When does Bit and Khz no longer matter?

[rokuez]

Will and if so when does the Bit and Khz no longer have an effect on audio clarity/fidelity for the human range of hearing?

[Cojo]

I probably going to regret this answer in the future (it's like when Bill Gates once said that in the future a PC doesn't have to be bigger than a kubic meter?!) but here it goes!

I think 96KHz and 24 Bits is more than enough! There's more to do with the conversation then the resolution in my oppinion.

/Cojo

[noiseflaw]

Originally Posted by Cojo

I probably going to regret this answer in the future (it's like when Bill Gates once said that in the future a PC doesn't have to be bigger than a kubic meter?!) but here it goes!

I think 96KHz and 24 Bits is more than enough! There's more to do with the conversation then the resolution in my oppinion.

/Cojo

I don't think too many would disagree with you there.

24bit, 96khz + good mics and pre's etc + quality conversion - great song, good players - good mix, good master.........surely what more do you need.

[AmbientSound]

When you buy a 2inch tape machine

[thenewyear]

I like this thread.

All the answers are very sensible.

Remember that many great albums were multi-tracked at 44.1kHz & 16 or 20 bit & realise that we are lucky to have such high resolution these days & that the numbers have essentially become irrelevant.

Lets all concentrate on the music now...

[thenewyear]

When you buy a 2inch tape machine

If digital had never happened & everyone was still tracking to tape Digidesign would bring out a 3" tape machine for 'superior audio bandwidth'.













But they'd still limit the track count...

[tomdarude]

when they stop to crush the masters to death....




....my guess: if everything in the entire involved digital chain would be 24/96....including the medium, people are listening to the music on ( DVD-A, SACD etc.)....now THAT would be a giant leap



good years to come......hopefully

tom

[Capstan Cappy]

Originally Posted by thenewyear

If digital had never happened & everyone was still tracking to tape Digidesign would bring out a 3" tape machine for 'superior audio bandwidth'.

telefunken was supposed to have a little experiment like that in the old days if i remember correctly 32 tracks 3" , it failed horribly

think numbers nolonger count when i can hear 3D in the digital digital , so bring on the 48bit/3db step, 384khz standard plz

[thenoiseflower]

96/24 is good enough, BUT 192/24 does sound better, period.

so if sample rates goes up again, it may be even better.

so my personal opinion, we are close but there definetly may be some dsd rates in multitracks(more than 2 ) in the next ten years.

[Bob Olhsson]

It never did mean anything but potential quality.

Good 44.1x16 converters sound better than mediocre 96x24 converters.

[Blind]

i use 44.1x24bit and i am happy..

[heyman]

A turd of a song is still a turd of a song regardless of what format /bit rate it is in..

dfegad

[David Spearritt]

Originally Posted by thenoiseflower

96/24 is good enough, BUT 192/24 does sound better, period.

Does it? Please present your evidence and how it was gathered, period!

[ghellquist]

Aaah, but I think it depends on where in the chain you are.

As for recording very few, if any AD converters, really give much more than 20 bits of resolution. The last few bits are mostly noise from the analog stages anyway. 20 bits is about 120dB of S/N which is plenty. So to be on the safe side add a few up to 24.

But when you start treating the sound, mixing and matching and whatever, it is quite possible that you want more than 24 bits. Inside some algorithms you clearly may hear a difference beyond the 24 bits.

Sort of same goes for sample rate. A theoretically perfect AD will be good enough at around 44.1 kHz. Nothing is theoretically perfect though, so lets add a bit. Then we end up at around the 60kHz that Dan Lavry thinks are perfect. So lets say 88.2 as a compromise. One thing to know is that as you increase the sample rate with an AD you compromise away with higher noise levels and less accuracy. As with any engineering discipline it is always gain some, lose some.

But inside the different processing algorithms, eq or compressing or whetever, 88.2 may be cutting it a bit short. Many algorithms will create spurious signals that has to be filtered off. The filters may sound much better run at a higher sampling rate. So some plugins elect to run at much higher frequencys internally. Nothing odd about that really.

Personally I run at 44.1/24. I find that when I use good converters and good programs those are much better than I am. But I have selected my main converters for this specifically.

Gunnar

[Duardo]

Will and if so when does the Bit and Khz no longer have an effect on audio clarity/fidelity for the human range of hearing?

As far as the human range of hearing is concerned, we're past that. There are reasons higher sampling rates may sound better (or worse) but the don't really have anything to do with the range of human hearing. As far as bit depth is concerned, we can't perceive the equivalent of a 24-bit dynamic range, nor do we have analog circuitry that can handle it.

96/24 is good enough, BUT 192/24 does sound better, period.

Aside from the fact that "better" is subjective...

Have you heard every converter out there? I don't think so, because there are some that sound worse at higher sampling rates. Some sound the same.

There are enough other factors that come into play that 44.1 kHz doesn't sound as good as it theoretically can yet...maybe it never will.

-Duardo

[thenoiseflower]

Originally Posted by David Spearritt

Does it? Please present your evidence and how it was gathered, period!

ok, sure, period may be a bit much, granted, the quality of the conversion, as well as the chain feeding it.
I feel like i have heard more noise on tracks at either 96 or 192 in a less than well isolated tracking room, Does this lend to what gunner stated above?
anyway,
i should have said "my ears have heard *blah*, on these machines *blah*; These,

A.)Motu HD192 @192/24 int. clock to DP & logic, sounds better than 96/24.

