  Of course, YMMV.

Frank
 
I would also recommend reading Dan Lavry's White Papers, especially his papers relating to 192kHz sampling rates.

http://www.lavryengineering.com/index_html.html

Click the "Support" link at the top.
 
Back
Top