|
|
Post by BradM on Sept 3, 2017 15:44:21 GMT -6
Dumb question... why are the DAW manufacturers not doing this automatically for us?
Brad
|
|
|
|
Post by Bob Olhsson on Sept 3, 2017 17:20:09 GMT -6
Good question! The only logical reason I can think of is that the overhead reduces the amount of other processing you can do. It's a non-issue with today's computers. I suspect a lot of them just throw public domain code libraries together with little or no understanding of the underlying math. I don't have the understanding but JJ who literally learned from folks who worked with Shannon and Nyquist has told me "don't even think about not using dither" and I'll take his word over most.
|
|
ericn
Temp
Balance Engineer
Posts: 16,107 
|
Post by ericn on Sept 3, 2017 21:38:06 GMT -6
Dumb question... why are the DAW manufacturers not doing this automatically for us? Brad At the low end al they hear is efficiency efficiency efficiency. Even today the bottom of the market speaks loudly! |
|
|
|
Post by Martin John Butler on Sept 3, 2017 22:00:39 GMT -6
I read up to page 4 and then jumped here. Now, someone please just tell me how to properly dither in Logic. When do I do it, when I'm done recording, done mixing, and I'm bouncing down?
I don't know if I will hear it or not, but even if I can't hear it, why not use it anyway, just in case.
|
|
|
|
Post by keymod on Sept 4, 2017 5:09:39 GMT -6
I read up to page 4 and then jumped here. Now, someone please just tell me how to properly dither in Logic. When do I do it, when I'm done recording, done mixing, and I'm bouncing down? I don't know if I will hear it or not, but even if I can't hear it, why not use it anyway, just in case. If someone "in the know" could post a sequence of events for doing this properly and efficiently, it would be useful. I would imagine it would be similar for most DAWs. And again, would someone please re-iterate why it's a good idea to do so? |
|
|
|
Post by jcoutu1 on Sept 4, 2017 5:25:34 GMT -6
What's going on is that the math is correct! I think truncation distortion and crappy analog stages are the main reasons analog frequently sounds better than digital. To be clear though, we can put 20 different plugins per channel, bus things around up the wazoo, add automation, etc. without having to worry about messing with any dither as long as we're still in the daw, right? It's only at the point that we're exiting the daw to go the da converters that we need to put on a dither plugin for every single output going to a separate converter channel, whether that be just two tracks for your mix bus, 16 channels for otb summing or outputs beings used for hardware inserts, right? And those dither plugins should be the absolute last thing in the chain for each output which is exiting the daw to go to a da converter channel, right? If all of this is correct, what am I still missing? Or is that it? And is there such a thing as too much dither? Does it hurt to err on the side of caution and add it in places where you're not sure if you need it, even if it turns out that you didn't need it? Hey Martin John Butler and keymod, here's the basic summary. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 4, 2017 7:01:13 GMT -6
It's easy really, it's mathematically correct (and often sounds better) to dither every time you reduce the word length, whether that be purely in the digital domain, or when going to an analogue output via DAC.
|
|
|
|
Post by swurveman on Sept 4, 2017 7:13:25 GMT -6
To be clear though, we can put 20 different plugins per channel, bus things around up the wazoo, add automation, etc. without having to worry about messing with any dither as long as we're still in the daw, right? It's only at the point that we're exiting the daw to go the da converters that we need to put on a dither plugin for every single output going to a separate converter channel, whether that be just two tracks for your mix bus, 16 channels for otb summing or outputs beings used for hardware inserts, right? And those dither plugins should be the absolute last thing in the chain for each output which is exiting the daw to go to a da converter channel, right? If all of this is correct, what am I still missing? Or is that it? And is there such a thing as too much dither? Does it hurt to err on the side of caution and add it in places where you're not sure if you need it, even if it turns out that you didn't need it? Hey Martin John Butler and keymod , here's the basic summary. But doesn't this presume that the DAW Mixer channel output is to the hardware? This doesn't pertain to people that are using " External FX" insert plugins in the channel, does it? Those "External FX" plugins have a seperate input/output bus structure. The DAW Mixer's individual channel bus structure outputs to the main stereo channel which outputs d/a to whatever hardware outputs the user sets up, or a DAW Mixer Group Channel (drums/guitars etc) which then ouputs to the main output channel. So, I don't see how just putting dither at the end of a DAW Mixer channel gets in front of the hardware da/ad loop for people using hardware External FX insert plugins. If I'm misunderstanding this I would really appreciate someone pointing out my error. |
|
|
|
Post by stormymondays on Sept 4, 2017 8:55:23 GMT -6
The summary is: put a dither plugin set to 24-but dither before any I/O plugin (sending audio to external hardware) and last in the master buss. Done.