B.)DIGI 192HD @192 into PT6.9 @ DP, *IMO* sounds better than 96/24

and my private studio, I share this fac w/ 3 others, and we all seem to agree also, we like 192,
C.)DIGI 192HD Dgital Clocked to the BigBen with 32 i ad/o da of apogee 16x hd cards yada yada @ 192/24, into PT 6.9 @ DP , sounds better (again to my ears), than 96/24.

hell, my composition rig at home is DP/C-sound based w/ a motu 24 i/o interface and those shite 'verters sound better (*IMO*) at 96/24 than 44.1/24, how much more budget/cut corner does ad/da get?,

thats what I've experienced, and those 3 192k capable rigs are the only 192 setps Ive worked on.
am i crazy?
is 192 useless? those converter $cost$ two of us our live in girlfreind$. , seriously, not kidding.

I feel like I have heard more noise on tracks at either 96 or 192 in a less than well isolated tracking room, and in a nicer, quieter room I feel like I hear it to be better/quieter, obviously, but, does this lend to what Gunner stated above?and will more low level background noise "cloud" a signal with a higher theroretical frequency limit? Theoreticly, higher sample rates means better/wider theoretical bandwith, does that mean external noises... ;(maybe noises at higher frequencys within the extended portion of our theorecical bandwith at higher sample rates , maybe HF electronic noise from the analog stages? like gunner stated? maybee outside noises in a "suspect" or poorly isolated tracking/iso room, what kind of frequencys up that high would get ino a recording environment like that?);.... are more likely to actualy be captured in the conversion and then represented within these the low level bits of a higher sample rate recording, and not reproduced on a lower sample rate/24bit recording beacause the frequencies above said lower sample rates theoretical limit were NOT captured like they were in the higher sample rate recordings that could theoreticly reproduce higher frequencies? This "extended bandwith" in higher sample rates, does it only thereticaly affect higher frequencies?

I am now very interested in this topic..

[Billster]

Originally Posted by tomdarude

....my guess: if everything in the entire involved digital chain would be 24/96....including the medium, people are listening to the music on ( DVD-A, SACD etc.)....now THAT would be a giant leap

Why do you think so ? With today´s dynamics in pop and rock music we only need 1 bit : on and off !

Seriously : I really would enjoy if music was mixed and mastered using the dynamic range even of 16 bits ! But I guess there´s not much need for that in recent pop/rock/hiphop/dance music history.

Bill

[moogus]

When the record hits number 1!



M

[Cojo]

Originally Posted by moogus

When the record hits number 1!



M

He said wisely, slowly leaning back in a comfy armchair, puffing his pipe...

[Capstan Cappy]

Originally Posted by Duardo

As far as the human range of hearing is concerned, we're past that. There are reasons higher sampling rates may sound better (or worse) but the don't really have anything to do with the range of human hearing. As far as bit depth is concerned, we can't perceive the equivalent of a 24-bit dynamic

you lost me there, what does the sample frequency have to do with the human hearing range , yeah sure more information we can perceive but where do the dislikes come from then?, i'm sure we "hear"what we don't like, and then i men quality not colouring etc, so as far as í'm concerned we're not "past" that...or do i misunderstand what ur saying

[Farview]

Originally Posted by Capstan Cappy

you lost me there, what does the sample frequency have to do with the human hearing range , yeah sure more information we can perceive but where do the dislikes come from then?, i'm sure we "hear"what we don't like, and then i men quality not colouring etc, so as far as í'm concerned we're not "past" that...or do i misunderstand what ur saying

I assume he is refering to the nyquist frequency of the different sample rates.

[bpatural]

bits and khz haven't mattered for years...people have been making great records recording to digital for at least a decade now and the public never even noticed the change.
-brian

[octatonic]

Originally Posted by Capstan Cappy

you lost me there, what does the sample frequency have to do with the human hearing range , yeah sure more information we can perceive but where do the dislikes come from then?, i'm sure we "hear"what we don't like, and then i men quality not colouring etc, so as far as í'm concerned we're not "past" that...or do i misunderstand what ur saying

Do a search on the Nyquist Theory.

It isn't just the bit depth and sample rate that matters.
I have 16 bit/44.1 effects units that sound far superior to some modern 24 bit devices.

Speaking generally when an Analog to Digital conversion occurs there can be an overlap of some frequencies.
This is called 'alaising'.
This aliasing can be prevented through the increase of the sample rate or by using an anti-alaising filter - the former being far more prefereable.

Nyquist theorem applies to the reconstruction of sampled material (ie your recordings) and requires it to be band limited and for the sampling frequency to be at least twice the bandwidth in order there to be minimal differences between the source material and the playback (Note to physics geeks- I know I am grossly simplifying things here.)

Then you need to consider the clock/jitter.

In short- there is a lot more to it than just sample rate and bit depth.
If one is using high quality A/D D/A stages with a good clock and low jitter then going above 24/44.1 should not make that much difference.

JR

[Dzoing]

As far as PCM is concerned, this Bit and kHz story will never end. For those who know about maths and digital electronics, there is not only the Nyquist Theory, but also artefacts like Gibbs phenomenon that are *growing* as sample frequency increases [I mean: this is a mathematical limit in the PCM approach, no matter how good the technology]

Bit and kHz will no longer matter when a new Theory+technology will appear -- think analog with *ideal* specs, think improved Delta-Sigma 1bit (DSD), think I don't know the new one that is still to be found/created/developped.

Today 24bit 44.1 is ok for me. Analog is nice -- but has artefacts too: his color, warmth, whatever...

[huarez]

Originally Posted by bpatural

bits and khz haven't mattered for years...people have been making great records recording to digital for at least a decade now and the public never even noticed the change.
-brian

Maybe they didn´t notice , but in fact they don´t buy music anymore. Real music freaks tend to collect old recordings . Obviously, the new digital recordings have a lack of Magic. Even in sequenzer based music nobody is using samples from new recordings, it´s all about the 70´s and 60´s stuff.