44.1 vs 96

[mickrich]

I am quiet this week in the studio so have been doing some tests.
This one was mainly for myself but I thought you guys might be interested,
I wanted to check if it's worth recording at 96k. I usually use 44.1k.

I used CannibalBarnes excellent Accept cover for this test.
The session files are at 44.1 so I reamped and printed the mix at 44.1 firstly.
Next I resampled the session to 96k and reamped and reamped then printed the mix without changing anything.
Reamped thrtough TS808 to 6505 to recto cab to 57 to API
The 96k version has the main left/ right guitar reamps recorded at 96k, Ezdrummer will have resampled automatically, all plugins are operating at 96k.

The mix is going out of stereo pairs for drums, bass, guitar, vocals from Apogee Ensemble TB into Zed-R16, through Drawmer 1968me back into Apogee.

So it's the same mix with the exception of the main guitars, plugins running at higher sample rate and the in/out through the console and bus compressor.
I also did a bounce of the 96k version converted by logic when bouncing to 44.1K

There is no processing on the mix prints. They are exactly what came back into the apogee. (so may be useful for mastering practice too).

There was a definite higher CPU and RAM load at 96k but my quad i7 mac mini handled it with ease.
I am aware that this test is not perfect due to the up sampling of the original files.
I didn't get a chance to listen to it yet with fresh ears but will tomorrow.
A potential problem is actually doing an AB test of these files which is something I hadn't thought about. I'm not sure if quicktime can flip sample rates on the fly when listening to the different sample rates and it's not possible to put all the files into the same DAW session without resampling so if anyone has any ideas re this I would love to hear them.

Download zip
https://drive.google.com/file/d/0B1EPqLUI5jInLTdXUTVGMVR4dFU/view?usp=sharing

 

[exoslime]

i havent listend to the files yet, but have an idea about the approach via different listening options, i would set it up the way to compare

a) a mix recorded and worked in 44.1/48khz and mixed down to a 16bit, 44.1khz .wav file
against
b) a mix recorded and worked in 88.2k/96khz and mixed down to 16bit, 44.1khz .wav file

i guess hardly nobody would listen to a 96khz in real life, so i assume i does not really matter how it would sound, so i would set it to the final output format (CD, 44.1,khz, 16bit)

i´m sure there is a notable difference between both mixdowns, even if you end at the same output format, as the original recorded material is different a) more information on the higher sample rate version, and b) the daw and internal summing, plugin processing does also differ (except for plugins that already upsample internally)