To dither or not to dither that is the? Waves IDR?
 

My sessions go from 24/48 001 ..mixed to 16bit/44.1 stereo interleave
and then into Soundforge 5.0 where i use Waves Ultramaximizer DirectX (can't use it in 001)to get my tracks leveled out volume wise. L1 has IDR dithering options...my question..
Should I use any of the dithering options?
Hasn't 001 already dithered when I converted to 16 bit?
Do I need to use IDR again after using plugins on my 16bit stereo master?
Should I leave the mixes 24bit and use Waves IDR at the end?
Am I an idiot ..I can't figure this out...ok don't answer that one [img]images/icons/rolleyes.gif[/img]
thanks everyone.

Re: To dither or not to dither that is the? Waves IDR?
 

Great question.. I do know that you will get dithered bouncing to disk-IF-you add the dither plug in to the master fader at bouncing. I wish someone would explain the chain of processing from mix to cd. Like,if dither should be last,then what about the using of soundforge,l-1,and the "filters" and processing you can do in cd burning s/w after you have bounced and dithered??? I'm using NERO. THANX [img]images/icons/grin.gif[/img]

Re: To dither or not to dither that is the? Waves IDR?
 

Hey,,Anybody got any answers?? [img]images/icons/wink.gif[/img]

Re: To dither or not to dither that is the? Waves IDR?
 

Mix from ProTools to a 24 bit master stereo track. I open that track in Sound Forge 5 and do any further processing (mastering) while the file is still 24 bit. After you have done the processing your last step should be running the Waves IDR in Sound Forge. I usually start with the "16 bit, highest resolution..." preset and work from there.

When you run the IDR the file will still display that it is 24 bit at the bottom of the screen. This confused me too. The Waves IDR has converted the file to 16 bit, but the last 8 bits will just be storing zeros which is why it displays "24 bit". You simply need to right-click on "24 bit" and change it to "16 bit". This will convert it to 16 bit (truncate the last 8 bits) without doing any further dithering.

The basic idea here is stay 24 bit as long as possible while doing any processing on the file. I hope this helps. Good luck.

Re: To dither or not to dither that is the? Waves IDR?
 

Bryan,
This is a really interesting topic. I don't think I have it all correct in my head yet, but I'll write down what I think I know and let people who know more correct me. I hope you'll let me know if I'm not clear or have somethign wrong.

(I.e. - I don't mind looking stupid if I learn something...)

1) Going to 16-bit the way you do, I believe dithering would do nothing inside of SoundForge. If it was needed, it's too late. Dithering should be done at the step where you reduce bits/sample. (see #3)

2) Going to 44.1 the way you do also requires a sample rate conversion from 48KHz, which is being done by PTLE. This means that the sound of the bounced file will be somewhat different from the sound you heard while you mixed. You should check this conversion carefully with your ears. Maybe bounce the file to both 16/48 and 16/44.1 and then compare the two. (Or also bounce 24/48 and 24/44.1 and listen to all four version.) I hear differences on acoustic stuff I've recorded. (Please read through a thread I started 3-4 weeks ago about opinions concerning recording in 48KHz. It was interesting.)

3) Dithering is NOT truncation. In the dither, it is not as simple as taking the bottom 8 bits and just removing them. Dithering is a mathematical operation which involves bit-DEPTH conversion, where the 24-bit number is converted into the closest 16 bit number. There are issues with how to best do this that require the insertion of a 'noise' variable. I know the Waves Dither has options. I do not remember if the PTLE dither does...

4) If you're going to dither inside of Pro Tools, then do it on a Master Fader AND make sure the dither is the very LAST plugin. Plug-ins are post fader on the Master Fader. This is important when dithering, because you will likely have volume automation there, and the dither will do a correct conversion of the very final audio signal, which includes the volume automation.

5) Logically speaking, there is no difference between doing the dither in PTLE as above, or bouncing to 24/44.1 and then doing the dither in Sound Forge. I think your choice would be based on which Dither plugin sounds better to your ears.

6) You're way gives you an additional option of making a better mp3, based on 24/44.1, and then down converting for your CD. (You didn't say you wanted a CD, but I assume this is why you're going to 16/44.1.)

I'm using the UltraMAximizer as my last plug-in right now. The limiting feature is really amazing, but I'm still trying to understand some of the sounds in getting from it. I think in the end, it really comes down to your ears.

Anyway, please accept this as information from someone who's learning. Not guarenteed to be correct, but given freely.

Cheers,
Mark

Re: To dither or not to dither that is the? Waves IDR?
 

I missed it when he said he mixes to 48k. I would recommend against doing this if your final goal is audio CD. You may gain a little bit of fidelity at 48k but I think you actually lose more when you finally resample to 44.1k.

Mark is correct in saying that there is really no difference in using the UltraMaximizer when you mix or using it after the fact in Sound Forge. For me personally I find that I don't want any "mastering" type plugins on my master fader because it affects the way I mix. It's almost a crutch to me. I like to get the mixes sounding as good as possible without any of that stuff and then bounce it. That way, any extra EQing/Compression/Maximizing I do just makes my exisitng mix sound better. This is not necessarily the right or wrong way, it's just what works for me.

Mark, I wasn't sure if you were agreeing or disagreeing with me on your third point. Let me clarify anyway. When you run the IDR it will dither and noise shape down to 16 bits. It just doesn't throw away the last 8 bits for you automatically. The file is still 24 bit but the last 8 bits are only storing zeros. You then need to let Sound Forge truncate those empty 8 bits. In Sound Forge, when you right click on 24Bit and change it to 16Bit, it is the same thing as going to the Process menu, choosing Bit-Depth Converter and selecting "16Bit" with Dither set to "none" and noise shaping set to "Off". That is what truncates it. I hope this clears it up a little.

Re: To dither or not to dither that is the? Waves IDR?
 

bstaley,
Hi. No, I wasn't disagreeing with you and much as saying I wasn't understanding you completely. (I think I didn't read it enough times. I can understandd it now!) If the final dither output is a 24-bit data structure that puts dithered data in the upper two bytes (16-bits) and disregards what it puts in the lowest byte, then that's interesting, except that would make all your dithered data files 50% larger than they need to be. Also, my low end version of Sound Forge is only 16-bit, so I can't really look at this anyway. (Poor me!)

I mostly agree with the 48K comments. Read the other thread I did on this subject and I think most people did. There was one post, however, where the engineer (with a much larger budget than my home recording budget) talked about using an outboard hardware converter to do the sample rate conversion, and how everyone in his studio felt immediately that the sound was better than recording at 24/44.1. Unfortunately, I can't afford that comparison right now! [img]images/icons/wink.gif[/img]

And again, the final data type is important when choosing a sample rate to do a session. If you are NEVER going to make a CD, you might as well choose 48K.

Cheers,
Mark


<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:<HR>Originally posted by bstaley:
I missed it when he said he mixes to 48k. I would recommend against doing this if your final goal is audio CD. You may gain a little bit of fidelity at 48k but I think you actually lose more when you finally resample to 44.1k.

Mark is correct in saying that there is really no difference in using the UltraMaximizer when you mix or using it after the fact in Sound Forge. For me personally I find that I don't want any "mastering" type plugins on my master fader because it affects the way I mix. It's almost a crutch to me. I like to get the mixes sounding as good as possible without any of that stuff and then bounce it. That way, any extra EQing/Compression/Maximizing I do just makes my exisitng mix sound better. This is not necessarily the right or wrong way, it's just what works for me.

Mark, I wasn't sure if you were agreeing or disagreeing with me on your third point. Let me clarify anyway. When you run the IDR it will dither and noise shape down to 16 bits. It just doesn't throw away the last 8 bits for you automatically. The file is still 24 bit but the last 8 bits are only storing zeros. You then need to let Sound Forge truncate those empty 8 bits. In Sound Forge, when you right click on 24Bit and change it to 16Bit, it is the same thing as going to the Process menu, choosing Bit-Depth Converter and selecting "16Bit" with Dither set to "none" and noise shaping set to "Off". That is what truncates it. I hope this clears it up a little.<HR></BLOCKQUOTE>