Topic: 24-Bit / 48kHz or 96kHz

[hi and lo]

If I'm going to lug this gear out, and tape music that I like enough to spend valuable time and money on, I may as well record at the highest settings reasonably possible.

I think I most strongly identify with this statement. I don't always record at 24/96, which is my perceived highest reasonable setting, but I like to when I can. I believe I can afford the extra time required in post-processing, the extra disc storage space needed, and ultimately I enjoy the prospect of listening to my masters at the highest resolution possible.

I generally won't tape at 24/96 if I know I'm going to have to share my masters with other tapers or the band/label/etc simply because it's a chore to transfer such large files. 24/48 is far more convenient for just about everyone.

[adrianf74]

I think I most strongly identify with this statement. I don't always record at 24/96, which is my perceived highest reasonable setting, but I like to when I can. I believe I can afford the extra time required in post-processing, the extra disc storage space needed, and ultimately I enjoy the prospect of listening to my masters at the highest resolution possible.

I generally won't tape at 24/96 if I know I'm going to have to share my masters with other tapers or the band/label/etc simply because it's a chore to transfer such large files. 24/48 is far more convenient for just about everyone.

It's not that hard to run a 24/96 file through R8Brain (or some such program) to downconvert to 24/48 or 16/44 if you need to distribute a show (especially when you're running a fast enough PC).  I'm just toying with the idea of 24/96 for the shows where it might be worth it or just moving to it flat-out since I have the ability to.   I've always been one to say running more than 24/48 on the gear I'm running seems to be a waste but I know I have the ability to do it.  I'm almost tempted to record at 24/96 and then dumb-down to 24/48 for my "master" allowing me to keep the original files as is should I need to revist them down the road.   It also gives me the ability to master at 24/96 for the shows that truly deserve it.  Whether I'll "hear" a difference is to be determined and whether using twice the space is worth it is another question.

I know the M10 can do seamless file splitting with NO lost samples.  It's been reported on quite a bit in these parts.  All of this being added, this is why I'm thinking about it...

[SmokinJoe]

I almost always record at 24/44.  Why?  Because I know I'm going to be listening at 44k most of the time (CD's or 16/44 flacs on my iRiver). I think about the mathematical process of resampling from 48k or 96k down to 44k... all that mathematical interpolation to get to 44k, I think it has to sound more accurate if I go direct at 24/44 than if I record at 48 or 96k and resample.   My "proof is in the pudding" test was taking a 1bit recording from my MR1, letting audiogate save it to 24/44, then again at 24/48.  Resample the 24/48 to 24/44, invert and combine with the 24/44, so all that's left is the inaccurate little crumbs left over from resampling.  There is a lot more left over than you might expect.

Back when I had Earthworks (which are documented to be able to pick up soundwaves up to 30k), if I was recording onstage I felt good about recording at 24/96 thinking I might actually pick up some HF information above 20khz.  But if I am recording a PA, I don't believe there is anything above 20khz hitting my mics that I might capture, except possibly the gentle harmonics from "loud whistling guy" and "loud clapping guy".  They are recording the show with SM57's and CM81's, running through the board, amp, and PA... no way any HF information is making it to me.

Jon has a good argument about the quality of brickwall filters... which is better?  the one in my rig, or the one on my computer?  I don't know.

And I can't really hear 16k any more.  Not so much hear it as "feel that there is something there".

[adrianf74]

Thanks Jon and Joe.  I think between the two last replies, my question has pretty much been answered and then some.  I've also spoken with a few others outside of here and everything points to running beyond 24/48 as being a general "waste" of space when recording from a PA source (and possibly little benefit from a soundboard output as well) -- not to mention the issues introduced with SRC.  That said, I'll just continue to stick with 24/48 for now and maybe consider 24/96 for those "special instances."   Hopefully this has helped a few people out as well.

[mosquito]

FWIW:

- I can hear a difference between 24/48 and 24/96, and sometimes between 24/96 and 24/192.

- To hear the benefits of a sample rate higher than 48 kHz, I have to be using a good outboard DA and good monitors.

- I used to record 24/96 and sometimes even 24/192 because "I wanted to have a great raw recording available if the band ever needed it."

- I've since decided on using 24/48 for most of my regular audience recording because:
--> I post-process nearly everything and want a higher data rate than my final output which is 16/44.1.
--> None of the bands I've recorded have ever asked for the raw recordings to do things with.
--> In the grander scheme 24/48 is better than the rest of my signal chain:  I normally have to set up in a non-optimal position, my mics are cheap, my recorder has cheap pre's and AD converters. 
--> If a band really wants a very high quality recording we'll set it up in a different way and make the whole thing a heck of a lot better.
--> Less data is cheaper:  In the past two weeks I've collected about 10 GB of data.  Through processing that will normally become about 25 GB.  Running at 96 kHz means I have to buy another HDD twice as soon.

But the most important for me is:

--> Less data is faster:  my single biggest issue with taping is time.  If I can cut out 15% of the time it takes to post-process I get more done, make more people happy, and get more sleep.

Like others have already said, choose what's best for you. Other than that, there is no best, only best for a particular situation.

[aaronji]

Second, anecdotally I have noted that peoples' ears seem to incorporate their own "brickwall" filters.  That is, hearing does not drop off at any low-order filter (6dB, 12dB per octave); my best ear hears at 17kHz, then it's gone, it doesn't matter how loud you (cleanly) play 18kHz, I can't hear it.  With that in mind, I would think that incremental differences in sample rate *could* be as dramatic as order-of-magnitude changes.  This is why I tend to think that the difference between 44.1 and 48 is larger than 48 to 88.1, because the gremlins that lurk below 20kHz are mostly already gone by 48.

