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
dithering and soft synths
Old 9th October 2005
  #1
Gear Maniac
 
beatzz's Avatar
 

dithering and soft synths

I have been curious that if you are making a dance song with many soft synths and you do not print any of them. At the end u go export: 16 bit 44 khz.

Is the program host doing the dithering? Is this preferrable to a Waves L2?
Old 9th October 2005
  #2
What exactly is it you want to do?

Normally you shouldn't use dithering until the very last stage, going from internal bit depth or 24 bits to 16 bits on the final mix, and then hands off.

Depending on what sequencer program you are using, there's usually a manual way of selecting the kind of dither you want. In Logic Pro you have the choice between POW-r#1, 2 or 3.

Using the Waves L2/L3, you can select any of the IDR types. Then turn off any dithering in the sequencer.

You can check out my thread called The Dithering Test to actually hear the difference between all these types of dither.
Old 9th October 2005
  #3
Gear Maniac
 
beatzz's Avatar
 

Thanks, i always assumed that if u had the L2 dithering on the master bus you wouldn't need to turn of the program's host own dithering. I guess i was wrong.
Old 9th October 2005
  #4
Oh yeah, you'll be getting dithering x 2. Those bounces must have been quite a bit more noisy and high freq annoying.
Old 12th October 2005
  #5
Lives for gear
 

Quote:
Originally Posted by Lagerfeldt
What exactly is it you want to do?

Normally you shouldn't use dithering until the very last stage, going from internal bit depth or 24 bits to 16 bits on the final mix, and then hands off.

.
FYI you should dither any and every time you reduce word length. I.e going from 48 bits to 24 dither...or 64 dither or 32 dither etc.....
Old 13th October 2005
  #6
Quote:
Originally Posted by Teacher
FYI you should dither any and every time you reduce word length. I.e going from 48 bits to 24 dither...or 64 dither or 32 dither etc.....
That's a bit of a misunderstanding. It all depends on where in the process you are.

You should never dither multiple files or dither several times in a row. Multiple layers of dithering add a veil to the sound and can also cause ultra high frequency build up. As dithering is a non-periodic noise signal (or periodic in the case of UV22) it also raises the noise floor, and the (noise-shaping) algorithms can add some nasty high frequencies, and will become very audible when layered. Also I would be hesistant to dither (depending on the dither type) when further processing of the file is required. Have you ever tried pitching or equalizing a file with dither on? Funny stuff happens to the noise-shaping, e.g. a drum beat will get a nasty high pitched sound when pitched even just a bit down.

The best solution is always to keep the bit depth as high as possible for as long as possible, and then dither the final result when going to 16 bit. Of course self-dithering is another issue as this happens without user control. Using 24 bits for individual files is IMO more than enough, and should pose no audible problems in a mix. Also notice that it will also depend on what software you're using. Using 32 bit float internally and then dithering to 24 bits fixed doesn't make a lot of sense, since the two formats are sonically speaking virtually identical.
Old 14th October 2005
  #7
Lives for gear
 

Quote:
Originally Posted by Lagerfeldt

The best solution is always to keep the bit depth as high as possible for as long as possible,
true.

The rest of what you said; not true.

You're not gonna hear a noise signal 130 db down, not even multiple ones. And even if you did hear them, you'd rather listen to that than truncation errors. Nothing sounds worse than that imo.
Old 14th October 2005
  #8
Lives for gear
 
catfish11's Avatar
 

i like dither
it sounds good,

tech people who cannot hear should not
respond to sound oriented disscussions
they lack the crucial qualification:

ears
Old 14th October 2005
  #9
Lagerfeldt knows his stuff.

I'm all for using your ears -- but for God's sake, use your brain, too.
Old 14th October 2005
  #10
Lives for gear
 

"your petty insults are of no consequense"

that's hilarious heh
Old 14th October 2005
  #11
Quote:
Originally Posted by juicemaster1500
true.

The rest of what you said; not true.

