Topic: 24bit ..... or 16bit...whats your experience?

[Nick's Picks]

i've told my stories w/these two formats over and over here.
my bottom line, 24/44.1 is absolutely perfect.  All the benefits of 24bit, plus the ease of dither for redbook.

high sample rates?
i've done 96k off an on since 2001, and i have very few recordings where I feel that it made a positive impact.
mostly, i am distracted by crowd noise FAR more in my high bit rate/sample rate recordings.

[scb]

i think the crowd noise thing is just a coincidence.  i haven't found my 24/96 recordings to have more crowd noise

[Brian Skalinder]

i haven't found my 24/96 recordings to have more crowd noise

I don't think it's a question of more crowd noise, but the higher resolution of 24-bit providing greater detail.  The greater detail in turn makes the crowd noise more lifelike, more real - easier to pick out individual voices, conversations, noises, etc. - and therefore more distracting.  But that's based only a very few 24-bit recordings I've heard.

[Nick's Picks]

you nailed it Brian.
Scott..you just record in nicer venues than I do.

[kindms]

i haven't found my 24/96 recordings to have more crowd noise

I don't think it's a question of more crowd noise, but the higher resolution of 24-bit providing greater detail.  The greater detail in turn makes the crowd noise more lifelike, more real - easier to pick out individual voices, conversations, noises, etc. - and therefore more distracting.  But that's based only a very few 24-bit recordings I've heard.

I would say that this is definitely true but there is no way that you can say the above and not also say that the recording has greater detail in general.

I think it comes down to any taping situation you just never know what your gonna get in the sense that the crowd might be a great 1 and would you be wishing you ran 24Bit ? Sure a killer recording in any bit depth gonna sound good but when the recording is great and 24 Bit watch out.

"These go to eleven"

[rustoleum]

i haven't found my 24/96 recordings to have more crowd noise

I don't think it's a question of more crowd noise, but the higher resolution of 24-bit providing greater detail.  The greater detail in turn makes the crowd noise more lifelike, more real - easier to pick out individual voices, conversations, noises, etc. - and therefore more distracting.  But that's based only a very few 24-bit recordings I've heard.

I agree with this... it also makes other annoyances like wind noise more defined and potenitally more distracting.. that said, I'll run pretty much exclusively at 24 bit/48Khz these days as the music has that greater detail, too.  When listening on my big playback system the sound is so much more lifelike and natural to my ears and I can always dither and resample for the car or office. 

[creekfreak]

I think when taping a PA the difference is really not that big...now if you are taping raw music..say like an orchestra in a nice hall, or a band with no PA, then the difference become more apparent. Just my 2 cents...if anything, when taping a crappy PA in your typical bar or music venue, the 24bit will just make it sound all the more crappy...just my 2 cents

[twatts (pants are so over-rated...)]

Some interesting reads I found this morning:

http://stereophile.com/features/404metrics/index.html
http://www.stereophile.com/features/282/index.html
http://studio-central.com/phpbb/viewtopic.php?t=10674&highlight=24bit+16bit

Very interesting questions raised, at least in my mind - if our mics only record up to 20kHz, what's the point of being abel to playback or record higher than 20???  How about your playback freq. range??? 

Terry

Edit:  added another link

[JasonSobel]

Very interesting questions raised, at least in my mind - if our mics only record up to 20kHz, what's the point of being abel to playback or record higher than 20???  How about your playback freq. range??? 

yes, it is very interesting.  a few different things going on here, 24 bit vs 16 bit, and then the whoel sampling frequency debate.

IMO, 24 bit is definitely worth it.  the added low level detail (in both the music and the ambiance) makes the recording much mroe realistic.  It just makes it that much easier to close my eyes and imagine that I'm back at the show.  Come on Nick, first you say that 24 bit isn't worth it, and 16/44.1 upsampled at playback to 24/96 sounds almost as good as real 24 bit.   then you change your mind think 24 bit really is the way to go.  and then you get the Oade mod PMD-660 and think 16 bit is all that.    I'm a strong believer in the advantages of 24 bit, regardless if you think it makes the background conversations stand out more.

now, what Terry was talking about with Frequency response.  there is more reason than just frequency repsonse to increase the sampling rate.  I've read (sorry, I'm at work and don't have time to look up articles to link to at this time), but I've read that the human ear can detect time differences that are smaller than 1/48000th of a second.  From that logic, a 96kHz recording will be able to more accurately represent when a note actually begins, while that exact point in time is missed somewhere between 2 samples from a 48kHz recording.  (I'm sure that I'm not describing it as clearly as I would like, perhaps someone else can chime in and help out as well).

