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
If I record at 16 bit should I bounce down at 24bit?
Old 28th July 2005
  #1
Gear Addict
 
Chrisac's Avatar
 

If I record at 16 bit should I bounce down at 24bit?

This may sound daft I know. My Daw is always set to 16 bit, or I should say that the sounds i use are all 16 bit, and rarely record organic material. I mostly use virtual synths and samples from CDs (the CD's are 16 bit)

Should I bounce down at 24 bit, then reimport to do my pre mastering.

I think I heard that when you use more than 2 plugins on the master buss at 16 Bit something is lost. Is this true?

Even if its not, would there be an advantage bouncing at 16 bit, or am I putting my muisc through a needlss conversion process?
Old 28th July 2005
  #2
Lives for gear
 
Ruudman's Avatar
 

You will get a larger file that includes 8 bits of nothing..

ruudman
Old 28th July 2005
  #3
Lives for gear
 
orange's Avatar
 

If you are going to apply effects to a sound then it is better to do this at a higher bit depth even if the original wav/aif is a lower bit depth. Those extra 8 empty bits will be filled with effected audio.

Converting to 24 bit but keeping the same sample rate involves no quality loss. In general if space/processing-power is not an issue i would work at 24 bit. If you are processing at 24 bit and need to convert to 16 bit at the end (for cd ?) then you would be wise to dither the signal.

Essentially 2 ways to convert 24 bits to 16 bits - truncate or truncate/dither. Truncate chops of the lower 8 bits whilst dithering adds a little bit of noise (usually weighted in some way) so that your 'scissor mark' when truncating is a little blurred. Seems odd that adding noise can give a clearer/better result but try both and you'll see.

si
Old 28th July 2005
  #4
Lives for gear
 
Riad's Avatar
 

If I record at 16 bit should I bounce down at 24bit?

Isn't that bouncing up?
Old 28th July 2005
  #5
Lives for gear
 
Ruudman's Avatar
 

Reminds me of "Spinning Wheel" BS&T.

ruudman
Old 28th July 2005
  #6
Gear Addict
 
Chrisac's Avatar
 

Quote:
Originally Posted by Riad
If I record at 16 bit should I bounce down at 24bit?

Isn't that bouncing up?
I suppose it would be ?
Old 28th July 2005
  #7
Lives for gear
 
zemlin's Avatar
 

If you are mixing or applying effects to the 16 bit tracks, I would make them 24 bit BEFORE you mixdown. A 24 bit bounce will be better for mastering - there may be 8 bits of nothing if it is a converted file, but any effects or processing will appreciate the extra headroom and the end result will be better.
Old 28th July 2005
  #8
Lives for gear
 
joaquin's Avatar
 

can the Mix Bus take advantage of those extra 8bits at bounce time??
...or what we are bouncing is the Mix bus and the audio has already been compress to fit???
Old 28th July 2005
  #9
Gear Addict
 
Chrisac's Avatar
 

Thanks for the replies but Im still no clearer if I should do this...sorry.

I will try and explain:

If I were recording guitars/vocals etc I would definitely work at 24bit. But i rarely record real instruments. All my sounds come from sampled stuf I did years ago on my Roland s750 whiich have been converted to audio (16 bit) and Audio sample CD's, again 16 bit.
So, apart from the extra headroom 24bit would give my reasonsing for sticking with 16 bit while writing/composing is because 95% of my sounds are 16 bit. I rarely use virtual synths which Im sure might benefit from working at 24 bit...but Im mainly a Samplist type.


....................................................................................
What Im asking about is when I have finished the track and am Bouncing down/rendering the stereo master which at the moment is always 16 bit. Once thats done I will import the stereo audio file back to the Daw and proceed to work on the stereo master adding EQs/multiband Compression etc. Iwould do this at the render/bouncing stage but I dont hav ethe CPU power to run all those heavy mastering type plugins. On some stuff I have used up to eight plugins.

So, would I be better to render at 24bit (even though all sounds are 16 bit)....import back to the Daw as a 24bit file, do my mastering thing using a lot of plugins (up to 8) and when bouncing/rendering downsample to 16 bit using the usual methods.


Its nothing to do with disk space. Its purely because someone mentioned that plugins work better at 24 bit especially if using a lot of them on one channel for mastering.

Am I making sense?

Old 28th July 2005
  #10
Gear Addict
 
Chrisac's Avatar
 

Quote:
Originally Posted by zemlin
If you are mixing or applying effects to the 16 bit tracks, I would make them 24 bit BEFORE you mixdown. A 24 bit bounce will be better for mastering - there may be 8 bits of nothing if it is a converted file, but any effects or processing will appreciate the extra headroom and the end result will be better.
I posted the long one before I saw this.

yes, thats exactly what Im talking about. Just treating the master.

I understand the extra headroom, but how do you know the end result would be better. Do you mean in sound quality?

If I had the CPU power would it be better to do this at the first bouncing/render stage, Or do the 24bit thing each time.?


