Digital Boards - Sound Quality
- Thread starter Chuck Simon
- Start date
Re: Digital Boards - Sound Quality
I don't know about Smaart or TEF, but a distortion analysis of an input signal vs an output signal should reveal any jitter or other nasty things happening to the signal. I suppose the trick is determining what components of the measured distortion are "desirable sweetening" vs "distortion".
Since this thread is in the "Junior" section, I'll go on a limb and say that the distortion caused by the transducers at each end of the chain and the venue that the system is being used in, is likely at least a couple orders of magnitude larger than whatever distortion is caused by the mixer. Relative to all the other issues of live sound, it seems a pipe dream to be at the level that everything else about the gig - equipment arriving safely, loading issues, setup happening in enough time, having enough power, troubleshooting, sound check, venue acoustic issues, band talent issues, etc. are so well satisfied that worrying about the mixer clock is a significant concern.
I don't know about Smaart or TEF, but a distortion analysis of an input signal vs an output signal should reveal any jitter or other nasty things happening to the signal. I suppose the trick is determining what components of the measured distortion are "desirable sweetening" vs "distortion".
Since this thread is in the "Junior" section, I'll go on a limb and say that the distortion caused by the transducers at each end of the chain and the venue that the system is being used in, is likely at least a couple orders of magnitude larger than whatever distortion is caused by the mixer. Relative to all the other issues of live sound, it seems a pipe dream to be at the level that everything else about the gig - equipment arriving safely, loading issues, setup happening in enough time, having enough power, troubleshooting, sound check, venue acoustic issues, band talent issues, etc. are so well satisfied that worrying about the mixer clock is a significant concern.
Last edited:
Re: Digital Boards - Sound Quality
If you can hear it, you should be able to document the change in AUDIO via FFT. JR likes the null test, and it would be another way to do this.
If you can hear it, you should be able to document the change in AUDIO via FFT. JR likes the null test, and it would be another way to do this.
Re: Digital Boards - Sound Quality
Hi Tim,
I've not seen any conclusive data either, but there is certainly a difference that I have found to be noticeable in terms of difficult to quantify terms such as space and depth, smoothness, etc.
In the recording world, some folks have done some rudimentary measurement which has shown there to be no significant difference in frequency response, while at the same time acknowledging that there was a very perceptible difference in sound.
From elsewhere:
"The soundstage seemed wider, but not in a “phase-y” way. At the same time, the center of the image—the picture of the guitar emanating from the speakers—seemed more solid or focused in the middle, between the speakers. I know this sounds like “hi-fi” talk, but it was noticeable and repeatable. The frequency response seemed more extended from bottom to top, even though I knew the response of both converters was ruler-flat. It sounded clearer and smoother, especially in the upper midrange and high frequencies. The natural decay and low-level information seemed more apparent, giving the sound more depth from the front to the back of the speakers "
I don't doubt it must be measureable, but suspect that frequency and phase response a la a traditional Smaart measurement is not the right target to examine in this case.
Hi Tim,
I've not seen any conclusive data either, but there is certainly a difference that I have found to be noticeable in terms of difficult to quantify terms such as space and depth, smoothness, etc.
In the recording world, some folks have done some rudimentary measurement which has shown there to be no significant difference in frequency response, while at the same time acknowledging that there was a very perceptible difference in sound.
From elsewhere:
"The soundstage seemed wider, but not in a “phase-y” way. At the same time, the center of the image—the picture of the guitar emanating from the speakers—seemed more solid or focused in the middle, between the speakers. I know this sounds like “hi-fi” talk, but it was noticeable and repeatable. The frequency response seemed more extended from bottom to top, even though I knew the response of both converters was ruler-flat. It sounded clearer and smoother, especially in the upper midrange and high frequencies. The natural decay and low-level information seemed more apparent, giving the sound more depth from the front to the back of the speakers "
I don't doubt it must be measureable, but suspect that frequency and phase response a la a traditional Smaart measurement is not the right target to examine in this case.