So, does that sampling difference make an audible difference?  well, I've only done one non-scientific comparison of a 24/48 recording vs. 24/96 recording.  based on that comparison, I'd say the difference is subtle, but I was able to pick up a slightly more defined and clarity when listening on my grace 901 > grado rs-2 headphones.  but very subtle.  subtle enough that I'm more than happy to record at 24/48 until 8gig CF cards are cheap enough.

but do I think 24 bit is worth it?  absolutely.  huge improvement over straight 16 bit recordings.

[scb]

Scott..you just record in nicer venues than I do.

oh i defintiely tape in some shitholes

[kindms]

if portable DSD becomes a reality im all for it. At this point ill be happy if i can leave the lappy at home

[creekfreak]

if portable DSD becomes a reality im all for it. At this point ill be happy if i can leave the lappy at home

QFT

[Nick's Picks]

I know Jason...i'm fickle.

I think 24bit is worth it.  but 16bit doesnt suck at all.
high sample rates...not worth it, IMO. 

its funny...
here is a band...whos cabinets are being "recorded" with sure SM57s > compressed to all hell > pumped through mostly crappy PA systems in MONO, more or less...and here we are recording the room with $2k+ worth of stereo gear.
and ...thinking its "all that"

[nickgregory]

The one thing I think some folks miss in this type of discussion is the effect of 24 bits being captured of not just the PA, but the entire environment.  True it equals more detail on conversations as stated earlier, but the other thing it does is more accurately capture the environment you were listening in...I think some of this is attributable to the "space effect" that I know I hear on 24 bit tapes.  In addition to the resolution of the recording I am hearing the environment it is in...which can be both a bad thing and an amazing thing...based on the surroundings.

[Ed.]

I know Jason...i'm fickle.

I think 24bit is worth it. but 16bit doesnt suck at all.
high sample rates...not worth it, IMO.

its funny...
here is a band...whos cabinets are being "recorded" with sure SM57s > compressed to all hell > pumped through mostly crappy PA systems in MONO, more or less...and here we are recording the room with $2k+ worth of stereo gear.
and ...thinking its "all that"

i've thought about that too, with some local bands i record - my rig costs more than the instruments of the entire band.

but it is still about the environment too.  i'm thinkign i'll probably do 24/96 only if i'm shooting video or if the venue is top notch.  i'll still probably run 24bit once i can tho, especially if i have the space on the cf card.

[Kindguy]

24bit makes me feel like a newbie again. 

One day I'll give it a whirl.

[nickgregory]

24bit makes me feel like a newbie again.

exactly how I felt with inital 722 issues, burning dvd-A issues and home playback upgrade...glad I made the shift though!

[BC]

i haven't found my 24/96 recordings to have more crowd noise

I don't think it's a question of more crowd noise, but the higher resolution of 24-bit providing greater detail.  The greater detail in turn makes the crowd noise more lifelike, more real - easier to pick out individual voices, conversations, noises, etc. - and therefore more distracting.  But that's based only a very few 24-bit recordings I've heard.

This is exactly one of the first things I noticed when I started recording 24 bit...

[cdevs]

its funny...
here is a band...whos cabinets are being "recorded" with sure SM57s > compressed to all hell > pumped through mostly crappy PA systems in MONO, more or less...and here we are recording the room with $2k+ worth of stereo gear.
and ...thinking its "all that"

sounds like somebody's got a case of the ...Thursdays (No offense intended, dude)

What you say is true...but two wrongs don't make a right, either. (although 3=left)

Universal law: every link in the chain affects the end result. Having said that, some of the best listening experiences I've had involved sub-$1K playback systems (albeit w/Klipschs) and average recordings of REALLY GREAT shows...cranked...

So what's my point? Guess it's all relative...and trust your ears. (And yer NUTS )

PS Nick check this out for km140 action:
http://www.dimeadozen.org/torrents-details.php?id=58869

[marc0789]

I think when taping a PA the difference is really not that big...now if you are taping raw music..say like an orchestra in a nice hall, or a band with no PA, then the difference become more apparent. Just my 2 cents...if anything, when taping a crappy PA in your typical bar or music venue, the 24bit will just make it sound all the more crappy...just my 2 cents

after listening to quite a few 24 bit sources, that's my take.....but I haven't run 24 myself....just think that for taping PA music from a distance, with venue issues, crowd issues that the payoff just isn't there. maybe I'd feel differently if I had a topnotch acoustically perfect venue to tape in, but not even close.

Or maybe that's my excuse, and I'm just too lazy and cheap to make the investment.

[marc0789]

I know Jason...i'm fickle.

