Bogus claims Re: OTB summing from Black Lion

Recording Techniques, People Skills, Gear, Recording Spaces, Computers, and DIY

Moderators: drumsound, tomb

Locked
User avatar
leigh
carpal tunnel
Posts: 1636
Joined: Wed May 07, 2003 11:16 am
Location: Portland, OR
Contact:

Post by leigh » Thu Jun 02, 2011 2:31 pm

TV Lenny wrote:I am curious what BLA says too on this.
Thus far their email responses to me have been lengthy, and they are standing by the assertions made on their page.

I don't want to get too into yet, but they are saying that by "clock based errors" they mean the "DSP clock". So, not the digital audio "word clock", but a computer's regular CPU clock.

Once I get some more answers from them (about what kind of errors they think a DSP clock causes), I'll share that info at greater length.

Leigh

User avatar
roygbiv
re-cappin' neve
Posts: 703
Joined: Sat Jun 16, 2007 6:02 pm
Location: Portland, Oregon

Post by roygbiv » Fri Jun 03, 2011 12:29 am

keep us posted leigh, I'm baffled by their assertion.

I'm no expert, but I have been using computers for some time now, ever since my older brother brought home a brand new "portable" Osborne computer in 1981, and made me ashamed of my CoCo Tandy Trash 80 (back in the day, I even had to do some some assembly programing to get a Z80 processor-based scientific system to talk to another instrument).

Anyway, I raise all this because well, I like to talk about myself. But also, I thought I had a pretty good grasp on how computers fundamentally worked, and this now is causing me some cognitive dissonance - I don't understand how the BLA stance makes any sense. Are they saying there is no Santa Claus too?

Presumably, after the initial DA conversion, the only "clock" any internal calculations done ITB would depend on would be the master CPU clock which drives the cycle time of the computer - but every calculation "waits" on this clock for its turn - there is no timing error there, just wait time for the next calculation (and that wait time is a lot less now than back in the 4.77 MHz days). The DSP clock they are referring to (I thought) was based on a different quartz oscillator that is only used to control the timing of sampling used in the AD/DA conversions.

(Maybe I am wrong here, please someone correct me if I am).

Moreover, even if I'm wrong about the above - if the internal summing calculations for some reason depend on the DSP clock, how could they introduce more clock based errors than would be introduced if the same DSP clock is used to play back the data (samples) out of the computer in the DA stage, to go through their summing box? Doesn't make any sense at all.

Maybe they are trying to say that playing back through their summing box will smooth out the clock based jitter/errors that originally occurred upon the AD/DA conversion (analogous to how dither works or something).

But if that is what they are trying to say, why don't they just say that?

Or, maybe I am wrong in my above assumptions and will learn something new here.
"Decisions by the Secretary pursuant to the authority of this Act are non-reviewable and committed to agency discretion, and may not be reviewed by any court of law or any administrative agency."

kslight
mixes from purgatory
Posts: 2968
Joined: Tue Oct 13, 2009 7:40 pm

Post by kslight » Fri Jun 03, 2011 5:26 am

KennyLusk wrote:At the time I had BLA do the TweakHead mod on my (2 year old) 1814fw I didn't have $850 for an RME (or better) interface, but I did have $500 for the mod. Since I already owned the 1814 I spent $500 to get something worth $850+, is the way I look at it.
I opted for a used Alesis HD24XR for $600 for use as my A/D and D/A conversion and clock, and I got a bonus 24 track recorder as well! The point is more about buying reasonable quality gear to begin with and taking advantage of the used market to get great deals on pro gear, rather than keep pumping money into upgrading throwaway gear like an 1814 (which I feel I have a right to say...because I have one). When I contacted BLA about the 1814 mod they even said that it did not make as great of an improvement as they would like, which is why they no longer offer it.

I consider the total amount invested, cost of soundcard and mod. If I recall the 1814 was $400 or $500 new.

And the 003 mods get real pricey, and you have got to assume that for what totals well over $2000 altogether you could find something better than hot rodding a mediocre soundcard.