Last edited:
Re: Digital Boards - Sound Quality
Only if the effect is linear time invariant, which I don't believe it to be.
Only if the effect is linear time invariant, which I don't believe it to be.
Re: Digital Boards - Sound Quality
I make a point of not arguing about what other people say they hear, and generally if someone claims a "massive" improvement, that suggests that something measurable happened.
There has been a lot of investigation into this phenomenon and I am not aware of a smoking gun, identified in bench testing, while subjective anecdotal claims persist.
Yes, jitter or clock uncertainty would reveal as a low level noise/distortion, so THD or linearity would be the obvious bench test. The subjective claims suggest that some very fine (small) time resolution uncertainties impact stereo sound field localization. Our brain often uses first arrival to localize so hypothetically, competing arrival times in a stereo sound field could alter perception, but jitter occurs to the clock so impacts the digital data on a per bit basis**. Per word would get some average jitter applied, and per cycle of a waveform or per transient, would average this out further over the multiplicity of samples it takes to capture a single transient event.
I try to be open minded about such things but the evidence for this being a "massive" phenomenon is not well supported. I am not even aware of successful blind testing, while such a success may reveal more about the shortcomings of a particular hardware platform than a general digital audio phenomenon.
This has been batted around in the press for years, if there was some there there, this would be well characterized by now and manufacturers would have stepped up.
...but the customer is always right so do what makes you happy. I remain skeptical and unconvinced.
JR
** this is actually a little more complex than that with actual modern A/D and D/A conversion technology (oversampling etc). Logically it seems this (jitter) would matter most at both conversions.
I make a point of not arguing about what other people say they hear, and generally if someone claims a "massive" improvement, that suggests that something measurable happened.
There has been a lot of investigation into this phenomenon and I am not aware of a smoking gun, identified in bench testing, while subjective anecdotal claims persist.
Yes, jitter or clock uncertainty would reveal as a low level noise/distortion, so THD or linearity would be the obvious bench test. The subjective claims suggest that some very fine (small) time resolution uncertainties impact stereo sound field localization. Our brain often uses first arrival to localize so hypothetically, competing arrival times in a stereo sound field could alter perception, but jitter occurs to the clock so impacts the digital data on a per bit basis**. Per word would get some average jitter applied, and per cycle of a waveform or per transient, would average this out further over the multiplicity of samples it takes to capture a single transient event.
I try to be open minded about such things but the evidence for this being a "massive" phenomenon is not well supported. I am not even aware of successful blind testing, while such a success may reveal more about the shortcomings of a particular hardware platform than a general digital audio phenomenon.
This has been batted around in the press for years, if there was some there there, this would be well characterized by now and manufacturers would have stepped up.
...but the customer is always right so do what makes you happy. I remain skeptical and unconvinced.
JR
** this is actually a little more complex than that with actual modern A/D and D/A conversion technology (oversampling etc). Logically it seems this (jitter) would matter most at both conversions.
Last edited:
Re: Digital Boards - Sound Quality
JR, in the studio world manufacturers are making a big deal out of this now. Even my low-midrange Focusrite Liquid Saffire 56 is marketed as having a vastly improved clock, let alone companies like Black Lion doing customizations to Avid gear to make improvements on that front. The Black Lion 192 interface is probably one of the best respected interfaces in the recording world at the moment, and it is based on the premise of modding Avid hardware primarily with respect to delta sigma converter and master clock design. From their site: "We start with an advanced master clock that has around 2 picoseconds of jitter and -130dBc/Hz of phase noise @ 10Hz. "
Of course that box does more than just clock improvements, however the Big Ben discussed previously does yield notable differences as well.
Of course, regardless of how many people say it makes a difference (though not an enormous one compared to differences in bad vs good speakers) shouldn't matter to you.
