Community Forums Archive

Go Back

Subject:Bus FX post Volume?
Posted by: BradlyMusic
Date:6/11/2006 7:33:01 AM

How do you get FX's that you insert into a Bus to be POST bus Volume slider? I noticed that insert FX's are all Pre Volume on the buses. Well shouldn't Dither be the last step in a processing chain? How do I insert the Dither plugin, so that dither is the last step before the audio signal gets sent to my sound cards output?

Message last edited on6/11/2006 7:33:57 AM byBradlyMusic.
Subject:RE: Bus FX post Volume?
Reply by: BradlyMusic
Date:6/11/2006 8:48:48 AM

Also, How do I assign multiple tracks to a bus with a compressor on the bus and have the compressor threshold follow volume adjustments? Are we not suppose to mix in Acid?

Subject:RE: Bus FX post Volume?
Reply by: vespesian
Date:6/11/2006 9:06:57 AM

Route that bus to a new bus with EQ and the compressor on it (?)...

Subject:RE: Bus FX post Volume?
Reply by: JohnnyRoy
Date:6/11/2006 9:11:22 AM

> How do you get FX's that you insert into a Bus to be POST bus Volume slider?

Normally, you would right-click on the send label you get a popup menu to select Pre Volume / Post Volume. You would then select Post Volume. Unfortunately, this appears to be a bug in ACID Pro 6.0a because selecting Post Volume doesn’t appear to work. :( It works fine in ACID Pro 5 so I assume Sony will fix this in an update.

~jr

Subject:RE: Bus FX post Volume?
Reply by: BradlyMusic
Date:6/11/2006 9:43:46 AM

Thanks Vespian, that's a good work around for the compressor problem....but there seems to be no solution for adding a Dither plugin. Dither should be the last thing in the processing chain before the audio goes to the outside world, but there is no way that I can find to do this.

It looks like another bug in Acid.

I'm a little confused also. Why is it when you create a bus that there is a pan adjustment on the bus, but then if you assign that bus to one of your sound cards outputs the pan disappears and you get 2 seperate volume adjustments instead?

Subject:RE: Bus FX post Volume?
Reply by: H2000
Date:6/11/2006 11:27:25 AM

"....but there seems to be no solution for adding a Dither plugin. Dither should be the last thing in the processing chain before the audio goes to the outside world, but there is no way that I can find to do this.

It looks like another bug in Acid."

It's not really considered a bug if it's not designed that way.
I use a dither plugin on a bus, but then I set to the output fader at 0. This is effectively doing nothing (no additional processing). But, even if I needed to set this fader lower I wouldn't worry too much about it. I doubt having one fader post dither would degredate the audio that much. But, I haven't really tried it. Have you done a listening test?

Subject:RE: Bus FX post Volume?
Reply by: BradlyMusic
Date:6/11/2006 2:27:35 PM

Well, what I'll do is do some rendering with and without dither and see how they compare as far as bit reduction distortion is concerned. That being said, this is like DAW 101 stuff where you should dither anytime you are doing a bit reduction as the very last step in a process chain. Considering we're working with 24bit files and/or working in 32 bit float and then we do a bit reduction down to 16bit, then the next following step is a 32 bit float volume fader then back to a sound card D/A bit reduction, it's just not what is expected from an app with the word "Pro" slapped on to the end of it, if you get my drift.

It seems every DAW developer knows this besides Sony, so I'm hoping this is not designed this way. Sorry for the blantant stab, but I would like to be proven wrong if this is not the case of how Acid's mixer routing has been establised.

Message last edited on6/11/2006 2:35:30 PM byBradlyMusic.
Subject:RE: Bus FX post Volume?
Reply by: JohnnyRoy
Date:6/11/2006 5:57:57 PM

> Dither should be the last thing in the processing chain before the audio goes to the outside world, but there is no way that I can find to do this.

Why don’t you just add the Dither as the last FX on the bus? This would make it the last thing if the bus it routed directly to hardware.

~jr

Subject:RE: Bus FX post Volume?
Reply by: BradlyMusic
Date:6/11/2006 8:55:55 PM

The problem is the FX is PRE Fader.....which means that following the FX is a volume Fader, which would mean that all fader adjustments are done in the 32 bit float, which means I just applied dither and then a bit reduction and then the following post FX fader just transformed it back to 32 bit float and then it does a truncation when the 32 bit float gets sent to the D/A converter, which means it defeats the entire purpose of adding dither in the first place.

Here's a good article which simplifies the entire purpose of dither, which is explained very well by Nika Aldrich.

http://www.users.qwest.net/~volt42/cadenzarecording/DitherExplained.pdf

I suggest reading it and we can then be on the same page, where hopefully you'll come to understand the flaw in the Acid mixer signal flow.

Message last edited on6/11/2006 8:57:08 PM byBradlyMusic.
Subject:RE: Bus FX post Volume?
Reply by: H2000
Date:6/11/2006 8:59:50 PM

The truncation occurs after the bus volume fader.

There is no bit reduction, then back to 32bit, then reduced again. There is no need for any bit reduction plugins - just dither. The bit reduction will occur as the very last step (after the bus fader) and will be whatever your soundcard bit depth is set to.

Message last edited on6/11/2006 9:00:34 PM byH2000.
Subject:RE: Bus FX post Volume?
Reply by: BradlyMusic
Date:6/11/2006 9:47:19 PM

Ok, that's a good explaination......but.....Let's say you have a 24 bit sound card. You have a digital out connected to an external CD recorder or DAT player which is 16 bit. Now what happens in your example?

If that was the case, then I have to ask, why does the Sony Dither plugin have a "Quantizaton depth" adjustment in it?

You very well might be correct in all of this, because I don't know the full routing of Acid yet. The only thing I have to go on is other Daws that I have used they all have the ability to put dither as the last process and they recommend that Dither being the last process when going from a higher bit dept to a lower one. If someone from Sony could chime in then that would be great. Otherwise I don't feel very comfortable the way it is and I'll have to check it out and see for myself to see if there are any unnecessary distortion side effects by not having dither as the last process. Right now I just don't feel very comfortable the way it is and what I've come to expect.

Message last edited on6/12/2006 1:53:57 AM byBradlyMusic.
Subject:RE: Bus FX post Volume?
Reply by: H2000
Date:6/14/2006 7:02:28 AM

AFAIK, you typically will want to use dither in "fixed point" bit reductions - like 24bit to 16bit. Also, in fixed bit systems like ProTools where there are many "fixed point" bit recuctions from 48bit to 24bit it is advised to dither. But, you wouldn't dither when going from a "floating point" to a fixed point (even when there is a bit reduction).

In your question of going from a 24 bit output to a 16bit CD burner, it would be advisable to dither there. One thing to do is to set the sound card (and Acid properties) to output 16bit. This will negate the need for dither. It doesn't make sense to send 24bits out of your soundcard and then truncate it once it hits your CD burner.

When I burn a quick listening CD from within Acid in a project that is 24bits, I will sometimes add a dither plugin to the master bus because I assume it is summing to 24bit and then truncating in the burn process. (This may or may not be the way it's done, but it makes sense.) Then you have a master bus fader after the dither, yes. But, like I said, if it is set to zero then it does no math and no additional processing.

Subject:RE: Bus FX post Volume?
Reply by: BradlyMusic
Date:6/14/2006 9:02:15 AM

"But, you wouldn't dither when going from a "floating point" to a fixed point (even when there is a bit reduction). "

Sorry, that's not correct information in my viewpoint. Please read the link I posted from Nika Aldrich, which goes into good detail of this. There was a lengthy discussion on this type of stuff in the Sonar forums, where Nika actually participated in the discussion and it was very informative. I don't know where it is now off the top of my head, but everything you just said, pretty much goes against everything Nika had said in that discussion.

"But, like I said, if it is set to zero then it does no math and no additional processing. "

That doesn't seem like correct information to me either, but let's just say you are correct? Doesn't this defeat the entire purpose of having a fader on the buses? Why don't we just elliminate them?

FYI, when the fader is set at 0dB it just multiplys the incoming data by ONE.....so yes it is doing something, and it is doing that something in floating point math since all calculations are done in 32 bit float in Acid.

Please read this, so you can better understand the purpose of dither and when it should be done.
http://www.users.qwest.net/~volt42/cadenzarecording/DitherExplained.pdf



Message last edited on6/14/2006 9:45:21 AM byBradlyMusic.
Subject:RE: Bus FX post Volume?
Reply by: H2000
Date:6/14/2006 11:50:33 AM

""But, you wouldn't dither when going from a "floating point" to a fixed point (even when there is a bit reduction). "

Sorry, that's not correct information in my viewpoint. Please read the link I posted from Nika Aldrich, which goes into good detail of this. There was a lengthy discussion on this type of stuff in the Sonar forums, where Nika actually participated in the discussion and it was very informative. I don't know where it is now off the top of my head, but everything you just said, pretty much goes against everything Nika had said in that discussion."

OK, maybe I generalized a little. But, there are many, many papers out there which discuss the problems associated with dither in a floating bit system. I've read some that say you should, some that say you shouldn't, some say you should but only with non-shaped dither. This particular paper of Nika's does not address these issues.

I will say that if you are going to 16bit then I do believe that you should do some dithering (even though my previous statement appears to say not to). The quantization errors going from a 32bit float to 24 bit should be inaudible. The quantization errors from going to 24bit to 16bit are much higher (48db higher) and therefore will be audible through the noise floor of your system.

My solution is to do no dithering when mixing down to 24bit master. Then I take that and apply dither and convert to 16bit for CD burning. I understand your case and for you to burn to an external CD at 16bit, you would want to dither (as I said before). I understand your concern regarding the master fader and since I don't work for Sony, I can't address why it's that way. That's why for critical stuff, I render at 24bit and then dither and convert that file.


""But, like I said, if it is set to zero then it does no math and no additional processing. "

That doesn't seem like correct information to me either, but let's just say you are correct? Doesn't this defeat the entire purpose of having a fader on the buses? Why don't we just elliminate them?"

I see your point, but I'm just trying to offer you a possible best case workaround.

"FYI, when the fader is set at 0dB it just multiplys the incoming data by ONE.....so yes it is doing something, and it is doing that something in floating point math since all calculations are done in 32 bit float in Acid."

Yes, well I don't know if they do the multiply by one or not. You would have to ask the coders. They could have simply said: if it's zero, then do nothing. But, in any case a mulitply by one does not change the wordlength and therefore should not require dithering after it. I'm not sure what your point is that it is at 32bit float because your dither plugin is only adding noise at a 24bit or 16bit level. The truncation to the final bit depth happens at the end, after dither and after the master fader.

"Please read this, so you can better understand the purpose of dither and when it should be done.
http://www.users.qwest.net/~volt42/cadenzarecording/DitherExplained.pdf"

I read it and I have read many papers on dithering. This is one of the simpler papers which is a good primer.

My interest is in getting a high quality result in an effecient way. I understand your concerns and I tried to suggest some things to help. Acid is a great program for what it is. It has shortcomings like all programs, and it's not the only program I use. It's definately not the end all be all for me, but I like what it does.

Subject:RE: Bus FX post Volume?
Reply by: BradlyMusic
Date:6/14/2006 1:38:46 PM

Ok, I've done my research and the results are. It looks like you are correct for the 0dB setting, in that this will give you similar results since it is multiplying by a nice even number, which will not therefore introduce further round off error.

I've installed the Vegas 6 demo for my other recent monitoring distortion problem and I see there is an option to insert FX's post fader or Pre Fader on the buses. I set up a test similar to Nika's. I created a 24 bit Sinewave at a max amplitude of -6dB. Within Acid I placed a triangular dither with 16 bit reduction and the master fader was set to 0dB and then rendered to a 16 bit file format. I confirmed that this gave me the same results if I had done the same thing in Vegas with the dither set to post fader. So long story short, what i first found is that Vegas gave me the exact same results as Acid when I changed the variables to be the same as in Acid. So to simplify the results of this study, I did the comparison in Vegas after I verified that the results where pretty much identical when working in Acid.

The test setup steps went like this:
1. Set Vegas to work at 24 bit /44.1Khz resolution.
2. Import a 24/44.1, 100Hz sinewave at -6dB peak amplitude
3. Set the master fader to -.1dB
4. Insert the dither plugin PRE fader with Triangular shape dither, and 16 bit depth reduction.
5. Render the mix to a 16 bit mono .Wav file
6. Move the dither plugin to POST Fader.
7. Render the mix to a 16 bit mono .Wav file
8. Import both .Wav files into Sound Forge and take a snapshot of each files S.A. so you can compare them.

The results:
1. If the master fader is set to 0dB, then this will give you the same results as having dither placed post fader as I'm asking for.

2. If you move the master fader down to any other value besides 0dB, then this will have an impact on the overall noise floor level and nullify the gains you get by adding dither to your mix.

Here's a summary graph from the Sound Forge SA, that shows the impact of having the dither located PRE master volume fader vs POST Master volume. What you see is that the RED graph is where the dither plugin was located PRE fader as it is in ACID. The BLUE graph is where the dither was moved to POST fader......thus being the last process in the chain before rendering as I'm trying to accomplish in Acid.

The result is that the Noise floor for the dither being located PRE master fader is +10dB higher than that of the dither which is located POST master fader. An increase in noise floor of +10dB is pretty substantial considering you LOWERED the master fader by -.1dB wouldn't you think? If this was the way the Acid's mixer is designed, then this is a pretty obvious design flaw, if you ask me. Like I said before, this is like DAW 101 stuff and I would have thought Sony would have known better. I am highly disappointed and am now wondering what other design flaws are in the near future for me.

Graph results:
http://www.stashbox.org/uploads/1150314771/Dither-Pre-vs-Post-fader.jpg

Message last edited on6/14/2006 1:42:56 PM byBradlyMusic.
Subject:RE: Bus FX post Volume?
Reply by: H2000
Date:6/14/2006 5:18:17 PM

Sounds like a good informative test.

I've always used Vegas for all my final mixing, and I use Acid to develop and sketch song ideas. So, I was hoping Acid would incorporate more of what Vegas offers (and then some). And, perhaps it will someday. But, for now it is what it is. I still find the version6 update useful for the price, but I will still use Vegas or Reaper for my final mixes.

Subject:RE: Bus FX post Volume?
Reply by: pwppch
Date:6/14/2006 5:23:42 PM

The post fader insert will be addressed in a future version of ACID.

The simplest work around is:

- Create a sub bus
- rout it to the master bus and all tracks or sub buses to it.
- Use it as your project master bus .
- Add the Dither to your master bus and set the gain to 0 db.

Peter



Subject:RE: Bus FX post Volume?
Reply by: BradlyMusic
Date:6/14/2006 6:33:31 PM

Thanks for the reply, that work around is pretty much what I surmized after my investigation. I think the suggestion found in this thread sounds like a simpler and more practical one:
http://www.sonymediasoftware.com/forums/ShowMessage.asp?MessageID=464275&Replies=25
That being since I have Sound Forge 8.0 installed with the Volume adjustment plugin slider.

Any chance that we could see those Sound Forge Volume and Pan plugins magically appear in an Acid 6 update?

Subject:RE: Bus FX post Volume?
Reply by: jumbuk
Date:6/14/2006 6:41:07 PM

Just to add my 2 cents:

I normally don't use dither in Acid or Cubase. I don't know what my final fade-outs/ins will be like until I assemble a track list and do my final mastering of all tracks together. That's when I do true "mastering" EQ, compression etc, and adjustment of overall relative track levels, time between tracks, fade-out/in etc. I typically do all of this in CD Architect, and dither is the last plugin applied.

My reasoning is there is no point dithering and then later making an adjustment that would affect low-level gain settings, and that would actually change the already-dithered data.

Having said all that, you have now set me thinking - do I really understand the setup of CD Architect? The answer is no, and I need to do some research.

As an aside, this is all kind of irrelevant for 90% of music makers on this forum (my guess). I bet most of the music made here is compressed to the point where there are few low-level dynamics to worry about dithering anyway! In my case, I do a lot of ambient and folk-related music, particulalrly wind instruments, where the dynamic range is wide and there are long-tailed envelopes that trail off to nothing. Dither is important to me!

Subject:RE: Bus FX post Volume?
Reply by: BradlyMusic
Date:6/14/2006 7:54:37 PM

Jumbuk,
The thing you are overlooking is that not applying dither does not only effect low level fade out material. The problem is that anytime you do a bit reduction, there are harmonic distortions that get added to your audio signal. Yes, even when working in Acid which has a 32 float audio engine and then rendering to a 24 bit file, since there is a bit reduction required. You would have to eccentially render at 32 bit float to avoid these distortions, which is possible in Acid and probably the best work around.

So further points of what you described in your workflow is that by not applying dither you are actually increasing the level of these harmonic distortions that get added due to the quantization error......then when you go to your mastering process and start using compression like you mentioned 90% of the Acid users probably do, then you are further moving this increased noise floor up closer to your audio signal where it becomes more of a factor.

FYI, the Master bus insert in CD Architect is Post Fader.

Message last edited on6/15/2006 12:05:25 AM byBradlyMusic.
Subject:RE: Bus FX post Volume?
Reply by: jumbuk
Date:6/15/2006 6:50:36 AM

BradlyMusic,

Well, you are right. I do render to 24 bit, and technically this does introduce harmonic distortion due to the bit reduction. I doubt if I can hear the distortion, but I agree with the principle that if you can eliminate a potential source of distortion by a workflow choice, you should do so.

Having said that, using dither alters the audio, and the theoretical cleanest way would be to render in 32 bit even if the Acid master FX chain was post fader, and then apply dither once as the last mastering process (I think!).

Subject:RE: Bus FX post Volume?
Reply by: Ben 
Date:6/15/2006 7:19:09 AM

>> The post fader insert will be addressed in a future version of ACID. <<

Peter, while you're fixing that, please also consider the pan to busses/FX issue in both ACID and Vegas, and also consider implementing sends from busses - which is very much needed.

Thanks

Ben


Subject:RE: Bus FX post Volume?
Reply by: pwppch
Date:6/15/2006 4:30:21 PM

Ben,

As I stated previously, we will consider these for future versions of ACID (and Vegas.) This is very unlikely to happen for an update to ACID 6.

Peter

Subject:RE: Bus FX post Volume?
Reply by: H2000
Date:6/15/2006 4:56:29 PM

"As I stated previously, we will consider these for future versions of ACID (and Vegas.) This is very unlikely to happen for an update to ACID 6."

I understand about the sends from busses not being seen in 6. But, "post volume" panning to a bus should work (in Acid) according to the current Acid6 user manual. The fact that it doesn't is clearly a bug or the manual is printed incorrectly. Can you tell us which it is?

Go Back