Become a Site Supporter and Never see Ads again!

Author Topic: Which to sacrifice? Bit depth or sample rate?  (Read 5016 times)

0 Members and 1 Guest are viewing this topic.

Offline down2earthlandscaper

  • Gear Collector (Slut/Hoarder) and Multiple-Rig Maniac
  • Site Supporter
  • Trade Count: (29)
  • Taperssection Member
  • *
  • Posts: 905
  • Gender: Male
Which to sacrifice? Bit depth or sample rate?
« on: March 28, 2015, 11:21:40 PM »
Ok, I could kick myself. I usually always bring fresh memory cards, but I got to the show last night and went to record 3 channels at 24/96, and realized that I did not have enough recording time left on the card :face palm:
Is there a "lesser of two evils"? I sacrificed one - I hope it was the correct choice. I put 96 down to 44.1 and left it at 24bit :-\
YMSB put on a great show btw
Mics: CA-14(cards & omnis) and CA-11(cards & omnis) ; AT853's(cards, hypers, mini shotguns); Busman BSC-1 (cards, hypers, omnis)
Nakamichi CM300's (CP-1,2,3,4) Nakamichi CM700's (cards, omnis)
Tascam PE-120's (cards, omnis) Countryman B2D
DPA 4061's DPA 4022's; DPA 4080; AKG 480 ck61 and ck63; Naiant AKG Active Cables
Preamps: CA-9100; Naiant Tinybox (12v/48v + PIP 8V); Naiant Littlebox;
DPA MPS6030; Sound Device Mix Pre-D
Decks: Mixpre 10T and 6; Roland R-07; Marantz PMD620; Sony PCM M10; Edirol R-4; Zoom H6; Marantz PMD-661; Sound Devices 722

Marshall7

  • Guest
  • Trade Count: (0)
Re: Which to sacrifice? Bit depth or sample rate?
« Reply #1 on: March 28, 2015, 11:53:20 PM »
Probably the best choice, although if you had enough space 24/48 would have been better.

Offline justink

  • Trade Count: (4)
  • Taperssection All-Star
  • ****
  • Posts: 1973
  • Gender: Male
Re: Which to sacrifice? Bit depth or sample rate?
« Reply #2 on: March 28, 2015, 11:53:46 PM »
yup.  i would have done the same thing.
Mics:
DPA 4023 (Cardioid)
DPA 4028 (Subcardioid)
DPA 4018V (Supercardioid)
Earthworks TC25 (Omni) 

Pres and A/D's:
Grace Design Lunatec V3 (Oade ACM)
Edirol UA-5 (bm2p+ Mod)

Recorders:
Sound Devices MixPre10 II
Edirol R-44 (Oade CM)
Sony PCM‑M10

Offline down2earthlandscaper

  • Gear Collector (Slut/Hoarder) and Multiple-Rig Maniac
  • Site Supporter
  • Trade Count: (29)
  • Taperssection Member
  • *
  • Posts: 905
  • Gender: Male
Re: Which to sacrifice? Bit depth or sample rate?
« Reply #3 on: March 29, 2015, 01:04:11 AM »
Phew! Thanks - I knew I read about this before but couldn't remember which one is considered more important. I think if I didn't link channels 1 and 2 as a stereo pair, I might have had enough room… Of course I didn't think of that until now.
I used Busman BSC-1 cards on channel 1 and 2 as the L and R (stereo) and I ran a Nakamichi 300 CP3 on channel 3 as a center omni. Just experimenting - turned out really nice.
Mics: CA-14(cards & omnis) and CA-11(cards & omnis) ; AT853's(cards, hypers, mini shotguns); Busman BSC-1 (cards, hypers, omnis)
Nakamichi CM300's (CP-1,2,3,4) Nakamichi CM700's (cards, omnis)
Tascam PE-120's (cards, omnis) Countryman B2D
DPA 4061's DPA 4022's; DPA 4080; AKG 480 ck61 and ck63; Naiant AKG Active Cables
Preamps: CA-9100; Naiant Tinybox (12v/48v + PIP 8V); Naiant Littlebox;
DPA MPS6030; Sound Device Mix Pre-D
Decks: Mixpre 10T and 6; Roland R-07; Marantz PMD620; Sony PCM M10; Edirol R-4; Zoom H6; Marantz PMD-661; Sound Devices 722

Offline Fatah Ruark (aka MIKE B)

  • Trade Count: (11)
  • Needs to get out more...
  • *****
  • Posts: 9940
  • Gender: Male
  • I dream in beige.
    • sloppy.art.ink
Re: Which to sacrifice? Bit depth or sample rate?
« Reply #4 on: March 29, 2015, 11:41:36 AM »
I'd also chose 24 bit...but personally I'd just start recording 24/48 all the time to save space.

