Quick Poll ||| How Many of You Are On 96KHZ ?

[25ghosts]

Been doing many tests myself... Between 48 and 96. When comparing in a blind test I am NOT able to tell which one is playing. Putting a low cuz at 16KHZ - then I can tell a small difference - but I dont know many listening to music with a low cut that high...

So - I would love to hear what you guys are doing and what your PERSONAL experiences are - Rather than what you know to be a fact on white papers

[Howardk]

44k1 here.

In my opinion the big problem with these comparisons, and resulting opinions, is they are generally meaningless. This is because the comparisons are most often apples to oranges, not apples to apples, or oranges to oranges. What I mean is there are so many variables and one must take 2 of the exact same sources with the exact some situations, at the exact same time, with the only difference being the same rate, in order to do any valid test comparision.

Years ago, when I first switched to HD, I tried to create a valid test and did as follows:

1) Started with a project that was recorded and mixed in HD at 24 bit 96k, resulting in a 96k stereo mix (a).
2) made a copy of the session down to 44k1, then printed the mix again at 44k1 (b).
3) reduced the 96k stereo mix(a) to 44k1 (c)
4) imported both the (b) + (c) into an HD session, lined the files up in time and did double blind comparisons, with associates to keep each other honest.
5) took the 44k1 mix (b) and increased the bit-rate to 96k (d)
6) imported (d) into a 96k session with the original 96k mix (a), and did the same type of listening comparisons.

At all times the audio remained at 24 bit.

In these comparisons, a number of different studio monitors and headphones were used. In the end, no one involved could make an repeatable accurate distinction between the mixes in either scenario 3) or 4). Either the test procedure scenario is flawed, we are all deaf, or people in general can't hear the difference. I suggest it is the former is correct.

Anyone else done a test like this? What were your results?

Can anyone suggest a better method?

My thoughts further to this, if there is an audible difference that some can hear (ie. superman, my wife, young people that haven't yet heard 1,000,000 snare hits from their iPods, and probably my wife), it is so slight that it is immaterial. Therefore I suggest it makes more sense to use HD's DSP horsepower for things that are more material/important (i.e. better sounding, often more DSP intensive plugins), so I continue to record and mix in HD at 44k1, at the same sample rate as I did in my Mix system prior.

[Vaporhead]

44.1 all of the time. Besides not being able to hear any difference, the colossal depletion of available DSP and the ridiculous file size at 96KHZ makes it a no brainer for me. Are we trying to make the best records, or whatever, with the largest amount of available creative tools or are we trying to kid ourselves into believing that the alleged benefit of a higher sample rate will make any perceivable difference in the end result of the creative process.

My two cents.................

[rnd]

44.1 (48 if that's how a project comes to me...but I don't like it!)

[25ghosts]

Hi - Thanks 4 sharing...

One of my tests was this:

I recorded a rap vocal onto 2" tape.

Then I routed the 2" tape into PT and recorded once at 96 and once at 48.

Each session had three settings on vocal.

1) Flat
2) Boosted Mid
3) Boosted Highs.

After the test was done - both sessions were bounced in their 3 respective versions and all 6 individual files were converted to 44.1/16 Bit.

After that I opened the songs in Bias Peak and did a blind test between the correct matches. In the blind test I fooled myself again and again.

So I did this:

added a low cut at 16KHZ - then took a listen again. The 96KHZ files still understandable and clear (what you could hear was of course only ssssssss) and the 48khz was very unclear...

Yet - In the blind test with the real vocal running - I was not able to tell the difference without engaging into a guessing contest.


I have been using 96KHZ sample rates for a while now... But now I am getting tired of those huge file sizes and less DSP power. And when I think about the fact that I still like that Korg M1 piano which is a compressed 44.1/16Bit sample ROM then I start to feel stupid for using a sample rate that high - especially when I cant tell the difference....

I switched to 48KHZ today.

[Howardk]

I switched to 48KHZ today.

I could be wrong, but I remember noticing certain plugins use more DSP when running at 48k vrs 44k1, and 48k is also a bit more data storage. Why not go all-the-way and run lo-fi at 44k1?

[M2E]

I've been using 96k for a while and I can hear a difference. To me it's when I'm mixing that I'm able to get a better mix for some reason. The plugins sound better when tweaking an EQ. I can get the sound I want faster. I know when I'm in 44.1k and when I start mixing, the plugins don't have the same "BAM" to me, whatever that iz.
I had a session the other day and I didn't know what it was but, soon as I sat down and started tweaking stuff, I was like, something ain't right. I looked at the session info and it was 24bit/44.1k. I was like ahhhh, that's why.
I could be wrong, as it could be my brain playing tricks on me but, for some reason, I can tell. I haven't did a test about it yet or token any test as well but, I wouldn't mind doing one to see if my mind iz playing tricks on me.
I know in Logic Pro 8 that when I play the EXS24 piano in 44.1k it is dull with no life at all, soon as I switch it up to 96k, WOW, it comes to life bigtime. The sounds in Logic sound huge at 96k. Not really sure in Pro Tools when it comes to Instruments.
Maybe that could be a way to see if there's a clear difference.
I'll do a track in both 44.1 48k and 96k, same song, same instruments and mix it down with the same plugins and see if there's a difference since VI's don't have any hiss as their very clean sounding.
Maybe I'll do it tonight and post it tomorrow if I have time for ya guys to see. I'll keep an eye out on this post if ya have any suggestions.

