The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Audio for both CD and DVD?
Old 1 week ago
  #1
Gear Maniac
 

Thread Starter
Audio for both CD and DVD?

Suppose one desires to record audio for both CD and DVD.

Is it preferable to record at sampling rate of 44.1 or 48 Khz?

My first thought was 48. But, then I thought wouldn't that require an additional stage of antialiasing filtering, so maybe 44.1 is better.

But, then I also thought that with good numerical technique, it probably doesn't really make a whole lot of difference.

Still, is there a generally accepted best practice?

Thank you.

DG
Old 1 week ago
  #2
Gear Maniac
 

Use 44k, definitely. No matter what SR you start out with, once it is down sampled, it can never sound as good as the native 44k recording. Do a test recording yourself by recording your speakers playing something twice, once at 44k native SR and once with higher SR. Down sample the high SR recording to 44k and compare it with the native 44k recording. You will agree with me in 5 seconds or less.

Best regards,

Da-Hong
Old 1 week ago
  #3
While I agree that the more processes you put your audio through damage the sound in minute ways. I record in 48kHz when part of the audio will be headed toward video. The vast majority of recordings today are made at something higher than 44.1, even when a CD is the end result, and the end result does not suffer because of it, at least not in any way all but the most highly trained ears in a direct comparison could detect. Modern sample rate conversion is rather good, though some programs are more ever so slightly transparent than others.

just my 2 cents.
Old 1 week ago
  #4
Gear Maniac
 

Thread Starter
I think Da-Hong's point, though, is that, if, at the end of the day, if a "red book" CD is to be produced from a signal sampled at >44.1Khz, then decimation will necessarily occur (and is best avoided).

DG
Old 1 week ago
  #5
So is there a consensus that upsampling to 48k doesn't introduce anything unwanted?
Old 1 week ago
  #6
Up sampling will give you the same problem. THough it has the added annoyance of being audibly pointless, since no additional frequency information is gained.

At any rate, the audio is decemated by the converter ship itself. Which some, more knowledgeable people than i, claim is worse than software. So there is a good argument that higher sample rates, brought down to 44.1 in the box, gives you better results. I have no doubt that there is a difference since I have experienced it myself, but I am not thoroughly convinced one yields "better" results. And in that case, I usually opt for the more convenient and practical option of 48kHz, as that is the highest sample rate requested.
Old 1 week ago
  #7
Gear Maniac
 

Thread Starter
I don't fully understand this. Doesn't upsampling involve only interpolation (not decimation)?

DG
Old 1 week ago
  #8
Lives for gear
 
loujudson's Avatar
Record and process and mix and master at 96k and then make two versions dithered for the two media. Simple, and it'll sound better!
:-)
Old 1 week ago
  #9
Quote:
Originally Posted by dgpretzel View Post
I don't fully understand this. Doesn't upsampling involve only interpolation (not decimation)?

DG
It doesn't involve any further limiting of the high frequencies, I don't think anyone implied it did. Nor does it open up any new high frequencies. All it really does is smear the image a little bit like down sampling. Since both ways are bad in some tiny way, I would personally prefer the method that contains more info in the high frequencies to begin with.

I see nothing wrong with starting at 44.1. Especially if you know it will only be for cd. But the reasons for that are practical rather than audible.
Old 1 week ago
  #10
Quote:
Originally Posted by loujudson View Post
Record and process and mix and master at 96k and then make two versions dithered for the two media. Simple, and it'll sound better!
:-)
There is a better chance they will both sound the same. Better? Eeeehhhmmmm.........
Old 1 week ago
  #11
Lives for gear
 
loujudson's Avatar
Quote:
Originally Posted by rumleymusic View Post
There is a better chance they will both sound the same. Better? Eeeehhhmmmm.........
It it a well known fact that you can get better sound with 96k, if you start there in the first place!

And think about it - both CD and DVD get equal treatment in the end.
Old 1 week ago
  #12
(WELL KNOWN FACTS)
Old 1 week ago
  #13
Lives for gear
 
loujudson's Avatar
"Everyone is entitled to his own opinion, but not his own facts."
Daniel Patrick Moynihan

But hey, in the age of trump, maybe your facts have changed...
Old 1 week ago
  #14
Gear Head
 
Simmosonic's Avatar
 

If your only options are 44.1k or 48k, I’d go for 44.1k on the grounds that the audio for video often goes through a perceptual encoding process of some kind anyway before being released, so keep the best quality for the version that is not going through such processing. Also, the CD audio has to stand alone whereas the video footage will be accompanied/distracted by video footage and does not have to carry the show on its own.
Old 1 week ago
  #15
Gear Maniac
 

