[xnor]

Sure, 24 bits during multitrack recording is obviously advantageous but recording at higher sample rates does not necessarily mean better sound.
 
Quote:

When recording in 24/96 we are at mix down using less equalization and the reverb tails just sound so much better,than at 16/44 or 24/44.

So you apply less effects to 24/96 files and say it sounds better than heavier processed files that happen to be sampled at 44.1 kHz? Isn't that dishonest to the customers?
Also, check if your nonlinear plugins/devices are broken (for example if they don't oversample).
 
That reminds me of the files that Linn offered for format "comparison" quite some time ago. The 24/96 files had a clearly visibly different waveform (my guess was less compression, EQ ...) than the 16/44.1 file. The different processing is what made the 24/96 file sound better, not the delivery format, but of course the 24/96 files are more expensive. Is constraining oneself to do less processing so hard that it justifies a higher price?

 

[peterBj]

no what I mean is we dont need to equalize as much when we are mixing the multitrack down to stereo.
And we only sell our own recordings.No remasters.We just sell the Wav file as a one to one copy of the multitrack mixdown.
So I dont know what you mean,dishonest?
 
take a look at our site

 

[xnor]

I didn't know you just offer 24/96 files. What I was going at is companies offering 24/96 and 16/44.1 files but processing the 16/44.1 more (more EQ, compression etc.) so they really do sound worse and then advertise that the more expensive and higher bitrate files sound better due to the format and suppress the fact that they processed them differently.
 
It's like offering two "identical" meals, one made from organic foods (more expensive) and the other one made from conventional foods (cheaper) but to make the one from conventional food taste worse the chef secretly throws in a "bit" of extra spices.
The analogy is not great because organic foods are actually quite a bit more expensive. Recording at 24/96 just requires more disk space which is really cheap these days.
 
I still don't understand why you'd need to use less EQ with 96 kHz.

 

[peterBj]

I still don't understand why you'd need to use less EQ with 96 kHz.

 
hi Xnor
 I don't now why that is is either.It is just a thing we noticed.Maybe some ''technocrat'' here can explain it.
 

 

 

[Jwooten]

Just a comment about being able to actually hear the difference in 16 bit vs 24 bit:
The main thing I noticed when, years ago, moving from cassette to CD is how clear the audio is on CD versus cassette. This is at least partly due to the dynamic range (noise floor) of CD vs Cassette. CD noise floor in many times better than cassette giving the audio a more up front, seemingly bottomless, clarity. It seems to me that similarly the same would be true for 16 bit vs 24 bit. Moving away from the noise floor gives a perceived increase of audio quality without actually changing anything else. Lower volume passages are much higher above the noise floor on the 24 bit recordings than on 16 bit recordings. I know that in a lot of music such as rock and other intense styles, the audible dynamic range envelope is much less, and in a lot of cases, it's hard to distinguish the music from the noise anyway; but, in certain cases, for instance when recording a soft acoustic guitar passage, a whispering vocal part, or other types of delicate passages, the extra dynamic range can be recognized as the "in your face" sound quality without actually being loud. The further away the material is from the noise floor, the clearer the audio sounds. Again, this isn't noticeable on louder, compressed, recordings, but when recording subtle passages, the extra dynamic range of 24 bit over 16 bit recordings is much appreciated and, whether real or imagined, welcomed.

 

[xnor]

With noise shaping the perceived noise floor of a 16 bit track can reach about -120 dB. The recording needs to have an annoyingly huge dynamic range in order for the noise floor to become a problem.
 
If you think you have recordings where 16 bits are audibly "unclearer" to the 24 bit version please post something like a 30 sec snippet of the 24 bit track.

 

[xnor]

Quote:

I don't now why that is is either.It is just a thing we noticed.Maybe some ''technocrat'' here can explain it.

 
Once you're done mastering the 96 kHz file all you need to do is resample it to 44.1 kHz and convert the bit depth to 16 bit, ideally with shaped dither. Should you have no quality resampler try SoX.
 
No further processing is necessary.

 

[bigshot]

In order to hear the difference in noise floor between a CD and high bitrate, you would need to turn the volume up so loud, you would incur hearing damage. It's identical at normal non deafening volume levels.

 

[audiosampling]

16bit vs 8bit, the myth exploded (again) !
 
http://www.audiocheck.net/blindtests_16vs8bit.php

 

[GSARider]

Thanks for the link - very interesting.