I think 24bit is worth it.  but 16bit doesnt suck at all.
high sample rates...not worth it, IMO. 

its funny...
here is a band...whos cabinets are being "recorded" with sure SM57s > compressed to all hell > pumped through mostly crappy PA systems in MONO, more or less...and here we are recording the room with $2k+ worth of stereo gear.
and ...thinking its "all that"

that about says it all for me.

[tonyvt]

here is a band...whos cabinets are being "recorded" with sure SM57s > compressed to all hell > pumped through mostly crappy PA systems in MONO, more or less...and here we are recording the room with $2k+ worth of stereo gear.
and ...thinking its "all that"

Excellent observation.
This thread reminds me of a conversation I had years ago in the section before a dead show. We were bragging about how our mics sounded better than our buddy's.

 My buddy said the funniest thing, "No matter how much you spend on your rig is still sounds like the show was taped in a boomy hockey rink".

[Chanher]

two things

1. for all the ua-5 users, I have found 24-bit to be worth it. I ran the ua-5 into a lappy at 24 bit since Jan. I recently busted out the jb3 for a couple shows. I had prime fob sweet spot location and was excited to hear the results; after listening on the stereo I was pretty dissapointed. The recording just didn't shine like past ones, it seemed "dull" as opposed to being "lively". Is this what people are calling "musical"?  Anyway, I REALLY want to run a head to head ua-5 comparison, one at 16 and one at 24. This brings up another question: Does the ua-5 have any sort of noise-shaping algorithm for 16-bit?

2. we all know the rate of improvement in technology is very fast. I like to think that in the future, maybe even past our lifetimes, digital audio technology could be beyond our very comprehension. By using the highest bit-depth and sample rate possible, we could be doing future audio technicians a favor for whatever advances techniques and solutions they might have. Think about all the remastering being done to timeless recordings from the past/analog era; they've worked wonders.

[Lil Kim Jong-Il]

if our mics only record up to 20kHz, what's the point of being abel to playback or record higher than 20???  How about your playback freq. range??? 

Terry

Nyquist is appropriate to determine the maximum required sampling rate only for a single point source.  For stereo recordings of multiple point sources, there is more to capture than just the wave amplitude.  Your ears and brain can detect sound arrival differences about as small as the time between two samples at 96KHz.  So if you record a widely dispersed collection of distinct sources, an orchestra for example, you will get better spatial resolution of the different instruments.  I don't have the supporting links handy but I posted references in a playback thread some time ago. 

I agree that it is overkill for the PA tapes we make mainly because I could not hear a difference in the comparison tracks Wayne sent to me.  But there is an advantage to 96k in some real situations. 

[admkrk]

if our mics only record up to 20kHz, what's the point of being abel to playback or record higher than 20???  How about your playback freq. range??? 

Terry

Nyquist is appropriate to determine the maximum required sampling rate only for a single point source.  For stereo recordings of multiple point sources, there is more to capture than just the wave amplitude.  Your ears and brain can detect sound arrival differences about as small as the time between two samples at 96KHz.  So if you record a widely dispersed collection of distinct sources, an orchestra for example, you will get better spatial resolution of the different instruments.  I don't have the supporting links handy but I posted references in a playback thread some time ago. 

I agree that it is overkill for the PA tapes we make mainly because I could not hear a difference in the comparison tracks Wayne sent to me.  But there is an advantage to 96k in some real situations. 

i'm surprised it took this long before the nyquist theory (frequency) was brought up. also i certainly can't quote it right now, but it's something like needing twice to get half??  i.e. you need to record at 40 khz to get 20 khz. it also is irrelivent to 24bit vs. 16bit.

frequency = the "range" of sound we hear.
bit rate / word length = the amount of detail.

 consider 16bit = mp3  and  24bit = flac


maybe not the best analigy, but i think it fairly represents the point.

[twatts (pants are so over-rated...)]

Oh, yeah.  I'm a total moron when it comes to the science...

I can understand the basics - more is better - 24/96 gives you much better detail, etc.  But I'm wondering how many of us are thinking of the cart before the horse by jumping to 24/96 before playback can handle it.  I know if I went 24/96 now, I wouldn't be able to reap any of the benefits. 

But at the same time, I'm not saying that we should tape to 24/96 if we have the ability.  I've started taping 16/48 a long time ago knowing one day I could use the 48k FLACs.  Thinking about the future...

T

[Lil Kim Jong-Il]

Terry implicitly brought up nyquist with his question of sample rate vs the fequency range of the mics.  I mentioned it because I understood that was the basis for his question and most people assume that nyquist is the last word on sample rate selection, which it isn't. 