When someone is watching a video, he or she will never listen to the audio as carefully as that person would to a CD. Besides, it is less likely people will have good audio systems hooked up to their video system. Audio is always 2nd class citizen when it comes to video world.
Old 1 week ago
  #16
Lives for gear
 
jwh1192's Avatar
Quote:
Originally Posted by Simmosonic View Post
If your only options are 44.1k or 48k, I’d go for 44.1k on the grounds that the audio for video often goes through a perceptual encoding process of some kind anyway before being released, so keep the best quality for the version that is not going through such processing. Also, the CD audio has to stand alone whereas the video footage will be accompanied/distracted by video footage and does not have to carry the show on its own.
yes, like a Dolby / DTS encoding process .. if the Main Finished Product has an Audio Only Delivery Format component .. the Favor the Audio Format !!! and let the video chips fall where they may !!! my .02 from doing a lot of audio for DVD authoring .. !! cheers john
Old 1 week ago
  #17
Lives for gear
 

I am in the FP32/96K camp for all tracks I now record in my studio. Archival storage is dirt cheap and a broad based assumption that retail packaging is in a perpetual regressive state is dark side neanderthal thinking. There is no objective evidence that mastering FP32/96K projects to Red Book or various compressed lower sonic levels is in any way a perceptual destructive process. In the event the clientele that pays your bills has no prayer of a performance future then by all means record it where it will ultimately be delivered then scrub the memory and start over clean with your next customer. However if archival factors exist and you possess the gear to work in FP32/96K arguments against it are pretty lame!
Hugh
Old 1 week ago
  #18
Gear Guru
 
UnderTow's Avatar
I would work at 48 KHz and downsample offline to 44.1 KHz for the CD. This let's the anti-aliasing filters in the ADC be a little further out of the audible range. Use a good quality offline sample rate converter to go to 44.1 KHz. (I'm a big fan of the free SoX converter).

Alistair
Old 1 week ago
  #19
Gear Guru
 
UnderTow's Avatar
Quote:
Originally Posted by hughshouse View Post
However if archival factors exist and you possess the gear to work in FP32/96K arguments against it are pretty lame!
Hugh
They are only lame if you don't understand the subject. For instance, no converter reaches even full 24 bit performance, going over 24 bit is therefore pointless for recording. Also, recording 8 extra bits so the files are FP instead of integer, adds absolutely nothing. A pointless exercise.

As for 96 KHz there are both advantages and disadvantages. There is no clear win either way but for the vast majority of productions, working at 96 KHz has more negatives than positives (hence the vast majority of commercial projects being done at 44.1 khz or 48 KHz for film, broadcast or DVD).

Alistair

Last edited by UnderTow; 1 week ago at 02:15 PM..
Old 1 week ago
  #20
Lives for gear
 
jimjazzdad's Avatar
Quote:
Originally Posted by hughshouse View Post
I am in the FP32/96K camp for all tracks I now record in my studio. Archival storage is dirt cheap and a broad based assumption that retail packaging is in a perpetual regressive state is dark side neanderthal thinking. There is no objective evidence that mastering FP32/96K projects to Red Book or various compressed lower sonic levels is in any way a perceptual destructive process. In the event the clientele that pays your bills has no prayer of a performance future then by all means record it where it will ultimately be delivered then scrub the memory and start over clean with your next customer. However if archival factors exist and you possess the gear to work in FP32/96K arguments against it are pretty lame!
Hugh
Quote:
Originally Posted by loujudson View Post
Record and process and mix and master at 96k and then make two versions dithered for the two media. Simple, and it'll sound better!
:-)
This thread runs the risk of becoming a parallel thread to "Is it time for a new (preferably higher) standard? Yes? No?". There, the promise of higher sample rates is pursued ad nauseam. The OP in this thread only asked: "Suppose one desires to record audio for both CD and DVD. Is it preferable to record at sampling rate of 44.1 or 48 Khz?". No mention was made of recording at sample rates higher than 48 kHz.

So which is preferable - up sample to 48 kHz or down sample to 44.1 kHz? Or is there any difference? I am also interested in the answer as I am sometimes asked by the videographer for a 48 kHz copy of a CD archival recording I do.
Old 1 week ago
  #21
Gear Guru
 
UnderTow's Avatar
Quote:
Originally Posted by jimjazzdad View Post
So which is preferable - up sample to 48 kHz or down sample to 44.1 kHz? Or is there any difference? I am also interested in the answer as I am sometimes asked by the videographer for a 48 kHz copy of a CD archival recording I do.
From a theoretical point of view, see my before last post. In practice, it will hardly matter as long as the tools used are at least half decent.

Alistair
Old 1 week ago
  #22
Lives for gear
 
Bruce Watson's Avatar
 

Quote:
Originally Posted by dgpretzel View Post
Suppose one desires to record audio for both CD and DVD.

Is it preferable to record at sampling rate of 44.1 or 48 Khz?
Yes. If you know for sure that the project requires both, the cleanest thing you can do is record both.