I am not into the whole "hifi audiofool" mentality, and tend to approach such things with skepticism.
This is one of those cases where you simply need to hear it for yourself. You WILL notice a difference, but I'll never be able to convince you based on some random forum post.
JR, in the studio world manufacturers are making a big deal out of this now. Even my low-midrange Focusrite Liquid Saffire 56 is marketed as having a vastly improved clock, let alone companies like Black Lion doing customizations to Avid gear to make improvements on that front. The Black Lion 192 interface is probably one of the best respected interfaces in the recording world at the moment, and it is based on the premise of modding Avid hardware primarily with respect to delta sigma converter and master clock design. From their site: "We start with an advanced master clock that has around 2 picoseconds of jitter and -130dBc/Hz of phase noise @ 10Hz. "
Of course that box does more than just clock improvements, however the Big Ben discussed previously does yield notable differences as well.
Of course, regardless of how many people say it makes a difference (though not an enormous one compared to differences in bad vs good speakers) shouldn't matter to you.
I am not into the whole "hifi audiofool" mentality, and tend to approach such things with skepticism.
This is one of those cases where you simply need to hear it for yourself. You WILL notice a difference, but I'll never be able to convince you based on some random forum post.
Re: Digital Boards - Sound Quality
And that's why I also suggested that we could be measuring the wrong things or going about it in wrong way to elicit physics-based evidence of a perceptual experience.
Whatever ain't physics, is psychology.
And that's why I also suggested that we could be measuring the wrong things or going about it in wrong way to elicit physics-based evidence of a perceptual experience.
Whatever ain't physics, is psychology.
Re: Digital Boards - Sound Quality
Jason,
You wouldn't, because the sampling rates are too slow to see the jitter on smaart/TEF. You could easily see it on a wideband sampling scope. Gaussian jitter is straightforward to observe.
Regardless, the external clocks are not an improvement to jitter, because the phase lock loop that they sync to in the console have worse jitter specifications by a significant margin over using the internal clocking.
I'm not saying that the external clock doesn't make the console sound better, as I have never had a situation of critical listening to test it out in, but if the external clock did create an improvement, less jitter is not the explanation for it.
Jason,
You wouldn't, because the sampling rates are too slow to see the jitter on smaart/TEF. You could easily see it on a wideband sampling scope. Gaussian jitter is straightforward to observe.
Regardless, the external clocks are not an improvement to jitter, because the phase lock loop that they sync to in the console have worse jitter specifications by a significant margin over using the internal clocking.
I'm not saying that the external clock doesn't make the console sound better, as I have never had a situation of critical listening to test it out in, but if the external clock did create an improvement, less jitter is not the explanation for it.
Re: Digital Boards - Sound Quality
Here's something to ponder. Back in the earlier days of digital boards there were offerings of digital compressors, digital gates and a few other digital rack mount tricks other than FX units, many were maybe still are offered from well respected companies XTA and Drawmer come to mind. They were greeted with a cool maybe somewhat skeptical welcome. Put those digitized units into a digital board and there welcomed with open arms. Way back in the day anyone who was serious would not want to be caught with a Yamaha Eq or compressor in their rack, lets hope the 2032 and 2020 are not what they used to model the digital versions from.
For my digital disclaimer - I am only moderately proficient on a LS9 and someday I will own a digital board of some flavor.
Here's something to ponder. Back in the earlier days of digital boards there were offerings of digital compressors, digital gates and a few other digital rack mount tricks other than FX units, many were maybe still are offered from well respected companies XTA and Drawmer come to mind. They were greeted with a cool maybe somewhat skeptical welcome. Put those digitized units into a digital board and there welcomed with open arms. Way back in the day anyone who was serious would not want to be caught with a Yamaha Eq or compressor in their rack, lets hope the 2032 and 2020 are not what they used to model the digital versions from.
For my digital disclaimer - I am only moderately proficient on a LS9 and someday I will own a digital board of some flavor.