 

[audiosampling]

Quote:

With noise shaping the perceived noise floor of a 16 bit track can reach about -120 dB. The recording needs to have an annoyingly huge dynamic range in order for the noise floor to become a problem.
 
If you think you have recordings where 16 bits are audibly "unclearer" to the 24 bit version please post something like a 30 sec snippet of the 24 bit track.

 
You are totally right. And for the purpose of building a convincing demo I used 8-bit files, but the principles are exactly the same for 16-bit files (only 48 dB quieter). Have a listen here, the audio files are online :
 
http://www.audiocheck.net/audiotests_dithering.php
 
A 8-bit file is able to capture intelligible speech as low as quiet as -66 dbFS (the dynamic of a 8-bit file is limited to 48 dB in theory). This would translate into -112 dBFS if a 16-bit file was used (-66-48). 

 

[MrTechAgent]

Interesting thanks

 

[obuckley]

[size=medium]It has been a while, sorry. The world and this thread have moved on.[/size]
[size=medium]However, I do have some issues with a couple of replies to my earlier posts:[/size]
[size=medium]Mastering speakers: We have all (or we should all have, as there are plenty out there) seen adverts for speakers claiming that these are THE speakers used by recording studios in mastering sound for the final cut. Examples include the Yamaha NS4 and NS10, also the Genelec 1031, even Auratones. A bit further up the scale, names like Lipinsky, Focal, Dynaudio. The only mastering studio I know quite well (no names no packdrill) used some rather tacky looking Genelecs coupled with a tiny (5" cube) Carver sub to check their final mixes. I suspect (not least from the price label) that the Lipinskys are quite good, but many of the others are not. Sorry, but they just aren’t. They are selected more than partly because they are popular budget speakers typical of the market that the mix will end up either pleasing or not.[/size]
[size=medium]And FWIW I was not talking about a $300k set-up, I was talking about $300k speakers (e.g. Wilson Alexandria specials – not that I have a pair). [/size]
[size=medium]Bass emphasis: So, allow that a pair of such final review speakers does not produce much below 80Hz, quite a lot less at 60Hz and not much at all below 40Hz. For ease of illustration, say flat down to 80Hz, then 80-60 -3dB, 60-40 a further -3dB, 40-20 a further -12dB. Our engineer masters his final mix based on what sounds good from his NS-10s. You then play the resulting CD on your Wilson Alexandrias (allow, flat down to 20Hz). What happens to the bass? Well, of course room acoustics play a part, but other things equal, you will be getting 3dB more than planned from 80-60 Hz, 6dB more from 60-40Hz and a whole dimension you could not tell even existed from 40Hz on down. [/size]
[size=medium]Even if you have equipment that can be “voiced” to produce a certain frequency response in your particular room, what is the objective? Of course, to minimise peaks and troughs caused by resonances and reinforcements/cancellations in your special environment, but what then? Are you aiming for a flat response 20 – 20k? Because that surely was not what our mixing engineer was hearing coming out of his NS-10s.[/size]

 

[xnor]

Well, there's a difference between studio and reference monitors. You use something like the NS-10 to check how well the mix translates to consumer stuff with little bass. But yeah, I've seen people (mostly in home recording) using something like NS-10's only. Then of course you can get into trouble with low frequencies, but respectable engineers know that.
The master studios I know use something like Lipinski L707's and L150's in surround setups calibrated to be "flat" down to close to 20 Hz, but also hi-fi speakers and grado headphones to check how the mix translates.
 
But this is kinda off-topic. If you want to continue the discussion in this direction maybe open a new thread? I'm sure some other guys would be interested as well, but here it's just getting buried.

 

[bigshot]

The mixing stages I have worked with all had lead engineers whose responsibility it was to design, maintain and calibrate the room. There were usually consumer speakers in the room to check the mix on home equipment at the end, but the main work was done on the calibrated monitors, which were sometimes built into the cowl overhead. Maybe you didn't see the main monitors because they were built ins.
 
The point to calibrating the frequency response to flat is it will sound the same if you have to switch to another mixing stage halfway through your mix. No need to start all over again. Every studio will provide the same sound.
 
Lousy studios don't pay for a lead engineer and don't calibrate their monitors. I only worked in a studio like that once and it was a catastrophe.