Marc...1ne

[dbjp]

44k1 here.

In my opinion the big problem with these comparisons, and resulting opinions, is they are generally meaningless. This is because the comparisons are most often apples to oranges, not apples to apples, or oranges to oranges. What I mean is there are so many variables and one must take 2 of the exact same sources with the exact some situations, at the exact same time, with the only difference being the same rate, in order to do any valid test comparision.

Years ago, when I first switched to HD, I tried to create a valid test and did as follows:

1) Started with a project that was recorded and mixed in HD at 24 bit 96k, resulting in a 96k stereo mix (a).
2) made a copy of the session down to 44k1, then printed the mix again at 44k1 (b).
3) reduced the 96k stereo mix(a) to 44k1 (c)
4) imported both the (b) + (c) into an HD session, lined the files up in time and did double blind comparisons, with associates to keep each other honest.
5) took the 44k1 mix (b) and increased the bit-rate to 96k (d)
6) imported (d) into a 96k session with the original 96k mix (a), and did the same type of listening comparisons.

At all times the audio remained at 24 bit.

In these comparisons, a number of different studio monitors and headphones were used. In the end, no one involved could make an repeatable accurate distinction between the mixes in either scenario 3) or 4). Either the test procedure scenario is flawed, we are all deaf, or people in general can't hear the difference. I suggest it is the former is correct.

Anyone else done a test like this? What were your results?

Can anyone suggest a better method?

My thoughts further to this, if there is an audible difference that some can hear (ie. superman, my wife, young people that haven't yet heard 1,000,000 snare hits from their iPods, and probably my wife), it is so slight that it is immaterial. Therefore I suggest it makes more sense to use HD's DSP horsepower for things that are more material/important (i.e. better sounding, often more DSP intensive plugins), so I continue to record and mix in HD at 44k1, at the same sample rate as I did in my Mix system prior.

Your test is valid to an extent, but I think your test only begins after the recording process, so you're likely to not getting the full result.
Your 44.1k session was a converted version of a session recorded in 96k. I think it makes a huge difference.

Personally, I can only tell the difference in quality when recording an acoustic instrument like a gut guitar, though I'm sure some engineers would beg to differ. The overall quality when mixing does seem better to me but that's possibly my brain telling me to think that way.

I'm seriously regretting going 96k on my current session which is R&B/pop style. My track count's reaching ridiculous new heights (snare alone's taken up 10 tracks) and I've still got to add more string parts and vocal tracks. I'm thinking of importing the session into a 48k one.

I'd love to know what the top engineers of say Rock, R&B, Jazz, Funk, etc use for their respective work in different fields.
Anyone have extracts from interviews?

[WorldStudios]

Hi!

Many people ask me about this - people who produce music for a living. They never seem to have a fixed opinion...

I use 44.1 almost all the time. Not for film work, because it needs to be 48 and sometimes I go all out and run 88.2if it is just a few live instruments, like vocals and guitar only.

When I first got my HD system back in 2002, when it came out, I did a lot of songs in 88.2 (because it is an even division of 44.1) and I still think those songs sound more open. I did, however spend a LOT of time on them, which has something to do with the result. But they also sound a bit soft. And bass light. They contained lots of guitars, electric and acoustic, vocals etc. I also did some dance tracks a year later at 88.2 again. Recently I decided to convert one of the dance tracks to 44.1 as a test. The resulting mixes sounded exactly the same, but we are talking all electronic sound sources, like synths, drum samples etc, that were probably never ever more than 44.1 to begin with... Last year I did a track for a major act and wanted to go 88.2. I did, but they felt the song was too polished sounding. I got the wild idea to convert it to 44.1. They now thought it sounded much better! Bad meaning good... Most of the time, folks I work with whant the sound hard and tough and cool. Then 44.1 is the ticket. If they want soft smooth and airy, 88.2 might be the way to go, but there is a huge price to pay. I run very big sessions sometimes with over 100 tracks, lots of plugins etc and I think I would have to do quite a bit of submixing and AS processing to get it to work at 88.2. And that is what, at the end of the day, makes me go 44.1 at all times these days.

It is going to end up on a crummy mp3 on a russian site tomorrow anyway. Or be played on commercial radio with stacks of multiband compressors on acid beating the life out of it...

[pmklein]

Like many, I started doing 96k when I got the HD rig. However, I missed the convenience of being able to patch all of my digital fx into the mix (Lexicon 300, TC3000, TC2000, Finalizer, etc.) Being able to have all those available with no drain on PT is a bigger plus to me than a 96k session. On top of that, I can't hear a big sonic difference in the end unless it's something like acoustic guitar and vocal. I think the quality of the converter is more important than the sample rate. IMHO.