Else I would record at a higher sample rate (88.2 probably because to my way of thinking the CD would have priority since it's audio only) and downsample for delivery.
Old 1 week ago
  #23
Lives for gear
 
loujudson's Avatar
Quote:
Originally Posted by Bruce Watson View Post
Yes. If you know for sure that the project requires both, the cleanest thing you can do is record both.

Else I would record at a higher sample rate (88.2 probably because to my way of thinking the CD would have priority since it's audio only) and downsample for delivery.
Someone ere will probably try to shoot me down, but 88.2 and 96k take about equal amonts of digital math to convert to either 44.1 or 48k. It is not true that 88.2 goes better to 44.1 than 48k, and 96k alos takes the same amount of calculations fo 48 and for 44.1.

It is nothing like divide by 2. So *I* would use 96k, not 88.2, no matter what th final conversion is to...
Old 1 week ago
  #24
Gear Guru
 
UnderTow's Avatar
Quote:
Originally Posted by loujudson View Post
88.2 and 96k take about equal amonts of digital math to convert to either 44.1 or 48k. It is not true that 88.2 goes better to 44.1 than 48k, and 96k alos takes the same amount of calculations fo 48 and for 44.1.
Correct.

Alistair
Old 1 week ago
  #25
Lives for gear
 
loujudson's Avatar
Quote:
Originally Posted by UnderTow View Post
Correct.

Alistair
Whew, thanks. I was half expecting someone to challenge me on that...
Old 1 week ago
  #26
Lives for gear
 
Bruce Watson's Avatar
 

Quote:
Originally Posted by loujudson View Post
Someone ere will probably try to shoot me down, but 88.2 and 96k take about equal amonts of digital math to convert to either 44.1 or 48k. It is not true that 88.2 goes better to 44.1 than 48k, and 96k alos takes the same amount of calculations fo 48 and for 44.1.

It is nothing like divide by 2. So *I* would use 96k, not 88.2, no matter what th final conversion is to...
OK. It's been so long since I did any investigation into this that I could well be wrong. I've forgotten more than I remember that's for sure.

My point was to record in a higher resolution and downconvert to the lower deliverable resolution.
Old 1 week ago
  #27
Lives for gear
 
loujudson's Avatar
Quote:
Originally Posted by Bruce Watson View Post
OK. It's been so long since I did any investigation into this that I could well be wrong. I've forgotten more than I remember that's for sure.

My point was to record in a higher resolution and downconvert to the lower deliverable resolution.
Agreed! I got it on excellent advice from Bob Ohlsson. I rarelly forget anything he has told me, and consider him a major authority on audio both digital and analog. Saved me lots of mistakes and time.

Once he demonstrated for me personally what a difference proper dither makes when reducing bit rate. With his superior monitoring and careful processes, the difference it stunning!

I was extremely lucky to have "Bob O." as a mentor for many years. He's the best!
Old 1 week ago
  #28
Gear Maniac
 

Thread Starter
If I am understanding you correctly, this is what I hear you to be asserting:

"If one records at sample rate 'A', and also records at a higher sample rate 'B', but decimates it back down to 'A', then the resulting derived product is superior to that originaly recorded at rate 'A'."

Is this an accurate summary of what you are saying?

Thank you.

DG
Old 1 week ago
  #29
Lives for gear
 
loujudson's Avatar
Quote:
Originally Posted by dgpretzel View Post
If I am understanding you correctly, this is what I hear you to be asserting:

"If one records at sample rate 'A', and also records at a higher sample rate 'B', but decimates it back down to 'A', then the resulting derived product is superior to that originaly recorded at rate 'A'."

Is this an accurate summary of what you are saying?

Thank you.

DG
No, not at all! The point is that both the CD and the DVD will have equal quality if they both come from a higher rate. If you record at 44.1 or 48, the one that gets resampled will be sightly inferior to the one that was not resampled. Might be very slight, but you asked!

and resampling is not quite "decimating" if done properly with good software. Well done, it will be imperceptible.

Hope this helps.
Old 1 week ago
  #30
Gear Maniac
 

What everybody forgets here is that the audio track of a DVD can easily be 44.1KHz. DVD spec allows the audio to be basically anything you want, up to 96KHz if I remember it correctly. All the DVD player can play whatever the sample rate the disc happens to be. You can even have different sample rate for different part of the video footage. There is really not need to insist on using 48KHz. Whereas for CD, you can only use 44.1KHz.

So, record 44KHz for CD and DVD. No up or down resample.

Da-Hong
New Reply Submit Thread to Facebook Facebook  Submit Thread to Twitter Twitter  Submit Thread to LinkedIn LinkedIn  Submit Thread to Google+ Google+  Submit Thread to Reddit Reddit 
 
Topic:
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