Some tracks I do I have the available Cpu, but not most.
Old 28th July 2005
  #11
Mastering Moderator
 
Riccardo's Avatar
Always bounce at the same sample rate and bith depth as the original session.
There are only a few reasonably good hardware SRC and very few "acceptable" software SRC ou there.
Old 28th July 2005
  #12
Gear Addict
 
Chrisac's Avatar
 

Quote:
Originally Posted by Riccardo
Always bounce at the same sample rate and bith depth as the original session.
There are only a few reasonably good hardware SRC and very few "acceptable" software SRC ou there.

Anyone else agree with this?

This is what I thought and is why I have always bounced at 16 bit, but there seems to be a difference of opinion. I see your a mastering engineer, but Ive been asked for 24 bit masters even when I tell them the session was at 16 bit?

Now im very confused
Old 28th July 2005
  #13
Lives for gear
 
zemlin's Avatar
 

I do all my SRC with Adobe Audition, which is supposed to be one of it's real strengths. The statement may be true with other software.

I do all my recording at 24 bit anyway, so SRC from 16 up is not something I do very often.
Old 28th July 2005
  #14
Lives for gear
 
blackcom's Avatar
 

When running plugins and digital summing at 32-bit internal precision (like most daws) I think bouncing to 24 would improve sound qualtiy, yes!

Just think about it.... when summing two 16-bit tacks into a new 16-bit track, something will be lost. However, most daws have 32-bit internal precision wich kinda takes care of this (I think...), so going 24-bit afterwards would serve you well.....and plugins adds materiall too...
Old 28th July 2005
  #15
Lives for gear
 
orange's Avatar
 

Quote:
Originally Posted by Riccardo
Always bounce at the same sample rate and bith depth as the original session.
There are only a few reasonably good hardware SRC and very few "acceptable" software SRC ou there.

agreed about the sample rate but disagree about the bit depth....converting up just pads with zeros.

si
Old 28th July 2005
  #16
Lives for gear
 
Ruudman's Avatar
 

Quote:
Originally Posted by zemlin
I do all my recording at 24 bit anyway, so SRC from 16 up is not something I do very often.

I'm glad to hear, since SRC is Sample Rate Converting

Quote:
Just think about it.... when summing two 16-bit tacks into a new 16-bit track, something will be lost.
Loose what? Other than dynamic range?


ruudman
Old 28th July 2005
  #17
Lives for gear
 

I recommend (and practice) that you bounce to the highest bit depth and sample rate that your system can handle. Then bounce down to your destination format with appropriate dither and all should be good.

I believe that staying at the same bit depth/sample rate as your destination is soooo 1998.

Of course, some SRCs are better than others, but it would be wrong to assume that hardware SRC is inherently better than software, or vice versa, or that expensive SRC is inherently better than software, etc.

Take a small section of your music. Bounce it straight across, bounce it at your highest available quality, bounce both with dither down to your destination (try a couple of flavors of dither while you are at it). After critical listening, if you hear a lot of difference, or no difference, you will know which route to take.

Nika's book on digital audio, available from Sweetwater, will make the technology and science clear.
Old 28th July 2005
  #18
Lives for gear
 
Ruudman's Avatar
 

Are you suggesting that when bouncing to a stereo file, you should convert both
bit depth AND samplerate up to highest rates, then down again? Who on earth would draw benefit of that?


ruudman
Old 28th July 2005
  #19
Lives for gear
 

Quote:
Originally Posted by Ruudman
Are you suggesting that when bouncing to a stereo file, you should convert both
bit depth AND samplerate up to highest rates, then down again? Who on earth would draw benefit of that?


ruudman
Seagate and Western Digital
Old 28th July 2005
  #20
Lives for gear
 

Quote:
Always bounce at the same sample rate and bith depth as the original session.
There are only a few reasonably good hardware SRC and very few "acceptable" software SRC ou there.
I'd agree on the sampling rate...but not on the bit depth. Any time you mix two files together the resultant file will have more bits than the originals, so if you mix two sixteen-bit files together you won't be just adding zeros if you bounce down (and it is bouncing down since you're going from more to less tracks). That's why digital mixers mix at a higher resolution than the tracks are recorded at. Also, depending on how dynamic your stuff is you could also wind up with more than sixteen bits' worth of dynamic range when you mix multiple sixteen-bit tracks together, and as mentioned your effects will benefit from the added resolution as well.

As has already been mentioned, SRC has nothing to do with this and since the system mixes at a higher bit depth it's going to be dithered/truncated anyhow. Might as well keep it at 24-bit resolution as long as possible.

-Duardo
Old 28th July 2005
  #21
Lives for gear
 
John The Cut's Avatar
 

I cant believe some of the mis-information in this thread.


Chrisac - wordlength (bit depth) always increases with ANY process you apply to the audio.

If you have a 16bit file on a track and you move the fader position, you now have >16bit bit depth.

This is the nature of digital audio

You need to set your project to 24bit. You also need to apply dither to get it back down to 16bit.