Last edited:
Re: Digital Boards - Sound Quality
For reference, that is just slightly worse than most high end oscilloscopes. Call me skeptical.
For reference, that is just slightly worse than most high end oscilloscopes. Call me skeptical.
Re: Digital Boards - Sound Quality
I've spent a fair bit of time with the compressors in the M7/LS9. Are they great? No, but they're not Yammy comps of 20 years ago, either. Likewise the graphic EQs don't appear to be emulations of the rightly disliked EQ2031. Wherever they came from, the comps and GEQ are adequate (maybe more so) for the level of mixer they appear on. They are what they are, however, and on these desks I don't see plug-ins running inside the console OS; the processing horsepower doesn't seem to be there for these extra tasks.
That said, I'm really starting to like certain plug ins in the Avid Venue package. The Joe Meek SC2 (I own the hardware, too) and Aggro compressors; ReelTape saturation, the Pultec EQ and comps. I'd never have a chance to use the vintage EQ and comps otherwise, and I'm coming to appreciate their uses and the tonal elegance behind some of the designs. Should I have the misfortune to become rich, I'm buying the WAVES Live bundle, too. Some cool things in that package...
Have fun, good luck.
Tim Mc
I've spent a fair bit of time with the compressors in the M7/LS9. Are they great? No, but they're not Yammy comps of 20 years ago, either. Likewise the graphic EQs don't appear to be emulations of the rightly disliked EQ2031. Wherever they came from, the comps and GEQ are adequate (maybe more so) for the level of mixer they appear on. They are what they are, however, and on these desks I don't see plug-ins running inside the console OS; the processing horsepower doesn't seem to be there for these extra tasks.
That said, I'm really starting to like certain plug ins in the Avid Venue package. The Joe Meek SC2 (I own the hardware, too) and Aggro compressors; ReelTape saturation, the Pultec EQ and comps. I'd never have a chance to use the vintage EQ and comps otherwise, and I'm coming to appreciate their uses and the tonal elegance behind some of the designs. Should I have the misfortune to become rich, I'm buying the WAVES Live bundle, too. Some cool things in that package...
Have fun, good luck.
Tim Mc
Re: Digital Boards - Sound Quality
No doubt... If i was trying to sell new expensive digital consoles I'd be telling customers how low my jitter was too....
. Maybe I'd lower it some more 2 years from now too.
It seems simple (but expensive) to set up a pair of these digital wonder consoles side by side, with identical real music inputs, one with a big Ben and one without then null them out.. The latencies and whatever should cancel out leaving pure "massive" clock jitter related ERROR... or not. Now perhaps an actual massive difference in the null residual could convince me... but i'm not a customer for big dog consoles, so I don't need to be convinced. You folks need to figure out for yourselves how to spend your money.
Another popular recording studio affectation is that somehow digital sum buses are inferior to analog combining (with passive sum buses even more loved). As an old analog designer I know better than that, for the theory part of it at least. However I can't account for every digital console platform out there so who knows what might be lurking in the woodpiles.
If somebody says they heard a massive something, I'm not trying to make them out as liars, maybe they did hear something significant, but perhaps it wasn't caused by jitter, or digital sum buses, or.... That's why I prefer to measure stuff... it makes it easier to identify the malefactors and drive them into submission.
JR
No doubt... If i was trying to sell new expensive digital consoles I'd be telling customers how low my jitter was too....
. Maybe I'd lower it some more 2 years from now too. It seems simple (but expensive) to set up a pair of these digital wonder consoles side by side, with identical real music inputs, one with a big Ben and one without then null them out.. The latencies and whatever should cancel out leaving pure "massive" clock jitter related ERROR... or not. Now perhaps an actual massive difference in the null residual could convince me... but i'm not a customer for big dog consoles, so I don't need to be convinced. You folks need to figure out for yourselves how to spend your money.