The hot rodding analogy is probably less accurate than just buying a new disposable car like a Kia to try to replace all the cheap parts to bring it up to standard like a Honda, or for the same or less money you could get a Honda with a couple years on it that will outlive and is markedly superior in build quality to the Kia.

KennyLusk
dead but not forgotten
Posts: 2037
Joined: Wed Sep 22, 2004 10:22 am
Location: Ramah, New Mexico

Post by KennyLusk » Fri Jun 03, 2011 6:19 am

kslight wrote:
KennyLusk wrote:At the time I had BLA do the TweakHead mod on my (2 year old) 1814fw I didn't have $850 for an RME (or better) interface, but I did have $500 for the mod. Since I already owned the 1814 I spent $500 to get something worth $850+, is the way I look at it.
I opted for a used Alesis HD24XR for $600 for use as my A/D and D/A conversion and clock, and I got a bonus 24 track recorder as well! The point is more about buying reasonable quality gear to begin with and taking advantage of the used market to get great deals on pro gear, rather than keep pumping money into upgrading throwaway gear like an 1814 (which I feel I have a right to say...because I have one). When I contacted BLA about the 1814 mod they even said that it did not make as great of an improvement as they would like, which is why they no longer offer it.

I consider the total amount invested, cost of soundcard and mod. If I recall the 1814 was $400 or $500 new.

And the 003 mods get real pricey, and you have got to assume that for what totals well over $2000 altogether you could find something better than hot rodding a mediocre soundcard.

The hot rodding analogy is probably less accurate than just buying a new disposable car like a Kia to try to replace all the cheap parts to bring it up to standard like a Honda, or for the same or less money you could get a Honda with a couple years on it that will outlive and is markedly superior in build quality to the Kia.
Nothing personal man, but every studio I ever recorded in that used those f'ing Alesis units was a nightmare. They crash too often and we lost tracks a few times because of it. Then the sessions would be put on hold while they brought up a redundant backup and sent the crashed unit to the manufacturer for repair. Alesis ADAT's?...not for me man. Anyway, that was just my experience.

When I asked BLA about the TweakHead mod they had nothing but positive things to say to me at the time so I spent the $500 to do it. Best $500 I'd ever spent in my studio if you ask me. The differences were very clear and it opened my eyes and ears to a whole new level of recording engineering.

I stand by BLA quality based on my experience with them. I don't know about this particular issue being discussed and I suppose we'll wait to hear what they have to say about this marketing language people are freaking over. But their product?...quality.

Funny our world, isn't it? 2 different people...2 totally different experiences. I'll never figure this planet out.
"The mushroom states its own position very clearly. It says, "I require the nervous system of a mammal. Do you have one handy?" Terrence McKenna

kslight
mixes from purgatory
Posts: 2968
Joined: Tue Oct 13, 2009 7:40 pm

Post by kslight » Fri Jun 03, 2011 6:41 am

No worries... Not an ADAT though, this is a hard disk recorder. Actually an upgraded model (this one has different converters than the cheaper one, and allows 96k recording. Considered by many to be poor man's Radar). I agree ADATs are a nightmare. I have had zero issues with this unit though, and I bring it on the road to do transfers off of 2" tape at a non-digitally equipped studio.

I'm not attempting to criticize BLA's quality or service because I've never directly used anything of theirs, just the price/performance ratio of the upgrades.

Jim Williams
tinnitus
Posts: 1135
Joined: Sat Jun 03, 2006 8:19 am
Location: beautiful Carlsbad, CA
Contact:

Post by Jim Williams » Fri Jun 03, 2011 8:22 am

kslight wrote:No worries... Not an ADAT though, this is a hard disk recorder. Actually an upgraded model (this one has different converters than the cheaper one, and allows 96k recording. Considered by many to be poor man's Radar). I agree ADATs are a nightmare. I have had zero issues with this unit though, and I bring it on the road to do transfers off of 2" tape at a non-digitally equipped studio.

I'm not attempting to criticize BLA's quality or service because I've never directly used anything of theirs, just the price/performance ratio of the upgrades.
I've used my HD24 since they were first released, I bought the first EC-2 upgrade kit to convert to 96k analog I/O. 8 or 9 years without one problem.

