Is 96khz better than 48khz for multitrack digital recording?

[JOHN]

Happy Xmas to all
Is 96khz better than 48khz for multitrack digital recording?
i want to record real instrument sounds

Cheers

[Fletcher]

Depends... I have found 96k better for a lot of stuff that I wanted to be a measure more "audiophile" in it's result... and 44.1k better for things I wanted to sound more "rock and roll" [have an attitude]... which is kind of amusing because I have found something very similar with 30 and 15 ips in the analog domain.

FWIW I don't like 48k sampling that much because it's a noticeable sonic improvement [in the 'audiophile' dept.] to 44.1k and should you stay in the box it will require a 'sample rate conversion' to make a CD... which I have always seemed to find more destructive to the audio than the higher sampling rate is beneficial.

As always... YMMV

[ISedlacek]

It also depends on the particular converter.

But - what prevents you from a simple trying and hearing for yourself ? Record somenthing in 44k, then the same in 96k, convert 96k files to 44k using some high quality SRC (like Voxengo r8brain Pro) and then compare the results. Do you hear any difference ? Do you prefer one over the other ? You don´t need to ask anyone about it - just listen.

Using Lavry Blue, 96k recodings sound a way better for me (even after converting them to 44k) - more details, depth, air and 3D. I would not be able to record in 44k any more.

[AllAboutTone]

Not sure about 96 that much, i have tired 88.2 before and liked it, 48 sucks IMO.
I much rather choose bit rate over most sampling rates.

[ISedlacek]

Originally Posted by AllAboutTone

Not sure about 96 that much, i have tired 88.2 before and liked it, 48 sucks IMO.
I much rather choose bit rate over most sampling rates.

I tried and found that 96k sounds yet slightly better than 88k

[The Byre]

It varies with the quality of the converters.

I often hear that higher sample rates are required if we are to hear music properly. Despite the rather obvious fact that our ears only go up to 18kHz at the very best for an adult and usually somewhere between 14 and 16Khz, (depnding on age, health and the amount of exposure to loud noise) and the fact that no studio mics go beyond 22kHz, people claim to be able to hear an improvement when listening to material recorded at higher sample rates.

At the same time, several scientific tests have shown that when the same material is played back at different sample rates (and where the source material was at the very highest rate possible) none of those tested could hear any difference between 192kHz and 96, 48, or 44.1kHz.

Many hi-fi companies and enthusiasts assert that the wider bandwidth and dynamic range of SACD and DVD-A make them of audibly higher quality than the CD format. However many carefully controlled double-blind tests with many experienced listeners showed no ability to hear any differences between formats. Therefore the perceived quality has nothing to do with additional resolution or bandwidth.

And yet even I can hear a difference on some occasions! So what is going on?

In general, yes it does sound better, but not because we need the extra bandwidth or dynamic range to appreciate the music, but because of the inadequacies of earlier equipment.

Earlier converters created all kinds of distortion and even self-generated noises at the higher frequencies. One DAW even had a constant whistle at about 18kHz. This lead to all kinds of unpleasant artifacts and distortion in the upper frequency ranges. The worst offenders were DAT and ADAT machines, cheap computer sound cards and even some prestige converters.

This upper range distortion was what some people called digital sharpness, or digital harshness. In reality, it was just good, old fashioned distortion. When ever I mastered a piece of music that had been recorded, using an older box, the spectrum analyser would show the line 'sticking' at the top somewhere above 17kHz. This is when I became interested in the subject and dug to find out more.

Now, if you double the sample rate to 96kHz (or 88.2kHz for a CD) when making the original recording, then all that distortion taking place between say, 14kHz and 22kHz is moved to between 28 and 44kHz - i.e. out of hearing range.

Suddenly, all that upper distortion vanishes and we come to the conclusion that a 96kHz sample rate is superior to 44.1kHz.