This is in accordance with a wide body of well-established audiology literature; especially in the high-end, the drop off is very steep.  Almost vertical.

I think this white-paper from Lavry sheds a lot of light on this subject: http://www.lavryengineering.com/documents/Sampling_Theory.pdf

The notion that more is better may appeal to one's common sense. Presented with analogies such as more pixels for better video, or faster clock to speed computers, one may be misled to believe that faster sampling will yield better resolution and detail. The analogies are wrong. The great value offered by Nyquist's theorem is the realization that we have ALL the information with 100% of the detail, and no distortions, without the burden of "extra fast" sampling.

Nyquist pointed out that the sampling rate needs only to exceed twice the signal bandwidth. What is the audio bandwidth? Research shows that musical instruments may produce energy above 20 KHz, but there is little sound energy at above 40KHz. Most microphones do not pick up sound at much over 20KHz. Human hearing rarely exceeds 20KHz, and certainly does not reach 40KHz. The above suggests that 88.2 or 96KHz would be overkill. In fact all the objections regarding audio sampling at 44.1KHz, (including the arguments relating to pre ringing of an FIR filter) are long gone by increasing sampling to about 60KHz.

He goes into a lot of detail that speaks to what Jon described as well as the physical limitations of some of the circuitry...He is particularly negative about 192 kHz.

[SmokinJoe]

- I can hear a difference between 24/48 and 24/96, and sometimes between 24/96 and 24/192.
articular situation.

You're a mosquito dude,  the rest of use are humans. 

[newplanet7]

- I can hear a difference between 24/48 and 24/96, and sometimes between 24/96 and 24/192.
articular situation.

You're a mosquito dude,  the rest of use are humans. 

Me and Joe talked about these things the other night and what we both gathered was what Lenny stated:

Everyones situation is different, and I think you should record in the format that works for you and your ideals of what you're getting out of this.  It's clear there's no right or wrong today and what's best is a matter of perspective, so just do what works for you and your situation.

[mosquito]

- I can hear a difference between 24/48 and 24/96, and sometimes between 24/96 and 24/192.
articular situation.

You're a mosquito dude,  the rest of use are humans. 

Hahaha!  Touché!

[justink]

i've already decided to stop releasing anything in 16/44.1.  to me, the cd is basically dead, and people are going to mp3 your uploaded files anyway...  so i'm just releasing/uploading in the format that i recorded in.

usually, i'll go 24/96, but if there's any reason that 96 is overkill, i'd love to hear more about that.

[acidjack]

i've already decided to stop releasing anything in 16/44.1.  to me, the cd is basically dead, and people are going to mp3 your uploaded files anyway...  so i'm just releasing/uploading in the format that i recorded in.

usually, i'll go 24/96, but if there's any reason that 96 is overkill, i'd love to hear more about that.

You make an excellent point.  I'm increasingly unclear who is burning things to CD, though I guess people do it.  I'd certainly prefer to just release my non-downsampled 24/48 FLACs and the VBR0 MP3s I already post on nyctaper.  Only people on DIME would complain, methinks.

Oh, but actually, the LMA only streams at 16/44.1, right?  That'd be another valid reason...

[justink]

i've already decided to stop releasing anything in 16/44.1.  to me, the cd is basically dead, and people are going to mp3 your uploaded files anyway...  so i'm just releasing/uploading in the format that i recorded in.

usually, i'll go 24/96, but if there's any reason that 96 is overkill, i'd love to hear more about that.

You make an excellent point.  I'm increasingly unclear who is burning things to CD, though I guess people do it.  I'd certainly prefer to just release my non-downsampled 24/48 FLACs and the VBR0 MP3s I already post on nyctaper.  Only people on DIME would complain, methinks.

Oh, but actually, the LMA only streams at 16/44.1, right?  That'd be another valid reason...

dime does complain, but they're retarded in their logic.  i'm not using dime for that reason anymore...  if i want to upload mp3's of a show i taped, i'll do it... elsewhere.

LMA will stream 24 bit files, they just derive from whatever you upload.  i'm not sure what exactly they do on their side, but i know it works.

[adrianf74]

@JustinK: I think Jon and Joe were trying to point out that 96/24 can be overkill and detrimental depending on what the final output is going to be.   I only convert my 48/24 recordings to 44/16 for VBR0 release to my friends (and for my phone when I'm at the gym or wherever).  To convert from 96/24 to 44/16 can introduce a lot of artifacts during the SRC so it's not overly advisable.   The other part of the equation is what do you plan to gain above the 20kHz threshold with a 96kHz recording.   

@AcidJack: What's a CD?    Seriously though, I haven't burned a show to CD since the mid-2000's as I've gone FLAC file-based to a) save room and b) prepare for the future in knowing that my tunes are easily available on my media computer at home.

@Both: DIME... Meh.

[newplanet7]

Sill burning discs and love it whorez!!!! 
I know quite a few people who put 16/44 flacs on their portable players after downloading as well as burn cd's.
Maybe it's a NH thang?
Maybe I am biased because I have been taping at 16/44 for four years but,
 I don't really see much improvement going to 24/48. I was going to just do it and get a dr100mkII
 but I am fully satisfied with my box. Probably don't have the playback and ears to notice anyway.
I get the not running levels so hot but once again, I know people who run and shoot for -6db hitting close to zero.
I do that with my 660 and have zero complaints. Raising the volume(if even necessary) -2db is nothing in the way of noise
that I hear.

[Belexes]

If I was in a controlled taping environment (studio), I would likely run 24/96. For us who are recording PA stacks, beer bottles being thrown away, chatty people near your gear, and sometimes a poor band...24/96? C'mon Man!