Another popular recording studio affectation is that somehow digital sum buses are inferior to analog combining (with passive sum buses even more loved). As an old analog designer I know better than that, for the theory part of it at least. However I can't account for every digital console platform out there so who knows what might be lurking in the woodpiles.
If somebody says they heard a massive something, I'm not trying to make them out as liars, maybe they did hear something significant, but perhaps it wasn't caused by jitter, or digital sum buses, or.... That's why I prefer to measure stuff... it makes it easier to identify the malefactors and drive them into submission.
JR
Re: Digital Boards - Sound Quality
Regarding digital compression, I recall the early stand alone digital comps and they were odd ducks. Well ahead of their time, and the market.
In theory, one of the hard parts about designing a compressor is VCA linearity and noise (or at least it was.). Inside the digital domain a gain change is a simple multiply that does not degrade signal linearity... both approaches sound will probably be dominated by side chain time-constants and techniques. Digital offers some extra capabilities that analog can not match, especially if you can tolerate some output latency to let the side chain look ahead in time.
Again, I will only argue theory (what should be) not execution (what is). Digital dynamics have a clear edge on theory, analog dynamics have more practice making it sound good.
JR
Regarding digital compression, I recall the early stand alone digital comps and they were odd ducks. Well ahead of their time, and the market.
In theory, one of the hard parts about designing a compressor is VCA linearity and noise (or at least it was.). Inside the digital domain a gain change is a simple multiply that does not degrade signal linearity... both approaches sound will probably be dominated by side chain time-constants and techniques. Digital offers some extra capabilities that analog can not match, especially if you can tolerate some output latency to let the side chain look ahead in time.
Again, I will only argue theory (what should be) not execution (what is). Digital dynamics have a clear edge on theory, analog dynamics have more practice making it sound good.
JR
Re: Digital Boards - Sound Quality
Are you talking about the cute virtual knobs on the left part of the screen (to which one has to cursor AND select - a bit of a nuisance), or a slick method I've yet to discover? I hope the latter, as I don't find the former to be much if any quicker than sends-on-fader, mix-by-mix.
Are you talking about the cute virtual knobs on the left part of the screen (to which one has to cursor AND select - a bit of a nuisance), or a slick method I've yet to discover? I hope the latter, as I don't find the former to be much if any quicker than sends-on-fader, mix-by-mix.
Silas Pradetto
Graduate Student
Re: Digital Boards - Sound Quality
I've mixed on an LS9 - get ready for this - ONCE.
Pushing an aux button once makes it the 'selected' aux such that I could mix any channel to that aux, and pushing the aux button twice got me sends on fader for that aux.
Both have a purpose, depending on if it's an initial soundcheck situation or making an adjustment on-the-fly during the show
PS - I hate the LS9!
I've mixed on an LS9 - get ready for this - ONCE.
Pushing an aux button once makes it the 'selected' aux such that I could mix any channel to that aux, and pushing the aux button twice got me sends on fader for that aux.
Both have a purpose, depending on if it's an initial soundcheck situation or making an adjustment on-the-fly during the show
PS - I hate the LS9!
Re: Digital Boards - Sound Quality
For the record, I never claimed anything "massive", and would be very hesitant to do so. Certainly what I have noticed in my own experiences would be classified as more subtle than massive. I too would be interested in your null test idea, though don't have a convenient way of doing such a test right now (the studio I am doing drum sessions in at the moment has Avid hardware clocked through a Big Ben, but I won't have an opportunity to chew up studio and engineer time to do such testing).
For the record, I never claimed anything "massive", and would be very hesitant to do so. Certainly what I have noticed in my own experiences would be classified as more subtle than massive. I too would be interested in your null test idea, though don't have a convenient way of doing such a test right now (the studio I am doing drum sessions in at the moment has Avid hardware clocked through a Big Ben, but I won't have an opportunity to chew up studio and engineer time to do such testing).