Don't confuse video tape based recorders with hard drive recorders, there is quite a difference.
Jim Williams
Audio Upgrades

User avatar
leigh
carpal tunnel
Posts: 1636
Joined: Wed May 07, 2003 11:16 am
Location: Portland, OR
Contact:

Post by leigh » Fri Jun 03, 2011 4:32 pm

roygbiv wrote:But also, I thought I had a pretty good grasp on how computers fundamentally worked, and this now is causing me some cognitive dissonance - I don't understand how the BLA stance makes any sense. Are they saying there is no Santa Claus too?

Presumably, after the initial DA conversion, the only "clock" any internal calculations done ITB would depend on would be the master CPU clock which drives the cycle time of the computer - but every calculation "waits" on this clock for its turn - there is no timing error there, just wait time for the next calculation (and that wait time is a lot less now than back in the 4.77 MHz days). The DSP clock they are referring to (I thought) was based on a different quartz oscillator that is only used to control the timing of sampling used in the AD/DA conversions.

(Maybe I am wrong here, please someone correct me if I am).
I am with you there. It's hurting my brain trying to understand the Black Lion Audio guy's emails, just sorting through all these assertions he's making that seem quite backwards to me. For the record, I don't know who I'm talking with at the company, because he's not signing the emails with a name, just as "BLA".

One point of clarity about the clocks: I believe that for a regular DAW on a computer, "DSP clock" and "CPU clock" are the same thing. That's the clock that controls the frequency at which processor calculations are performed. On a modern computer, the speed of that clock is measured in gigahertz.

The other kind of clock would be a "wordclock" or "sample clock". That, of course, typically runs at one of the standard digital audio speeds, like 44.1 kHz or 96 kHz.

The quality of a wordclock (like whether it has a lesser or greater amount of jitter) can greatly affect the digital audio signal, because a wordclock is synchronous with the audio data. The quality of a DSP clock doesn't matter so much, as long as it's happy playing with the rest of the computer system. The DSP clock is asynchronous in relation to the audio data - as you say, "every calculation 'waits' on this clock for its turn - there is no timing error there."

Leigh

User avatar
Scodiddly
genitals didn't survive the freeze
Posts: 3958
Joined: Wed Dec 10, 2003 6:38 am
Location: Mundelein, IL, USA
Contact:

Post by Scodiddly » Fri Jun 03, 2011 4:54 pm

"DSP clock" affects calculations? Man, that sure smells funny. If I get a better clock will my spreadsheets be more accurate? Will my tax returns have a bigger refund?

User avatar
leigh
carpal tunnel
Posts: 1636
Joined: Wed May 07, 2003 11:16 am
Location: Portland, OR
Contact:

Post by leigh » Fri Jun 03, 2011 5:43 pm

Scodiddly wrote:"DSP clock" affects calculations? Man, that sure smells funny. If I get a better clock will my spreadsheets be more accurate? Will my tax returns have a bigger refund?
Yeah, it does smell funny.

I'm debating whether to share the entirety of my discussion with "BLA" here. First I would like to at least make sure he and I are on the same page, and I'm trying to reach some kind of conclusion to our dialog. But for those of you dying to know more, here's the first part of his response.

I had sent him a slightly edited version of my original post in this thread. To recap, my first assertion was that "The process of summing ITB is, in fact, not prone to clock-based errors - because there is no clocking going on!"

His response:
First, the summing done in the digital domain requires the use of DSP. As you may know, DSP processes are not conducted in a vacuum, and they are not all the same. They are subject to real-world limitations, such as DSP math processes, logic switching and current movements, gate input capacitances, cross-talk related anomalies, resonant poles on the supply line, etc. When we say "clock-based errors," we are not referring sample or master clock signals, we are talking about issues created within DSP by logic switching noise, eye pattern errors, cross-talk issues, and other real-world imperfections that manifest themselves in our otherwise simple and ideal "sum all samples at a time" scenario.
I dunno, what do you make of that?