At the moment you are throwing away the internal resolution offered by your system.
Old 28th July 2005
  #22
Lives for gear
 
joaquin's Avatar
 

Quote:
Originally Posted by Duardo
I'd agree on the sampling rate...but not on the bit depth. Any time you mix two files together the resultant file will have more bits than the originals, so if you mix two sixteen-bit files together you won't be just adding zeros if you bounce down (and it is bouncing down since you're going from more to less tracks). That's why digital mixers mix at a higher resolution than the tracks are recorded at. Also, depending on how dynamic your stuff is you could also wind up with more than sixteen bits' worth of dynamic range when you mix multiple sixteen-bit tracks together, and as mentioned your effects will benefit from the added resolution as well.

As has already been mentioned, SRC has nothing to do with this and since the system mixes at a higher bit depth it's going to be dithered/truncated anyhow. Might as well keep it at 24-bit resolution as long as possible.

-Duardo
Hey Duardo. I make my question again....do you think that there would be any benefit for the summing process in the Mix bus if you use a higher bit depth than the actual session's at bounce??
Thanks......Joaquin.
Old 28th July 2005
  #23
Lives for gear
 
zemlin's Avatar
 

Quote:
Originally Posted by zemlin
I do all my recording at 24 bit anyway, so SRC from 16 up is not something I do very often.
uhhh - I do know that going from 16 bit to 24 bit isn't SRC. Maybe I should have had that extra cup of coffee today.



hey - I think I smell a fresh pot brewing now!

I still say (with absolutely no doubt in my head) that you should use 24 bit for as much of your processing as possible.
Old 28th July 2005
  #24
Gear Addict
 
Chrisac's Avatar
 

So many different opinions..but thats good I think..

Im just going to try bouncing at 24 bit, do my mastering thing and dither down to 16 bit. Ill do one at 16 Bit and see if I can hear a difference.

Apart from all my sound sources being 16bit another reason I dont use 24 bit is because Ableton Live (the daw I use) is very heavy on the cpu. My Daw struggles enough as it is with a track count of 16-20 tracks as it is now, and I dont use tons of plugs.......its not my computer as I get much more milage with SX3....If it werent for that then all sessions would be 24bit.......Now this is where I miss my Pro Tool days ...

Thanks all for the advice

Old 28th July 2005
  #25
Lives for gear
 
Riad's Avatar
 

Quote:
Originally Posted by Chrisac
Im just going to try bouncing at 24 bit, do my mastering thing and dither down to 16 bit. Ill do one at 16 Bit and see if I can hear a difference.
Give that man a cigar.
Old 29th July 2005
  #26
Motown legend
 
Bob Olhsson's Avatar
 

You may not hear much difference until after the audio has been processed additionally. Higher resolution creates a less fragile signal which I think is more important than simply sounding a little better at one stage.
Old 29th July 2005
  #27
Mastering Moderator
 
Riccardo's Avatar
Quote:
Originally Posted by orange
agreed about the sample rate but disagree about the bit depth....converting up just pads with zeros.

si
Sorry I should have made it clear.
The point I was trying to make is that if you have for example a 24 bit 48 kHz session , you should then bounce at the same sample rate and bith depth and leave the rest for the mastering.

The same concept applies to dithering and SRC.

I am not saying that software SRC are inerently worse than hardware just that there are very few good ones out there.

Now, to judge a src - hardware or software - AND DITHER you do need a reasonably good monitoring chain (i.e. d/a converter - amp - speakers - ROOM )

I am still surprised at how so many people for example star threads about a zillion mic preamps costing 1, 2 or 3K and go on talking about the differences when they are listening on 2k or less worth of monitors (not to into room acoustics, amplification and d/a conversion)

It is obvious that if he can set up the session at 24 bits from the start there will be advantages.

Sorry for the off topics.

Old 29th July 2005
  #28
Lives for gear
 
joaquin's Avatar
 

Quote:
Originally Posted by joaquin
I make my question again....do you think that there would be any benefit for the summing process in the Mix bus if you use a higher bit depth than the actual session's at bounce??
Thanks......Joaquin.
...................................................???
Old 29th July 2005
  #29
Gear Maniac
 
smarsland's Avatar
 

The original question was do I bounce to 24-bit with a 16-bit session, I do believe.

I don't understand what the benefit would be. If it was originally a 24-bit session, then there is 24-bits worth of useful information. If it's a 16-bit session, only 16-bit words are being recorded.

What's the advantage of adding zero's? Makes a hell of a bigger mess on your hard drive. I like the response about Seagate getting a benefit from that bounce.

Convince me otherwise, smart people.

s
Old 30th July 2005
  #30
Lives for gear
 
joaquin's Avatar
 

If I have a session at 16 bit, with 32 tracks of audio, they will have to be summed into the Master Mix Bus, wich in the case of most native systems run at 32 bit floating. Now, is recommended from digidesign that you dither to 16bit even when bouncing from a 16 bit session due to the higher (32bit floating) internal resolution. Now, would I benefit amplitud/resolution wise if I acually dither to 24bit instead of 16??????.......................Thank you.......Joaquin.
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
Similar Threads
Thread
Thread Starter / Forum
Replies
passionmax / So Much Gear, So Little Time
27
studjo / So Much Gear, So Little Time
5

Forum Jump
Forum Jump