With respect to the original topic, I hear a dramatic improvement with 24-bit in my system and I don't find the additional resolution distracting when it come to the crowd noise.  So for me the 24-bit is the way to go.

Edit:  BTW, sampling rate and bit depth are both related to the detail in a recording.  Sample rate determines the maximum spatial detail within the image for recordings of multiple point sources.

[Lil Kim Jong-Il]

I've started taping 16/48 a long time ago knowing one day I could use the 48k FLACs.  Thinking about the future...

what is the advantage of running 16/48, vs. 16/44.  I have always wondered.........

I don't think there is much advantage to 48K over 44.1K.   If you will always burn to CDs for playback, it is probably better to use 44.1K so that you don't have to down-sample from 48K for the CDs.

On the other hand, if you plan to burn your entire show to one DVD and you intend to use the DVD-V format instead of DVD-A, then you might prefer 48K so that you don't have to up-sample from 44.1K to 48K since the DVD-V format supports only 48K and 96K LPCM streams. 

[Brian]

24/44.1 or 48k = grainy

24/88.2 or 96 = not grainy

it has to do with the sample rate's relationship to time.  this is simply why higher sample rates will always be better.  more on this when i can pull together my thoughts to better explain this.

[BC]

sample rate -> Frequency response (Nyquist theorem as mentioned by others)

Bit depth -> Dynamic range, 6 dB/bit, so 24 bit word lengths give us a theoretical limit of 144 dB, although most preamps and AD's top out at 115 to 120 dB dynamic range. One benefit of greater dynamic range is increased resolution of low level sounds. Another is the ability to run more conservatively and still wind up with resolution greater than a 16 bit recording with perfectly hot levels.

[Brian]

ok here is a simple explanation of what i meant int he previous post.

Studies have shown that human beings can perceive time, or audio bursts in the range of 1/50000 of a second.thus you can HEAR the holes, or pauses in the 44.1 or 48k signal, but not in a 96 K signal or even a 60K sample rate.    People also used to say there was no reason to record in 24 bits too, the claim was that the human ear couldn't perceive the difference in bits.

Those holes we hear are perceived as "grain" in the recording.

[Lil Kim Jong-Il]

Brian, what you posted about temporal perception is what I was trying to write above.  Somewhere there is a link to a detailed explanation and I think there is a link to a study that provides emperical results.  I just dumped a bunch of bookmarks and lost the links.

I completely agree with this on a theoretical level, which is why I record everything at 96K.   But I have to keep in mind the results of the 48K vs 96K test Wayne did with the Deva.  I could hear a difference between the two recordings and the audible difference could easily be blamed on the different nights and slightly mismatched levels of the two samples.   I could not hear a difference that made me say "Ah, that one is 96k". 

Either my gear isn't up to the task or maybe my player upsamples with excellent results.   I can hear very distinct difference between the 24 vs 16 but the 96k isn't distinctly better than than the 48K in that test.  I have not done my own test.   At least one person here claims that he hears the difference with 96K in his system. 

Do you actually hear a difference in your playback system? 

[Brian]

at this point I can't really say with simple two track recordings.

now with some of the multi-track stuff i've worked on, I can definitely tell a difference when I'm listening back over really nice monitors.

over computer speakers? no.  over my motu 828mkII > KRK V6's setup at home...I can't quite tell with teh two track stuff yet.  I need to do some more critical listening though.

The scientific tests and results are enough to make me record at higher sample rates.  I mean, we spend lots of money buying all this great gear and take the time to record in the best spot we possibly can.  Why go that far to only to introduce grain or the "susceptibility of grain" into recording by utlizing a low sample rate? I thought we were all about trying to make the best possible recordings we can.

[dklein]

Adding a comment on the Nyquist thing - there's another reason for higher sample rates.  When doing the a>d, you must filter out all frequencies above Nyquist to avoid aliasing.  So when we record at 44.1, nothing above 22kHz is allowed to hit the a>d.  At the same time, we want frequencies up to 20 kHz.  This abrupt filtering (no effect up to 20kHz, full blocking at 22kHz) is supposed to be quite challenging from a design standpoint.

A major advantage of going to something like 96k is that Nyquist is now 48k.  Creating a filter that cuts everything above 48k while leaving everything up to 20k unaltered is supposed to be much easier to accomplish.

It's been written that much of what we don't like about the digital sound is a by-product of the required filtering vs. just being digital.

All theory...I haven't done any 96k recordings myself (or even played back someone elses).

I always record in 24 bit these days but haven't necessarily concluded the benefits are there.  I like the idea, have the gear and figure it couldn't hurt, especially with respect to nailing the levels.  One day I'll get a proper a/b test going but that's not so easy to do and would still only cover off one particalur a>d.