Very often, it is not even the converter's fault, but can be the result of poor cabling, or interference from other equipment. Discussions with hardware engineers would suggest that clocking was a major problem for some systems.
So if you want to play it safe, record at the higher rate to ensure that no distortion caused by jitter or other gremlins come in with the original material, but the improvement is not because you have golden ears!

[Flying_Dutchman]

if the converter is better, do it, already mentioned, +1

i want to add:

you could try a large project with many instruments, do it in 44,1kHz and in 96kHz
bring the 96kHz one down with r8brain pro.
Compare.

To me its really dramatic what happens, which is a result of my RME converters and how the DAW and plugins work an get together, in my opinion. When using external hardware fx it really gets more dramatic to me...

[jamwerks]

Originally Posted by The Byre

And yet even I can hear a difference on some occasions! So what is going on?

In general, yes it does sound better, but not because we need the extra bandwidth or dynamic range to appreciate the music, but because of the inadequacies of earlier equipment.

Earlier converters created all kinds of distortion and even self-generated noises at the higher frequencies. One DAW even had a constant whistle at about 18kHz. This lead to all kinds of unpleasant artifacts and distortion in the upper frequency ranges. The worst offenders were DAT and ADAT machines, cheap computer sound cards and even some prestige converters.

This upper range distortion was what some people called digital sharpness, or digital harshness. In reality, it was just good, old fashioned distortion. When ever I mastered a piece of music that had been recorded, using an older box, the spectrum analyser would show the line 'sticking' at the top somewhere above 17kHz. This is when I became interested in the subject and dug to find out more.

Now, if you double the sample rate to 96kHz (or 88.2kHz for a CD) when making the original recording, then all that distortion taking place between say, 14kHz and 22kHz is moved to between 28 and 44kHz - i.e. out of hearing range.

Suddenly, all that upper distortion vanishes and we come to the conclusion that a 96kHz sample rate is superior to 44.1kHz.

Excellent post.

[stevetgn]

Originally Posted by The Byre

It varies with the quality of the converters.

I often hear that higher sample rates are required if we are to hear music properly. Despite the rather obvious fact that our ears only go up to 18kHz at the very best for an adult and usually somewhere between 14 and 16Khz, (depnding on age, health and the amount of exposure to loud noise) and the fact that no studio mics go beyond 22kHz, people claim to be able to hear an improvement when listening to material recorded at higher sample rates.

At the same time, several scientific tests have shown that when the same material is played back at different sample rates (and where the source material was at the very highest rate possible) none of those tested could hear any difference between 192kHz and 96, 48, or 44.1kHz.

Many hi-fi companies and enthusiasts assert that the wider bandwidth and dynamic range of SACD and DVD-A make them of audibly higher quality than the CD format. However many carefully controlled double-blind tests with many experienced listeners showed no ability to hear any differences between formats. Therefore the perceived quality has nothing to do with additional resolution or bandwidth.

And yet even I can hear a difference on some occasions! So what is going on?

In general, yes it does sound better, but not because we need the extra bandwidth or dynamic range to appreciate the music, but because of the inadequacies of earlier equipment.

Earlier converters created all kinds of distortion and even self-generated noises at the higher frequencies. One DAW even had a constant whistle at about 18kHz. This lead to all kinds of unpleasant artifacts and distortion in the upper frequency ranges. The worst offenders were DAT and ADAT machines, cheap computer sound cards and even some prestige converters.

This upper range distortion was what some people called digital sharpness, or digital harshness. In reality, it was just good, old fashioned distortion. When ever I mastered a piece of music that had been recorded, using an older box, the spectrum analyser would show the line 'sticking' at the top somewhere above 17kHz. This is when I became interested in the subject and dug to find out more.

Now, if you double the sample rate to 96kHz (or 88.2kHz for a CD) when making the original recording, then all that distortion taking place between say, 14kHz and 22kHz is moved to between 28 and 44kHz - i.e. out of hearing range.