|
|
|
|
Post by swurveman on Sept 4, 2017 9:07:24 GMT -6
The summary is: put a dither plugin set to 24-but dither before any I/O plugin (sending audio to external hardware) and last in the master buss. Done. Yes, I believe this is correct. I think inserts work like daisy chained guitar stomp boxes. So, if you want the dither plugin to effect the hardware insert plugin it must be before the hardware insert. |
|
|
|
Post by jin167 on Sept 4, 2017 9:11:32 GMT -6
interesting comments. make your own decision.
|
|
|
|
Post by Martin John Butler on Sept 4, 2017 9:12:30 GMT -6
I checked, and I have the Waves IDR360 and I think there's a dither option in the L3 and L3 mastering limiter. Does it matter which of those I use, any particular options?I think thee's an optional dither setting in Logic, but I've always bypassed it. Should I use that or the Waves?
|
|
|
|
Post by stormymondays on Sept 4, 2017 9:14:33 GMT -6
The summary is: put a dither plugin set to 24-but dither before any I/O plugin (sending audio to external hardware) and last in the master buss. Done. Yes, I believe this is correct. I think inserts work like daisy chained guitar stomp boxes. So, if you want the dither plugin to effect the hardware insert plugin it must be before the hardware insert. Absolutely. You can check it with the Bitter bitscope plugin. |
|
|
|
Post by stormymondays on Sept 4, 2017 9:16:09 GMT -6
I checked, and I have the Waves IDR360 and I think there's a dither option in the L3 and L3 mastering limiter. Does it matter which of those I use, any particular options?I think thee's an optional dither setting in Logic, but I've always bypassed it. Should I use that or the Waves? I'd advise to get the aforementioned Good Dither plugin ($19). Logic Pro has no dither plugins. I think (from memory), L3 has 16-bit dither, and we're talking about 24-bit dither. |
|
|
|
Post by stormymondays on Sept 4, 2017 9:23:57 GMT -6
interesting comments. make your own decision. I like this video, VERY interesting! My take is: We can make a choice of a simpler workflow (no dither) and create inaudible truncation distortion artifacts. Or we can choose a slightly more complicated workflow (insert an extra plugin), and create inaudible dither noise. I had chosen the former, and after this thread I've decided to go for the latter now. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 4, 2017 9:47:21 GMT -6
I checked, and I have the Waves IDR360 and I think there's a dither option in the L3 and L3 mastering limiter. Does it matter which of those I use, any particular options?I think thee's an optional dither setting in Logic, but I've always bypassed it. Should I use that or the Waves? Stick with plain old TPDF at 100% and no noise shaping, if you have that option. I heard (probably from Bob O) it's the least likely to cause any problems down the line if further changes are made to the digital word length or file (lossy encoding etc.) |
|
|
|
Post by matt on Sept 4, 2017 10:10:53 GMT -6
To summarize my case: I am on Pro Tools (32 bit float for internal processing, and 32 float at all output buses according to Bitter) and going out to an Apogee Symphony Mk1 via Thunderbridge. Apogee touts the ESS Sabre 32 DAC as the latest-and-greatest in 32 bit processing. The real question, then, is how the Symphony is processing the incoming digital audio stream:
- If 32 floating point is preserved at the DAC input, no dither is needed (and would be mildly detrimental due to added noise)
- If 32 bit float is changed/interpolated/modified/truncated to 32 bit fixed by the DAC, then I should dither to 32 bit using Goodhertz's dither plug, which does 32 bit dithering (it's the only one I've seen so far that has a 32 bit setting)
- For me, dithering to 24 bit is not the thing to do. Or is it? Enquiring minds want to know!
