The No.1 Website for Pro Audio
96kHz vs 44.1kHz tests again.. sorry
Old 4 weeks ago
  #1
Here for the gear
96kHz vs 44.1kHz tests again.. sorry

Hi Folks
Im really sorry for bringing up the same old question, but can anyone direct me to any published papers or serious tests made in the comparison between 96k and 44.1k that don't include the same file being converted or downsampled? Im trying to collect some data and Im finding it hard to find published tests
Many thanks and apologies
joe
Old 4 weeks ago
  #2
Here for the gear
 

Listen to Dan Worrall explaining everything about it: https://youtu.be/-jCwIsT0X8M
Old 4 weeks ago
  #3
Here for the gear
 

Not published but some 1st hand experience

Hi,

I also couldn't find anything published on this that fit what you were after.

I actually don't think anyone has ever done a true test on this - probably because I can't see any organisation justifying funding research on this, and anyone doing it by themselves - even if it was done professionally, would be subject to the usual controversy/arguing that ensues when people get into this. Not worth the hassle!

I was doing a string sample pack about a year ago and decided to record it all at 96kHz, to enable me to do my own tests - although, even this was not as you would probably hope - the main problem being that how do you record simultaneously at 44kHz and 96kHz?
I don't have the capabilities to do that with two identical systems and some kind of splitting device at the right impedance.

But I did get my most complex loop - made from about 20 layered string samples - and I downsampled all the samples, saved a different version of the same project file, replaced all the 96kHz samples with 44kHz ones, and listened to the result.

The difference was clearly audible to anyones ears (the loop in question can actually be heard here - it's the first sound you hear when you play the demo - hover the cursor over the album cover and play button appears), and the difference was still there even once I'd downsampled the mixed down version of the 96kHz project to a 44kHz file (doing that didn't change the sound).

I wish I'd have saved the resulting files now - I cleared out a bunch of unneeded files recently and ditched the key ones.

The 44kHz project version lacked lower mids, and had more mids to high mids. But this result could've been completely by chance - it's one comparison after all.

I own personal opinion is that the difference only comes in once you start mixing complex sounds that are similar. Because I was layering string parts that had been recorded on the same instrument, same mic, same setup, etc. Each take was very similar, which means when they're all layered there are many phase cancellations occurring. Even when you have two parts layered like this, moving one of them a tiny bit either way can make the bass disappear, or become really big.
This is only a loose guess - I have studied acoustics, but rather than making me more sure, it actually makes me less sure, because I know that without a scientific process it's all just guess work.

And of course, if I'd recorded into two identical systems - one at 96kHz and one at 44kHz, that the same loop could've well ended up sounding identical. I really couldn't guess either way on that.

One further point of interest, I did the test with different sample rates - 44kHz, 48kHz, 88kHz and 96kHz, and the result from the 48kHz project sounded almost identical to the one from the 96kHz one.
Not that surprising given that one is half the other - so the samples that were downsampled from 96kHz to 48kHz would have theoretically changed quite a bit less than the ones downsampled to either 88kHz or 44kHz.
Or, any phase cancellations occurring from the mixing of all the 48kHz downsampled samples would have been very similar to those that occurred from the original 96kHz ones - because every sample point of the 48kHz samples would correspond exactly with every 2nd sample point of the corresponding 96kHz sample.

Old 3 weeks ago
  #4
Lives for gear
 
esldude's Avatar
https://www.researchgate.net/publica...kHz_vs_882_kHz

You can download the entire article as PDF. Using quality gear recorded concurrently at both 44 and 88 khz rates. Then using blind listening tests. Done by McGill University.

Summary, no difference heard between native 44 khz and native 88 khz. However, a difference was heard comparing native 88 khz vs downsampled 44 khz. Apparently downsampling was audible. The downsampler software was Pyramix if I recall rightly.

Mics were Senn MKH8020, Millenia mic preamps, Micstasy ADCs,
Old 2 weeks ago
  #5
Here for the gear
thanks for the replies.
Its a great help, i thought i was missing something.
It does appear that almost all tests are comparing a filtered/ converted file and so what people may hear is the filter as a difference.
thanks for the link to the paper!!!
its interesting to note that there is a lot of tests not published when there is a negative result (cant hear a difference)

thanks for the help

Joe
Old 2 weeks ago
  #6
Quote:
Originally Posted by esldude View Post
https://www.researchgate.net/publica...kHz_vs_882_kHz

You can download the entire article as PDF. Using quality gear recorded concurrently at both 44 and 88 khz rates. Then using blind listening tests. Done by McGill University.

Summary, no difference heard between native 44 khz and native 88 khz. However, a difference was heard comparing native 88 khz vs downsampled 44 khz. Apparently downsampling was audible. The downsampler software was Pyramix if I recall rightly.

Mics were Senn MKH8020, Millenia mic preamps, Micstasy ADCs,
I clearly hear a difference between 44 and 88 in the higher frequencies but hey what ever works for you.
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump