Quantcast
48k vs 44.1 - Page 2 - Gearspace.com
The No.1 Website for Pro Audio
48k vs 44.1
Old 8th February 2009 | Show parent
  #31
Lives for gear
 
Glenn Bucci's Avatar
 
41 Reviews written
🎧 10 years
Quote:
Originally Posted by Bob Olhsson ➑️
I think 96k sounds more analog than 44.1. It isn't about how much top-end there is. It's about how much crispy distortion there is. 44.1 often sounds brighter than 96k for this reason.

I also don't buy the theory that degradation is what makes analog sound great.
So there is distortion at 44/24 with 32 bit floating even when your meters don't get that close to 0? What is causing the distortion?
Old 8th February 2009 | Show parent
  #32
Lives for gear
 
1 Review written
🎧 10 years
Quote:
Originally Posted by elan ➑️

ANOTHER QUESTION:

what about IRs? if i'm using 44.1k during mixing, can i use 48k IRs? in convolution reverbs.. or the reverb will be pitched.. ?

thanks
anyone can answer to this?

thanks
Old 8th February 2009 | Show parent
  #33
Gear Maniac
 
Agreed's Avatar
 
🎧 10 years
Quote:
Originally Posted by elan ➑️
anyone can answer to this?

thanks
I would say it depends on the plug you're using.
Old 8th February 2009 | Show parent
  #34
Lives for gear
 
1 Review written
🎧 10 years
space designer
Old 8th February 2009 | Show parent
  #35
Gear Maniac
 
Agreed's Avatar
 
🎧 10 years
I don't mean "if you tell me the plug, I'll tell you the answer," I mean that whether it would cause problems or not depends on whether the plugin you're using resamples either your audio or more likely the impulse response when the two are not matched. I know that KeFIR, for example, which is a guitar cabinet simulator-intended IR plugin, has no problem working at various sample rates.
Old 8th February 2009 | Show parent
  #36
Gear Addict
 
🎧 10 years
Quote:
Originally Posted by Glenn Bucci ➑️
So there is distortion at 44/24 with 32 bit floating even when your meters don't get that close to 0? What is causing the distortion?
There is always distortion. The question is (given it is not a WANTED distortion), will it be audible?

Some digital effects (mostly compressors but also EQ) have the side effect that distortion in frequencies close to the maximum codeable frequency (the so-called Nyquist frequency, which is basically half the sample rate) have an impact on the lower frequencies. This side effect is called aliasing and has nothing to do with distortion caused by digital clipping and little with low sample wordlength.

The higher sample rate you work at, the lower the chances the aliasing takes place in audible frequencies.
Old 9th February 2009 | Show parent
  #37
Motown legend
 
Bob Olhsson's Avatar
 
🎧 15 years
Another advantage of high sample rates is that they move more of any truncation distortion (or better yet dither!) above the audible frequency range.
Old 10th February 2009 | Show parent
  #38
Lives for gear
 
Tube World's Avatar
 
5 Reviews written
🎧 10 years
Does analog gear add this undesirable distoration too or is it just digital plug ins? I am afraid this topic is very interesting to me, but I have no knowledge of why there is distoration as claimed and if I could even train myself to hear it.

When I hear that Madonna's last CD was recorded at 44, and many artist release CD's that were recorded at 44, it makes me beleive that only in a pro studio with $4,000 monitors with proper acoustic treatment and someone showing you a particular distoration they found at 14hz in a mix will you find it.

But I really respect Mr. Olhsson and others here with their experience and trust that they do know a lot more than me. If someone could explain this a little more, I would appreciate it.
Old 10th February 2009 | Show parent
  #39
Lives for gear
 
Tube World's Avatar
 
5 Reviews written
🎧 10 years
Ok, I did a little more research and understand it a little better. Then analog gear won't give you this distortion.

The other question is if you only have a decent but not great SRC like the one in Wavelab, or Samplitude, will you still hear an improvement when you downsample to 44?
Old 10th February 2009 | Show parent
  #40
Gear Addict
 
🎧 10 years
Quote:
Originally Posted by Tube World ➑️
Does analog gear add this undesirable distoration too or is it just digital plug ins? I am afraid this topic is very interesting to me, but I have no knowledge of why there is distoration as claimed and if I could even train myself to hear it.
Any analog device will add some sort of distortion. But the aliasing I was talking about happens in digital effects only. A lot of distortion created by analog gear is considered desirable by many, that is 1 out of 2 reasons why so many programmers try to emulate analog gear in software (the other is off-topic: it's about familiarity for those that are used to analog gear).

Quote:
When I hear that Madonna's last CD was recorded at 44
I am not sure how major an aspect pristine sound quality is in a Madonna recording. I also guess they used a lot of analog gear so aliasing may not have been much of an issue.

Quote:
it makes me believe that only in a pro studio with $4,000 monitors with proper acoustic treatment and someone showing you a particular distortion they found at 14hz in a mix will you find it.
You'll hardly be able to do an A/B test of a complete mix with otherwise exactly the same recording. Tests say that:
- the difference between 44kHz and higher sample rates is not noticeable to most people
- the impact of lower sample rate (let's say 44 vs 96) on effects quality is noticeable to average listeners on decent stereo equipment.
Old 10th February 2009 | Show parent
  #41
Gear Addict
 
🎧 10 years
Quote:
Originally Posted by Bob Olhsson ➑️
Yes and even sometimes 48 vs 44.1. The issue is how much garbage gets generated below 20kHz.

I think UAD says they do an on the fly up-sample. I don't have their stuff but every on the fly DSP up-sampling I've heard hasn't sounded as good as simply working at the higher sample rate.
What means : working at the higher sample rate?
Do you mean :

1/working with a file (44,1 sampled for exemple) BUT using higher sample rate processing? (88,2 or 96 / ITB or digital EQ and compressor)

2/working with a file recorded at higher sample rate since the beginning?

Quote:
Originally Posted by Bob Olhsson ➑️
Another advantage of high sample rates is that they move more of any truncation distortion (or better yet dither!) above the audible frequency range.
Same question here,
are u talking about the advantage of high sample rate recording? or high sample rate processing (with a lower sample rate recorded sample for exemple)?

I think many people are confusing the 2 things when the speak about 96 vs 44,1
Old 10th February 2009 | Show parent
  #42
Gear Addict
 
🎧 10 years
Quote:
Originally Posted by Spartacus ➑️
What means : working at the higher sample rate?
2 things should be avoided:

1. too many stages that introduce aliasing (such as digital EQ or digital compression)
2. too many upsampling/downsampling stages.

So I think it is perfectly acceptable to record at 44.1, then upsample everything to 88 or 96, do all the processing in 96 and sample back down to 44 second last to dithering.

You save another upsampling step by recording at 96 right away but I think the difference is negligible, compared to the impact by the recording situation, miking, preamp and ADC quality.

A "project sample rate" of more than 44kHz for ITB mixing will give you an advantage.
Old 10th February 2009 | Show parent
  #43
Lives for gear
 
Tube World's Avatar
 
5 Reviews written
🎧 10 years
Quote:
Originally Posted by waltermusik ➑️
2 things should be avoided:

So I think it is perfectly acceptable to record at 44.1, then upsample everything to 88 or 96, do all the processing in 96 and sample back down to 44 second last to dithering.

You save another upsampling step by recording at 96 right away but I think the difference is negligible, compared to the impact by the recording situation, miking, preamp and ADC quality.

A "project sample rate" of more than 44kHz for ITB mixing will give you an advantage.
I don't get it, when you upsample you are increasing the size of your mix just as you recorded it at 88/96. So what is the advantage of recording at 44 and then upsampling compared to recording at 88/96?
Old 10th February 2009 | Show parent
  #44
Gear Addict
 
🎧 10 years
Quote:
Originally Posted by Tube World ➑️
I don't get it, when you upsample you are increasing the size of your mix just as you recorded it at 88/96. So what is the advantage of recording at 44 and then upsampling compared to recording at 88/96?
you DO get it. If your ADC supports 88 or 96, recording at 44 and then upsampling makes not much sense. My current ADC (a new one is on my wish list) supports only up to 48 so for my next project I plan to either have a new ADC or upsample right after recording.

About the "size of the mix": If your projects are big enough (or your computer small enough) to cause performance issues, a lower sample rate may be preferred over the usage of inferior plugins or engineer annoyance.
Old 10th February 2009 | Show parent
  #45
Motown legend
 
Bob Olhsson's Avatar
 
🎧 15 years
Quote:
Originally Posted by Spartacus ➑️
are u talking about the advantage of high sample rate recording? or high sample rate processing (with a lower sample rate recorded sample for exemple)?
If you are going to be doing signal processing at a higher rate, it just makes sense to record it that way rather than upsampling it. Obviously any conversion needs to be first rate too.
Old 11th February 2009 | Show parent
  #46
Gear Addict
 
🎧 10 years
Quote:
Originally Posted by Bob Olhsson ➑️
If you are going to be doing signal processing at a higher rate, it just makes sense to record it that way rather than upsampling it. Obviously any conversion needs to be first rate too.
Agree,

but I wanted to point that even with a 44,1kHz file, we could wish "working" (= go through 96kHz plugs like Eq, Limiter, etc..) because these plugs give better results while processing at higher sample rate.

-For exemple, with an EQ, the shape of a filtering is better preserved by processing data at higher sample rates.
-Or a limiter can avoid inter-sample peak while processing at higher sample rates too.
Old 11th February 2009 | Show parent
  #47
Lives for gear
 
Tube World's Avatar
 
5 Reviews written
🎧 10 years
If there is an improvement recording at the higher sample rate, I am then surprised why so many choose to record at 44 or 48. Is it because they don't have a 2 quad computer yet, and lack enough power for the plug ins, or is there another reason?
Old 13th October 2010 | Show parent
  #48
Here for the gear
 
🎧 10 years
No matter how you slice it, your signal degrades regardless of what misinformation is out there. If you have to do it (for some good reason), the best algorithm is by iZotope in their most expensive current version of RX. If you are currently recording at 48k, switch back to 44.1, that is unless you score for DVD a majority of the time. If you want to show off your computer's CPU, use 88.2k, not 48 or 96 if your target is a CD or MP3 (or any other 44.1/16-bit target).

Here is the cutting edge of sample rate conversion shoot outs. This will help dispel any skepticism you may have on the matter.

SRC Comparisons
Old 13th October 2010 | Show parent
  #49
Here for the gear
 
🎧 10 years
Bit of a thread-necro there, but it's an opportunity for my thoughts:

I really don't believe that downsampling from 88.2 to 44.1 is any better than from 48 or 96. The 'divide by 2' reason given by so many suggests to me that the downsampling process is a case of removing every other sample, when surely the actual processes involved are a lot more complicated than that. For that reason I don't see why 88.2 should be any better than 96 for the purposes of downsampling. Yes, it may require a little extra computing power but I imagine that to be negligible, particularly because the downsampling process is not real-time.

And if operating at 88.2 or 96 within a DAW, the processing applied by plugins and such would likely smear over any such sonic differences (if they exist) surely?

I'm no DSP expert but I really feel that the belief that 88.2 is better than 96 shows a lack of understanding of what really happens in the downsampling process and is instead just a result of hearsay. I haven't seen anyone on this forum give a detailed analysis of why 88.2 is better other than that their 'engineer friends have said so.'

That said, I could be totally wrong, but am not willing to change my opinion unless there is an in-depth explanation available somewhere.
Old 13th October 2010 | Show parent
  #50
Lives for gear
 
🎧 10 years
People have done tests demonstrating that conversions from 48 or 96 can sound just as good as from 88.2.

I think it's pretty simple, if you're going to record at 44.1 or 48, do it at the rate you're going to release your final versions at.
Old 13th October 2010 | Show parent
  #51
Here for the gear
 
🎧 10 years
This sample rate debate has been beaten to death on various forums.

Me I use

24/44.1 for CD and MP3 conversions

24/48 for DVD material

Don't forget, dithering plays a huge role as well

Then you have the converters and quality of downsampling from 88.2 or 96. Lots of variables to take into consideration. Also, it is documented fact that the human ear is more sensetive to the bit rate than it is to the sample rate.


Question: - How much of the sample rate push is coming from manufacturers of equipment? I don't have an answer but it makes you think.
Old 13th October 2010 | Show parent
  #52
Lives for gear
 
jupiter8's Avatar
 
🎧 15 years
Quote:
Originally Posted by mpeskett ➑️
That said, I could be totally wrong, but am not willing to change my opinion unless there is an in-depth explanation available somewhere.
You're not.
Old 13th October 2010 | Show parent
  #53
Lives for gear
 
jupiter8's Avatar
 
🎧 15 years
Quote:
Originally Posted by gabriel_s ➑️
Don't forget, dithering plays a huge role as well
No it doesn't.
Quote:
Originally Posted by gabriel_s ➑️
Then you have the converters and quality of downsampling from 88.2 or 96. Lots of variables to take into consideration. Also, it is documented fact that the human ear is more sensetive to the bit rate than it is to the sample rate.
What's one thing got to do with the other ? It's not like you can't work at 96 kHz just because you choose to record at 24 bits.
Quote:
Originally Posted by gabriel_s ➑️
Question: - How much of the sample rate push is coming from manufacturers of equipment? I don't have an answer but it makes you think.
I have no idea but personally i have no problem making my mind up when higher sample rates are needed and when they are not. It isn't that hard really.
Old 14th October 2010 | Show parent
  #54
Lives for gear
 
atma's Avatar
 
🎧 15 years
as far as i understand it--the main benefit of using higher sample rates in a daw is to attenuate aliasing which occurs anytime you apply dsp to a signal.

i think people have the misconception that the purpose is simply to have a higher frequency bandwidth, which turns into the argument of whether or not anyone can hear those ultrasonic frequencies, but that isn't really the point.

the higher the sample rate when performing dsp operations on a signal in a daw (i.e., running a sound through an fx plugin, etc.), the less aliasing will tend to occur, or at least the aliasing that does occur will be way up in the ultrasonic range, and should be truncated out by an anti-alias filter when downsampling to cd specs.
Old 8th November 2016
  #55
Here for the gear
 
🎧 10 years
I agree with most the posts that came after mine here. My perspective evolved considerably since my earlier entry here. Here's a more up-to-date perspective from yours truly.

https://gearspace.com/board/10504452-post216.html
πŸ“ Reply

Similar Threads

Thread / Thread Starter Replies / Views Last Post
replies: 13 views: 9502
Avatar for jml designs
jml designs 7th December 2017
replies: 8 views: 4432
Avatar for Threshold
Threshold 2nd November 2006
replies: 126 views: 10103
Avatar for matskull
matskull 10th September 2009
replies: 94 views: 18791
Avatar for Jim Williams
Jim Williams 14th November 2020
Post Reply

Welcome to the Gearspace 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…

Forum Jump
Forum Jump