I can't tell the difference between 48 and 96. I figure I save space on my hard drives as well as for what you experienced...longer record times per card.

You never know when some band is going to break out some epic set.
||| MICS:  Beyer CK930 | DPA 4022 | DPA 4080 | Nevaton MCE400 | Sennheiser Ambeo Headset |||
||| PREAMPS: DPA d:vice | Naiant Tinybox | Naiant IPA |||
||| DECKS: Sound Devices MixPre6 | iPod Touch 32GB |||
|||Concert History || LMA Recordings || Live YouTube |||

stevetoney

  • Guest
  • Trade Count: (0)
Re: Which to sacrifice? Bit depth or sample rate?
« Reply #5 on: March 29, 2015, 01:31:37 PM »
I'd also chose 24 bit...but personally I'd just start recording 24/48 all the time to save space.

I can't tell the difference between 48 and 96. I figure I save space on my hard drives as well as for what you experienced...longer record times per card.

You never know when some band is going to break out some epic set.

I agree about the 24bit vs. 16 bit.  That's a no brainer.  I personally wouldn't feel badly if I had to drop my bitrate down to 48 or 44.1 if I had to. 

However, with that said, up until recently I also felt there was no sound quality loss at 24/48 vs. higher bitrates, but I recently did some testing with higher end audio gear (Fiio X3 > Sennheiser HD700 > Pan Am headamp) that changed my mind.  My testing was 24/48 versus 24/196 and 24/48 vs. DSD files. 

First off, if you're not listening through high end playback, then I don't think you're going to hear much difference in the higher bitrate, but with higher end gear, the difference is there for the higher quality recordings I've made.   I doubt that in a venue where the sound isn't all that great, that it would make much difference, but on my better recordings, after listening to the higher bitrate files, going backwards is one of those experiences where, after you've listened for awhile at the highest quality then the sensation of the recording being open, airy, realistic just isn't there anymore.  It's that A vs B thing we've all experienced...you don't perceive the lower quality on the 24/48 files until after you've re-set your frame of reference at 24/196.  Then when you go back, you hear the difference.  Come back an hour later (or whatever) and the 24/48 files sound great again.

Obviously, whether its worth 4x the disc space for that extra bit of sound quality is a personal choice...might not be worth it.  But since I'm kind of a quality fiend, I've changed my thinking as a result of my recent experience.  I've been recording everything in DSD format with my D100 as my master and still archiving to LMA in 24/48 and 16/44.1.  If I ever sell the D100, I'm going to be recording in 24/196.  Just yesterday I bought a 2tb portable drive on sale for $88 as a third redundant drive for backups, so space is just not an issue to me.
« Last Edit: March 29, 2015, 01:54:18 PM by tonedeaf »

Offline yug du nord

  • ...til things never seen seem familiar…
  • Trade Count: (56)
  • Needs to get out more...
  • *****
  • Posts: 5533
  • made with natural flavor
Re: Which to sacrifice? Bit depth or sample rate?
« Reply #6 on: March 29, 2015, 03:43:55 PM »
^24/196 must be incredible!!..  i've only ever heard of 24/192....   ;)
just messin with ya t-deaf..  :)

i'm still curious on the dsd..  unknown territory for me.  the flexibility is what intrigues me.

...but yeah good call to the OP for keeping the 24bit!!
.....got a blank space where my mind should be.....

Offline DSatz

  • Site Supporter
  • Trade Count: (35)
  • Needs to get out more...
  • *
  • Posts: 3347
  • Gender: Male
Re: Which to sacrifice? Bit depth or sample rate?
« Reply #7 on: March 29, 2015, 04:35:41 PM »
Both settings are of the kind where, once they've been set high enough for your actual circumstances, nothing of a non-masturbatory nature is gained by setting them any higher.

Neither setting is the kind where the higher it is, the better things are in an absolute sense (i.e. where we should always wish that they could be higher than the highest setting we have, because that would be even better).

Very few people can hear a difference between well-implemented 44.1 kHz recording and recording at any higher sampling frequency, even on a direct comparison of revealing material played back on a good playback system. Many more people believe that there must be a substantial, audible difference than can actually identify such a difference by ear--i.e. people who could, at all reliably, tell you whether a recording was made at 44.1 kHz vs. 96 if you had equivalent samples of both. Hell, in my professional work with prominent producers and artists, I rarely met anyone who could tell whether they were listening to a direct analog feed vs. 16/44.1 with 1980s equipment, although I met plenty of people who were oh-so-sure that they could. (They didn't always know when I was testing them ...)

There's a rather high probability that in any given case other than a specially-set-up laboratory situation, 24/96 wouldn't produce any more audibly accurate results than well-implemented 16-bit at either 44.1 or 48 kHz would do. Depending on the specific equipment, sampling accuracy can even be reduced at the higher rate (speed and accuracy are not normally friends in the performance of any given physical task). And the additional bit depth is completely useless, literally just a waste of storage space, if the noise level at the input to the converter is at or above the 16-bit level, which it often is in live recording situations. (It helps being old enough to recall the panic in the recording industry when digital recording first came in, and the studios and record labels realized that they had to deal with all the low-level noises being revealed on CDs that had never been audible on vinyl.)

So you did the right thing as long as you were able to make your recording successfully. But other possible choices might have been good, too, depending on the circumstances.

--best regards
« Last Edit: March 29, 2015, 04:51:42 PM by DSatz »
music > microphones > a recorder of some sort

Offline 2manyrocks

  • Trade Count: (12)
  • Taperssection All-Star
  • ****
  • Posts: 1664
Re: Which to sacrifice? Bit depth or sample rate?
« Reply #8 on: March 29, 2015, 04:49:37 PM »
Is there a particular reason why audio for video is often 48K? 

Offline DSatz

  • Site Supporter
  • Trade Count: (35)
  • Needs to get out more...
  • *
  • Posts: 3347
  • Gender: Male
Re: Which to sacrifice? Bit depth or sample rate?
« Reply #9 on: March 29, 2015, 04:53:32 PM »
Yes: 48 kHz is evenly divisible by the color video frame rate.

--edited later to add: Sorry--the above is true, but it's not the whole story. It has to do with progress in A/D converter technology as well.

The CD standard came first, but at that time the input filters on A/D converters were all-analog and quite complex (typically 11-pole) and expensive to manufacture. They were also subject to drift as components aged, even though they were almost always used indoors in a relatively narrow range of temperatures. You had to send them in for an expensive recalibration every year or so, or else your recording system would have higher and higher levels of distortion over time (I suspect that failure to do this in a timely manner may have been the cause of some people's poor opinion of digital audio quality at the time; there were also adjustments on the A/D converters themselves that needed to be made periodically, which I was trained how to do for the Sony professional system, but a lot of other people weren't).

The complexity of the filters was necessary because the steepness was necessary. These were linear (not oversampling) converters, and the "stop band"--the distance between 20 kHz and 1/2 of 44.1 kHz--was a mere 2.05 kHz. A higher sampling rate could not be chosen because at that stage in the manufacture of converters, it would have produced worse performance rather than better (the tradeoff between speed and accuracy that I mentioned earlier); the converters would have needed almost constant realignment (one reason the Soundstream recording system, which ran at 50 kHz, was leased and not sold, and all sessions were supervised by company-trained technicians; the converters were calibrated and tweaked for every session).

48 kHz converters were feasible by the time the video standard emerged a few years later, and the higher sampling rate gives a stop band of 4 kHz, which is twice as wide as what you get at 44.1. This allowed for simpler but still adequate filters (e.g. 5-pole) which were less expensive to manufacture and less likely to drift with temperature and age.

But one thing that it's not is because "more data points per second means a more accurate reconstruction of the analog waveform" the way some people imagine. That's not the way digital audio works or has ever worked.
« Last Edit: March 29, 2015, 10:10:43 PM by DSatz »
music > microphones > a recorder of some sort

Offline DSatz

  • Site Supporter
  • Trade Count: (35)
  • Needs to get out more...
  • *
  • Posts: 3347
  • Gender: Male
Re: Which to sacrifice? Bit depth or sample rate?
« Reply #10 on: March 29, 2015, 04:59:09 PM »
> That said, the difference in the audible band (20kHz) is greater between 44.1 and 48kHz than 48kHz and any higher sample rate.  This is easily demonstrated with measurement, so if you can't hear a difference between 44.1kHz and 48kHz, then the effect of higher sample rates is either implementation of the equipment (suboptimal at one rate or the other) or placebo.

I completely agree (not that the vote count should matter for a statement of basic fact, but I'm just saying ...)
music > microphones > a recorder of some sort

Offline yug du nord

  • ...til things never seen seem familiar…
  • Trade Count: (56)
  • Needs to get out more...
  • *****
  • Posts: 5533
  • made with natural flavor
Re: Which to sacrifice? Bit depth or sample rate?
« Reply #11 on: March 29, 2015, 05:06:19 PM »
^I've read somewhere, at some point (real accurate eh, :P) that "audiophile" gurus actually claim that 48, 96, or 192kHz recordings sound worse than 44.1kHz recordings.  And that if a person wants to run higher than 44.1kHz... 88.2kHz or 176.4kHz is the way to go.  Some sort of algorithm theory if I remember correctly.  That's way out of the realm of my knowledge..  but the theory exists somewhere in the "audiophile" world.

And through various comments regarding 16/44.1, that our own gurus have made here on site..  which had led me to some research, I have read numerous accounts that claim a precise 16bit/44.1kHz recording can be as good as it gets.  And that is why the "standard" may never change.
Now Neil Young and his PONO player might argue that theory..  but once again, that theory exists.

I'm happy with maximizing to the best that I can, in the situation at hand 24bit/48kHz just to play it "safe".
But to each their own.

.......and no, I do not have any links to support my comments.   :P
« Last Edit: March 29, 2015, 05:23:38 PM by Snowman »
.....got a blank space where my mind should be.....

Offline down2earthlandscaper

  • Gear Collector (Slut/Hoarder) and Multiple-Rig Maniac
  • Site Supporter
  • Trade Count: (29)
  • Taperssection Member
  • *
  • Posts: 905
  • Gender: Male
Re: Which to sacrifice? Bit depth or sample rate?
« Reply #12 on: March 29, 2015, 08:53:11 PM »
Thanks everyone. Great conversation and great bunch of info.
I didn't know that the largest audible difference is between 44.1 and 48 vs. 48 and any other…
I'll remember that.
Mics: CA-14(cards & omnis) and CA-11(cards & omnis) ; AT853's(cards, hypers, mini shotguns); Busman BSC-1 (cards, hypers, omnis)
Nakamichi CM300's (CP-1,2,3,4) Nakamichi CM700's (cards, omnis)
Tascam PE-120's (cards, omnis) Countryman B2D
DPA 4061's DPA 4022's; DPA 4080; AKG 480 ck61 and ck63; Naiant AKG Active Cables
Preamps: CA-9100; Naiant Tinybox (12v/48v + PIP 8V); Naiant Littlebox;
DPA MPS6030; Sound Device Mix Pre-D
Decks: Mixpre 10T and 6; Roland R-07; Marantz PMD620; Sony PCM M10; Edirol R-4; Zoom H6; Marantz PMD-661; Sound Devices 722

Offline John Willett

  • Trade Count: (0)
  • Taperssection All-Star
  • ****
  • Posts: 1550
  • Gender: Male
  • Bio:
    • Sound-Link ProAudio
Re: Which to sacrifice? Bit depth or sample rate?
« Reply #13 on: March 31, 2015, 08:04:56 AM »
Ok, I could kick myself. I usually always bring fresh memory cards, but I got to the show last night and went to record 3 channels at 24/96, and realized that I did not have enough recording time left on the card :face palm:
Is there a "lesser of two evils"? I sacrificed one - I hope it was the correct choice. I put 96 down to 44.1 and left it at 24bit :-\
YMSB put on a great show btw

Yes, you did the right thing - and 44.1 was the better option if the recording was to end up on CD.

All the sampling frequency does is to extend the top-end frequency response so you definitely made the best choice.

 

RSS | Mobile
Page created in 0.145 seconds with 38 queries.
© 2002-2024 Taperssection.com
Powered by SMF