Suddenly, all that upper distortion vanishes and we come to the conclusion that a 96kHz sample rate is superior to 44.1kHz.

Very often, it is not even the converter's fault, but can be the result of poor cabling, or interference from other equipment. Discussions with hardware engineers would suggest that clocking was a major problem for some systems.
So if you want to play it safe, record at the higher rate to ensure that no distortion caused by jitter or other gremlins come in with the original material, but the improvement is not because you have golden ears!

+1 great reply!

[Fletcher]

Originally Posted by The Byre

This upper range distortion was what some people called digital sharpness, or digital harshness. In reality, it was just good, old fashioned distortion.

In addition to the distortion [and yes... some definitely exists] you have the phase shift created by the aliasing filters which can also add to the perception of "digital sharpness/harshness"... the quality and construction of those filters is of paramount importance [as well as the clock, etc., etc., etc.].

Peace.

[Geert van den Berg]

Originally Posted by JOHN

Happy Xmas to all
Is 96khz better than 48khz for multitrack digital recording?
i want to record real instrument sounds

Cheers

Do a search on posts written by Dan Lavry, here, but also on other audio forums and you'll find a lot of good info!

[Ethan Winer]

Originally Posted by JOHN

Is 96khz better than 48khz for multitrack digital recording?

See this:

The Audio Critic Articles | CD vs. SACD/DVD-A

--Ethan

[tubehead]

FWIW, I agree with Fletcher. I like rock n' roll lower, though for me it's at 48k and I mix out of the box.

If you're into Classical or Jazz the higher rates seem to work better for that material.

[chrispick]

Originally Posted by Fletcher

Depends... I have found 96k better for a lot of stuff that I wanted to be a measure more "audiophile" in it's result... and 44.1k better for things I wanted to sound more "rock and roll" [have an attitude]... which is kind of amusing because I have found something very similar with 30 and 15 ips in the analog domain.

FWIW I don't like 48k sampling that much because it's a noticeable sonic improvement [in the 'audiophile' dept.] to 44.1k and should you stay in the box it will require a 'sample rate conversion' to make a CD... which I have always seemed to find more destructive to the audio than the higher sampling rate is beneficial.

As always... YMMV

I agree with this. 96K for audiophile stuff/jazz/classical. 44.1K for rock/pop.

Personally, I find the sonic difference between 44.1 and 48 to be negligible at best. Certainly not worlds apart.

Lastly, you need to have data throughout that can sustain 96K files. Depending on your system, this may not be practical.

[Bob Olhsson]

The problem with converters is that they are a moving target and almost more like comparing different mixing speakers rather than comparing line level analog gear. Musical translation is the whole name of the game. Once you start listening for the changes different converters make to musical balances, converter and sample rate issues become quite a headache. Less than stellar converters can mask problems and even the presence of instruments in a mix. I don't think this issue gets examined nearly enough.

I suspect a major reason analog summing often works so well, in addition to the obvious phase randomization, is that it's a bit like optimizing a mix for NS-10s and then putting a really neutral microphone on the speakers and recording their sound accurately. The whole idea is to have the listener hear as close as possible to what we heard while we mixed. This is why a technique that technically degrades audio remains a perfectly legitimate production choice. The bottom-line is that we are creating a transparent communication link between the artist and listener as opposed to simply transparent reproduction of the sounds created by the artist.

My experience has been that most inexpensive converters do a more transparent job at 48k and that if I want less transparency, some kind of analog treatment ahead of the converter is more likely to translate well to both cheap and high-end converters.

[dispensate]

hello everyone, I just bought a Presonus BluetubeDP off E-bay and we hooked it up for the first time the other night and......well here we go.

the #1 input will not send a signal above the VU meter -10 (this is a M-audio 1010Lt system we're using so some of the signal was picked up anyway)

however the #2 input rocks great. anybody got any ideas as to whats broke in the thing?? My thoughts...bad input jack....bad gain pot??

you guys tell me....