You're not gonna hear a noise signal 130 db down, not even multiple ones. And even if you did hear them, you'd rather listen to that than truncation errors. Nothing sounds worse than that imo.
Not right again.

It seems weird that first (in another thread) you claim to hear truncation from 32 bit float to 24 fixed (when there really is none), and then to now not being able to hear the implications of dithering noise signals? Truncation errors from 24 to 16 are perceptible, not so from 32 bit float to 24 fixed e.g. You really need to get the facts straight before transferring your audio paranoia unto others. :-)

A lot of dithering sounds horrible, and multiple layers will definitely sounds increasingly worse. What I'm objecting to is not the (broadband) noise as such but the HF noise shaping.

And regarding soft synths (that are of course by nature, digital) multiple layers of dithering will definitely worsen matters with some nasty HF digitalis, not very analog or warm to my ears. Again, truncation is a big issue when going to 16 bits, not so with 24 bits. The solution is, as I already said, to just stay in 24 bits, and the dither the final master.

The original poster asked about dithering soft synths for dance music.

So having years of experience in producing and mixing #1 selling dance and pop music using soft synths, I'd recommend the original user to NOT use multiple layers of dithering.

That's not tech advice, that's just plain good advice.
Old 15th October 2005
  #12
Lives for gear
 

if you don't like noise shaping, just turn it off. use the plain dither if you like it better.

the thing is, I hear the difference going from 32 to 24 bits loud and clear, even with a crap soft synth.

there's different types of dither though, a dither to 24 bits is lower in volume than a dither to 16 bits. I have a bunch of plugins that use dither to 24 bits.. waves and sony etc.

if multiple layers of dither was so bad, how do you explain that all respectable 64 bit internal prsessing plugins DITHERS the output down to 32 bits..

dither is everywhere.
Old 15th October 2005
  #13
Lives for gear
 
catfish11's Avatar
 

i would imagine multible dithering would be a bit much

i use powr #2 generally on the mix buss in protools,
once

sorry about the tech comment, it was not really directed to anyone here, just a general commentary of what one
comes into contact here so often, regarding the tech approach,as opposed to the ear approach
Old 20th October 2005
  #14
Lives for gear
In my host, I mix in a 64 bit engine. I then bypass rendering, opting instead to use the Tapeit2 plugin on the 2 bus to record a stereo wav of the mix straight from the host.

I then import the wav into soundforge and master it using Izotope. I also dither down to 16 in it. So I dither once..from a 32 bit wav. down to a 16.

To me, the sound translates well, but I was wondering if anyone had opinions on the way I do it?


also, I always track in 44.1
Old 20th October 2005
  #15
Gear Addict
 

Dithering is always last !

Is your host is running at larger then 16-bit, and you wanna render/bounce a final stereo master for CD, then dither.

Or you make a 24/32-bit stereo pre-master, and dither that in a seperate process later.
Old 20th October 2005
  #16
Lives for gear
Quote:
Originally Posted by AngeloClematide
Or you make a 24/32-bit stereo pre-master, and dither that in a seperate process later.
yeah, thats what Im doing...no rendering of any kind..a 24/32 bit .wav captured out of the host. then mastering/dithering to 16.
Old 20th October 2005
  #17
Gear Addict
 

Yes, as easy as that !

Sample Player Plug-Ins have a dithering engine build in, resp. are transcoded by the host engine, and don't need extra dithering. With Soft Synth's it doesn't matter, since the sound is calculated, and not stored as with a Sample player.

For example: Logic's EXS-24 lets you use samples of different sampling rates or bit depht in the same project without altering the pitch, so it doesn't matter if the samples are stored in another bit depth, the audio engine of Logic and EXS-24 will handle that in near realtime.

AIF/WAV audio files on the timeline, have to have the same sampling frequenz, but Bit depth can differ.
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
dangeorge6 / So Much Gear, So Little Time
19
Igotsoul4u / Music Computers
0
rolo / Gearslutz Secondhand Gear Classifieds
2
Igotsoul4u / Music Computers
9
tluke / So Much Gear, So Little Time
2

Forum Jump
Forum Jump