Silas Pradetto
Graduate Student
Re: Digital Boards - Sound Quality
I don't remember where I read it, and I believe it's been alluded to in this thread, but someone did measure the difference between an internally and externally clocked Yamaha desk. It turned out that the externally clocked board had more distortion and a higher noise floor than the internally clocked one, mainly because of how the board locks onto the external clock. Even though the externally clocked board measures worse, there is nothing saying that coloration is always a bad thing.
I don't remember where I read it, and I believe it's been alluded to in this thread, but someone did measure the difference between an internally and externally clocked Yamaha desk. It turned out that the externally clocked board had more distortion and a higher noise floor than the internally clocked one, mainly because of how the board locks onto the external clock. Even though the externally clocked board measures worse, there is nothing saying that coloration is always a bad thing.
Re: Digital Boards - Sound Quality
And therein lies the rub. Midas analog desks distort and color the input sounds; some folks use techniques that deliberately exploit this because they prefer the results over running a different desk or running the Midas differently. The thing is, we have a pretty good idea of what's going on in the analog realm.
Providing an external master clock should not affect the audio because of WHAT master clocking does and that any failure of clocking, internal or external, will manifest itself in plainly audible ways that won't be mistaken for "better." Andy Peters (among others) did a pretty good job of describing why this dog don't hunt (it's in the archived PSW forums).
Another thing JR has said is "I no longer try to tell people what they think they are hearing." When presented in person with this topic I usually just smile and say "oh, that's very nice." My listening doesn't confirm that the addition of an external clock changes the analog output of a digital console... but I'm over 50 and have 30+ years of live sound "experience" on my ears, so it's possible I'm not hearing the Emperor's New Sound. YMMV.
Have fun, good luck.
Tim Mc
And therein lies the rub. Midas analog desks distort and color the input sounds; some folks use techniques that deliberately exploit this because they prefer the results over running a different desk or running the Midas differently. The thing is, we have a pretty good idea of what's going on in the analog realm.
Providing an external master clock should not affect the audio because of WHAT master clocking does and that any failure of clocking, internal or external, will manifest itself in plainly audible ways that won't be mistaken for "better." Andy Peters (among others) did a pretty good job of describing why this dog don't hunt (it's in the archived PSW forums).
Another thing JR has said is "I no longer try to tell people what they think they are hearing." When presented in person with this topic I usually just smile and say "oh, that's very nice." My listening doesn't confirm that the addition of an external clock changes the analog output of a digital console... but I'm over 50 and have 30+ years of live sound "experience" on my ears, so it's possible I'm not hearing the Emperor's New Sound. YMMV.
Have fun, good luck.
Tim Mc
Re: Digital Boards - Sound Quality
Just to clarify, I don't argue with people about what they are sure they are hearing either. I can't hear through their ears... and in some cases they may be hearing something else and drawing an erroneous conclusion. Or, maybe I'm wrong..
Jeff: yes you didn't say "massive", you said "notable" and "you WILL notice a difference". Perhaps... I can't argue about things I haven't experienced first hand.
Noise and distortion similar to quantization are expected results from clock uncertainty. I am not aware of this sounding euphonious, but that may be subjective. IIRC wideband analog hiss has been positively perceived as extended frequency response, so perhaps a win-lose depending on your personal perspective about both.
JR
Just to clarify, I don't argue with people about what they are sure they are hearing either. I can't hear through their ears... and in some cases they may be hearing something else and drawing an erroneous conclusion. Or, maybe I'm wrong..
Jeff: yes you didn't say "massive", you said "notable" and "you WILL notice a difference". Perhaps... I can't argue about things I haven't experienced first hand.
Noise and distortion similar to quantization are expected results from clock uncertainty. I am not aware of this sounding euphonious, but that may be subjective. IIRC wideband analog hiss has been positively perceived as extended frequency response, so perhaps a win-lose depending on your personal perspective about both.
JR