[6strings]

Originally Posted by Ethan Winer

See this:

The Audio Critic Articles | CD vs. SACD/DVD-A

--Ethan

Interesting read. For those who skipped the link:

"They prove beyond a shadow of a doubt, with literally hundreds of double-blind listening tests at matched levels, conducted over a period of more than a year, that the two-channel analog output of a high-end SACD/DVD-A player undergoes no audible change when passed through a 16-bit/44.1-kHz A/D/A processor. That means there’s no audible difference between the original CD standard (“Red Book”) and 24-bit/192-kHz PCM or 1-bit/2.8442-MHz DSD."

I know they're talking distribution medium here, but interesting study nevertheless...

[MickeySmid]

The lavry ad-10 can add transformer and tube sound to the signal to be converted, but it only does this at 88 or 96khz... not at 44.

So, having just bought this one... we will have to switch!

[Ethan Winer]

Originally Posted by 6strings

"They prove beyond a shadow of a doubt

Yet life goes on and people continue to ignore the proof.

--Ethan

[Sounds Great]

Originally Posted by 6strings

"They prove beyond a shadow of a doubt, with literally hundreds of double-blind listening tests at matched levels, conducted over a period of more than a year, that the two-channel analog output of a high-end SACD/DVD-A player undergoes no audible change when passed through a 16-bit/44.1-kHz A/D/A processor. That means there’s no audible difference between the original CD standard (“Red Book”) and 24-bit/192-kHz PCM or 1-bit/2.8442-MHz DSD."

One word:



Bull

[6strings]

Originally Posted by Ethan Winer

Yet life goes on and people continue to ignore the proof.

--Ethan

LOL!

[severe]

Originally Posted by Sounds Great

One word:

Why not the whole word?

"Meyer and Moran do not say that 14 or 15 bits in a truncated CD are just as good as 20. What they say is that spot-on 16-bit/44.1-kHz processing is as good as it gets, audibly."

Love that part.

[Sounds Great]

Originally Posted by severe

Why not the whole word?

I think the message is there.

I have CD's that sound great! No question that a great sounding record can come off of of 44.1 PCM. But to say that is as good as recording can possibly be and we proved it in double blind tests is just laughable.

Is the claim also that if you were sitting in the control booth and heard a live take that there would be NO audible difference in playing it back on a PCM recording? Not only ridiculous, but impossible to "scientifically prove" one way or the other. You cannot conduct such a double blind test because you could not continually repeat the same live perfomance identically.

Analog tape recordings have a different sound than PCM digital recordings, and DSD digital recordings sound different still. And all these sound in some way different to the actual live recording. On a properly set up good quality listening system, almost anyone can hear these differences.

[jamwerks]

Originally Posted by Ethan Winer

Yet life goes on and people continue to ignore the proof.

--Ethan

Interesting read. If I understand what the artical is saying, then I have to say that I don't agree. Put us in a great sounding room with Barefoot monitors feed by Prism convertors, and play us back 96/24 files and then these same files chopped off at 44.1/16 and I don't think they would sound the same. But then again..........

[DenHerder]

The final word is no matter what medium or rate is, if the performer sucks
the performance sucks.
If it's ****in' beautiful it's ****in' beautiful......

[Sounds Great]

Originally Posted by DenHerder

The final word is no matter what medium or rate is, if the performer sucks
the performance sucks.
If it's ****in' beautiful it's ****in' beautiful......

Not really. I want to have it all.

I have a great performance, 'Kind of Blue', on both CD and SACD. The difference is a clear as day. The CD gets the performance across well enough, but it has not nearly the depth and the clarity of the SACD. I feel more in the room with the players with the latter.

[A LaMere]

huh??

The answers prove that everyone prefers something different I think...

I say.... listen to them all and decide which is best for you.

I sometimes record at 16 bits and think it sounds fine....

The differences between 44.1 and 48 in terms of rock or pop music are a complete wash in my opinion.