|
|
|
|
Post by drbill on Sept 4, 2017 10:25:34 GMT -6
Just a quick anecdotal question :
Has anyone ever HEARD noise from dither in normal mixing conditions? I've got sometimes up to 80 channels of inserts going, and I have yet to hear any noise from dither. I'm not saying it can't happen, I just am amused at people who are concerned with dither noise, but would give their left nut for an MCI or Studer 16T machine or an 8068. LOL I'm sure under microscopic conditions one might be able to scientifically hear it......
I would love to hear from someone who could say "the noise from dither wrecked my mix".
|
|
|
|
Post by matt on Sept 4, 2017 10:53:38 GMT -6
Just a quick anecdotal question : Has anyone ever HEARD noise from dither in normal mixing conditions? I'm wondering this too! The noise floor at my 2-bus ADC is at around -80db and concentrated where you might expect, at above 10khz (damn compressor hiss!). So for me, the whole dither conversation is not about reducing distortion artifacts, whether it be hiss, or the zipper effect, or any other unwelcome audio oddity. Rather, it's about whether dithering bus outputs has an audible impact to the depth/width/clarity of a mix. I guess I'll just have to try it and see. The main problem with this topic is that it appeals to my inner geek. Instead of recording, I'm studying dither. It's always something! |
|
|
|
Post by drbill on Sept 4, 2017 11:09:56 GMT -6
I don't want to answer my own question, but even with my personal use and longstanding interest in this question, I have never heard any anecdotal or scientific evidence that dither can mess up the noise floor. I'm not saying it couldn't happen, but I have yet to hear or read about it. Then of course, the question is "why wouldn't you dither" if it's the correct mathematical and engineering thing to do?
|
|
|
|
Post by ragan on Sept 4, 2017 11:36:28 GMT -6
I thought one of the skeptical side points against the need to dither in this case was that it's all taking place well below the noise floor of any of the gear involved and thus wouldn't be audible.
Note I'm not making that point, cause I don't know what the hell I'm talking about but I thought it was an issue someone raised?
|
|
|
|
Post by drbill on Sept 4, 2017 11:37:59 GMT -6
If it's inaudible, and if it's mathematically "correct" to do it, then why not?
|
|
|
|
Post by ragan on Sept 4, 2017 12:10:13 GMT -6
If it's inaudible, and if it's mathematically "correct" to do it, then why not? Right, I'm not saying not to. I was just wondering if anyone had expressed concern about dither "noise". I thought part of the whole deal was that it's possibly below the noise floor of everything else anyway, so definitely not a noise concern. |
|
|
|
Post by Bob Olhsson on Sept 4, 2017 14:08:20 GMT -6
Again, the distortion is going to be more audible than properly designed dither will be!
|
|
|
|
Post by EmRR on Sept 4, 2017 14:57:29 GMT -6
Just a quick anecdotal question : Has anyone ever HEARD noise from dither in normal mixing conditions? I've got sometimes up to 80 channels of inserts going, and I have yet to hear any noise from dither. I'm not saying it can't happen, I just am amused at people who are concerned with dither noise, but would give their left nut for an MCI or Studer 16T machine or an 8068. LOL I'm sure under microscopic conditions one might be able to scientifically hear it...... I would love to hear from someone who could say "the noise from dither wrecked my mix". Not to my knowledge. If you use noise shaped dither at low bit depths (not what we're talking about here) I can hear it as opposed to not when it's 'flat', but that's it. You might hear noise shaped versus not at standard rates under discerning listening conditions, or you might not, which is why we want to stay away from noise shaping. Maybe I'll get to some charts and graphs soon, maybe not..... |
|
