1. Register NOW and become part of this fantastic knowledge base forum! This message will go away once you have registered.

Dithering

Discussion in 'Recording' started by John Murphy, Apr 29, 2002.

  1. John Murphy

    John Murphy Guest

    Today i mixed down a track recorded on cubase vst32.
    One mixdown was mastered internally using the cubase internal dithering uv22(ie audio export)the other was mixed to a cdr digitally via spdif.
    The difference in quality was too much for me to ignore.The cdr version sounded like my mix, the internal cubase mixdown sounded a bit naff in comparison.It was unfocused,a lot of presence was missing,in short i didn't like it.
    Can anyone tell me of better ways to mixdown internally using Dithering and what are the best dithering plug-ins to use.

    John
     
  2. Doublehelix

    Doublehelix Well-Known Member

    John...any time you are changing the resolution and sample rate, you are dithering in some fashion. Dithering is by definition, intentionally introducing some noise into your file, it is just controlled! Each dithering algorithm differs in the type and location (frequency) of the added noise. This is done as bits are removed from the end of the file and rounding errors occur.

    At what resolution were the tracks originally recorded? (24/44.1K? 24/48K? 32/96?)

    You mention that you "mixed to a cdr digitally via spdif"...how was this done? Was it just mixed down to a WAV file, or was it converted to a regular audio format that can play in any standard consumer CD player? If it was converted, then it had to have been dithered as well. Again, please help us by explaining how this was accomplished...through Cubase again, or was there another method?

    You also mention that you exported the file using the UV22 dithering...I am assuming that it was exported as a WAV file also...

    Now you burn the CD...were both files burned as standard audio CD files, or were they both burned as WAV files? How did you listen to them to compare? In order to fairly compare the files, they should both be in the same format (WAV or audio), and auditioned from the same format (CD or Hard drive) and sound system (monitors vs. headphones).

    Sorry to ask so many questions, but we need some more info here to help you!

    Cheers!
     
  3. John Murphy

    John Murphy Guest

    DH thanks for your interest,i'll try and answer as much as possible.

    "At what resolution were the tracks originally recorded? (24/44.1K? 24/48K? 32/96?)."

    32bit/44.1Khz.

    "You mention that you "mixed to a cdr digitally via spdif"...how was this done? Was it just mixed down to a WAV file, or was it converted to a regular audio format that can play in any standard consumer CD player? If it was converted, then it had to have been dithered as well. Again, please help us by explaining how this was accomplished...through Cubase again, or was there another method?"

    The mix was sent through the spdif of my Delta 1010 soundcard using cubase,it then went to a HHB CDR which recorded it digitally as an audio format,which i could instantly play back within the recorder and was able to play it back on ordinary CD players.

    "You also mention that you exported the file using the UV22 dithering...I am assuming that it was exported as a WAV file also..."

    This was exported to a 16bit/44.1Khz .aif(mac)file within cubase,then burnt to a CDR using toast CD writing software.

    I listened to both the CDs(audio) at the studio using the HHB CDR through some genelic 1030A monitors.I then took them home and listened on my CD player through my hifi unit.
    The reason i went to the studio was because i was sure something wasn't right with the internal audio export i originally did.It just didn't sound the same as what was coming through the mix on my computer via the soundcard.

    John
     
  4. Doublehelix

    Doublehelix Well-Known Member

    Thanks for clearing things up!

    You say you went through Cubase to the HHB CDR? That means that the file had to be dithered from 32 bit to 16 bit *somehow*. If you had UV22 selected, then that was the protcol used to dither this file also. If not, then it used the "standard" Cubase dithering algorythm, which it sounds like you like better.

    The new version of Wavelab (also by Steinberg) has the new and improved UV22 protocol by Apogee, called UV22HR (high resolution?). I think the sound is superior to the "regular" UV22. There are other plug-ins available for dithering also. The first one to come to mind is the Waves IDR (this is bundled with most of the Waves plug-in sets, and is integrated into the L1 Ultra-Maximizer.

    Click on this link to follow a similar conversation thread that was going on here a few months ago, and that may also answer some of your questions.

    (Dead Link Removed)

    Good Luck!!!
     
  5. Opus2000

    Opus2000 Well-Known Member

    If you put the UV22 plugin on the last chain of your master effects you are indeed dithering to 16bit. In Cubase I believe it's on the master section allready and all you have to do is enable it, set it to 16bit and Normal and you're rocking!
    Now, UV22HR indeed is a higher resolution dither. The "tables" as we call them are a series of algorythmic mathemtical equations to reduce, add noise and dither at the same time while maintaining the original quality of the signal.
    Now, when you outputted the signal from your Delta1010 via the SPDIF to the HHB, what setting was the delta set to? Was it set to 24bit? If so then I'm wondering whether the HHB has it's own internal dithering IC components?
    If you have the UV22 on in Cubase and you hit export audio with 16bit set, you are dithering twice and that is why your mix didnt sound as good as it did going straight to the HHB. If you want to do it hat way select the LOW operation on the UV22 section if applicable. Low is for when you know you are going to dither more than once.
    I hope this helps explain things a little more. There's still a major part of dithering that I dont' understand only because I have seen what it takes to do that! It's mind numbing when you speak to the guy who created the UV22!!! The numbers that spew out of his mouth are too technical even for me! :p
    Opus
     
  6. Doublehelix

    Doublehelix Well-Known Member

    Oh yeah??? Take this:

    10010010 10010111 00010101 11000011

    Hehe!!! ;)
     
  7. John Murphy

    John Murphy Guest

    Opus said

    "Now, when you outputted the signal from your Delta1010 via the SPDIF to the HHB, what setting was the delta set to? Was it set to 24bit? If so then I'm wondering whether the HHB has it's own internal dithering IC components? "

    The bit rate is set in cubase as far as i know and that was set to 32bit floating point.All the Tracking files at mixdown were 32bit.
    I've got a feeling that it was the HHB that did the dithering.When i get back to the studio i'll read the manual and let you know.

    John
     
  8. Opus2000

    Opus2000 Well-Known Member

    Yes, Cubase was set to 32 bit but what was the card set to? This might be a reason why it worked..you probably had the Delta card set to default which is 16bit. Cubase doesnt care that the card is set to 16bit and the program is set to 32bit..Cubase will still create 32bit files no matter what..it will just add bits to get it there!
    Opus
     
  9. John Murphy

    John Murphy Guest

    Opus as far as i am aware there is no way to change the bit rate in the Delta card,only the sample rate.There is no mention of it in the manual and no parameters for it in the Delta soundcard control panel(software).

    John
     

Share This Page