[TornadoTed]

This is pure conjecture here but what I have encountered. I think a lot has to do with how trained your ear is.

I have done a few tests with a way less experienced musician friend who has a good ear and he can hear subtle differences in frequency in the low and high end for example but can't hear differences in depth.

I have also found this with other friends of mine when I try to bang on about the virtues of good analogue compression and EQ compared to plugins. They can hear the character of my G-1176 and UAD plugin being generally the same but can't hear the extra depth of the G-1176.

There is more depth and space to 96kHz recordings IMO but someone like my friend who hears in 2D! may not notice any difference from a 44.1kHz recording.

I have to admit that while I can hear the differnce between a 24/96 mix and a 24/44.1 mix but struggle once they are both put down to 16/44.1

[Ethan Winer]

Originally Posted by jamwerks

Put us in a great sounding room with Barefoot monitors feed by Prism convertors, and play us back 96/24 files and then these same files chopped off at 44.1/16 and I don't think they would sound the same.

Whether you want to accept it or not, this is exactly what was proved.

Also, the original article was in the AES journal. So it's not like one person is giving his opinion about what he believes. The more interesting question is why you and others prefer not to believe the original AES article. What exactly do you think was flawed about the tests?

--Ethan

Originally Posted by Ethan Winer

Whether you want to accept it or not, this is exactly what was proved.

Also, the original article was in the AES journal. So it's not like one person is giving his opinion about what he believes. The more interesting question is why you and others prefer not to believe the original AES article. What exactly do you think was flawed about the tests?

--Ethan

Many parameters can biased one way or another a listening test:

- Having visual cues of materials being played, equipment usde, etc.
- Not using A/B/X listening procedure ... double blind setup
- Listening with already an answer in mind about the listening 'dilemma'
- Untrained listeners or lack of listening experience
- Lack of objective vocabulary for subjective listening qualification
- Levels
- Low resolution or biased audio equipment in especially poor room acoustics and setup
- Who has the highest authority*

Unfortunately a lot can go wrong during audio tests

(*) A little story on audio amplifiers listening test with professionals. A friend organized a listening test to compare pro amplifiers: 3 recent switching power supply touring amplifiers with approx. 2x1200W @ 4 ohms (1 black colored amplifier from US manufacturer, 1 blue colored amplifier from EU manufacturer and 1 silver colored amplifier from US manufacturer), 1 Jeff Rowland Hi-Fi amplifier (still 2x 250W @ 4 ohms) and 1 Crown D-75 (warehouse speakers amplifier ... just for kidding). 2 sets of speakers: 1 pair of 2-way pro speakers (15" + 1.4") and 1 pair of 3-way Hi-Fi speakers from Finland (extremely flat and high resolution). Listeners could see speakers but not amplifier switching which were matched in level within less than 1dB gain (Bruel & Kjaer sound level meter at listening position to confirm)

Listeners were 2 big names from amplifier and console pro audio industry accompanied by 2 country distributor main Sales representatives, my friend and an assistant.

It was the funniest test because of the 2 special 'authority' visitors the Sales guys could not resist but try to guess what would be the 'right' answer to give by guessing on 'authority' listeners. Performing 5 times in a row an A/B/X test with different materials the 'best' amplifiers result was changing all the time depending on the music, the mood and soon the drinks!

Best amplifiers judged super Hi-Fi, natural sound, totally neutral at the beginning of test with excellent acoustic music ... were judged worst at the end lacking body and highs ... while soon the D-75 was claimed having more life and grit (sure heavy clipping) by the end of listening test on more live rock music. Totally biased and not objective.

My friend kept laughing for the rest of the day: it was like a piece of improvised audio theater. Everyone enjoyed the test without knowing the results of test sheets (better keep them non disclosed to have everyone in good terms) ... and enjoyed the drinks and conversation.

Listening test results and crietria depend also on who is inside the room...