I should mention that I do agree that "DSP math processes" are one of the limitations of digital summing. The math is carried out with a finite number of bits, and at some point you're probably scaling down the signal (reducing gain). This reduction (via multiplication) results in a tiny rounding error. These errors can be cumulative and, in theory, audible - but on a well-designed 24-bit system it's not a big issue.

Other than that one point, the rest makes my brain hurt.

Leigh

User avatar
Scodiddly
genitals didn't survive the freeze
Posts: 3958
Joined: Wed Dec 10, 2003 6:38 am
Location: Mundelein, IL, USA
Contact:

Post by Scodiddly » Fri Jun 03, 2011 6:33 pm

leigh wrote: His response:
First, the summing done in the digital domain requires the use of DSP. As you may know, DSP processes are not conducted in a vacuum, and they are not all the same. They are subject to real-world limitations, such as DSP math processes, logic switching and current movements, gate input capacitances, cross-talk related anomalies, resonant poles on the supply line, etc. When we say "clock-based errors," we are not referring sample or master clock signals, we are talking about issues created within DSP by logic switching noise, eye pattern errors, cross-talk issues, and other real-world imperfections that manifest themselves in our otherwise simple and ideal "sum all samples at a time" scenario.
I dunno, what do you make of that?
I'd call it rank stinky audiophool bullshit?

Digital logic is digital logic. Plugin programmers don't worry about "eye pattern errors" because such issues have been dealt with already by the chip designers. DSP and other logic chips can (and are) tested to make sure that they do exactly what they're designed to do.

Nate Dort
tinnitus
Posts: 1039
Joined: Mon Mar 08, 2004 6:07 pm
Location: Detroit
Contact:

Post by Nate Dort » Fri Jun 03, 2011 6:48 pm

I'm inclined to agree with Scott. Sounds like BLA is trying to spin the tiny amount of error you would get on the die of the CPU as being something you could hear and pick out of a lineup.

"AMD CHIPS SOUND SO MUCH WARMER THAN INTEL BECAUSE THE CPU IS PHYSICALLY LARGER!!!!!1!!1!"

Last time I checked, analog circuits also have to deal with stray capacitance, crosstalk, and current movements, among other things, and on a much larger scale.

FWIW, I'm not disagreeing with the assertion that analog summing sounds different or "better." I'm disagreeing with the reasoning behind their assertions.

User avatar
rhythm ranch
mixes from purgatory
Posts: 2793
Joined: Wed May 07, 2003 8:45 pm
Location: Corrales, NM

Post by rhythm ranch » Fri Jun 03, 2011 9:34 pm

leigh wrote:I'm debating whether to share the entirety of my discussion with "BLA" here.
Wouldn't it be easier (and more open) to provide BLA with a link to this thread so that they can respond directly to the various concerns being expressed here?

User avatar
masteroftracks
audio school
Posts: 6
Joined: Thu May 06, 2010 1:40 pm
Location: Kansas City

Post by masteroftracks » Sat Jun 04, 2011 8:19 am

I took some bad acid last week while mixing and had serious "eye pattern errors". Do you think BLA could mod my drugs so I get less of that?
I think I might be able to help with the whole Pan-Pam dilemma.......

User avatar
leigh
carpal tunnel
Posts: 1636
Joined: Wed May 07, 2003 11:16 am
Location: Portland, OR
Contact:

Post by leigh » Sat Jun 04, 2011 9:08 am

rhythm ranch wrote:Wouldn't it be easier (and more open) to provide BLA with a link to this thread so that they can respond directly to the various concerns being expressed here?
Yeah, good call. When I started this thread, I thought BLA would jump in to the conversation sooner or later. But, since they haven't, and the dude seems to have given up on our email dialog, it may be time to extend the invite directly.

User avatar
leigh
carpal tunnel
Posts: 1636
Joined: Wed May 07, 2003 11:16 am
Location: Portland, OR
Contact:

Post by leigh » Sat Jun 04, 2011 9:35 am

Scodiddly wrote:Digital logic is digital logic. Plugin programmers don't worry about "eye pattern errors" because such issues have been dealt with already by the chip designers. DSP and other logic chips can (and are) tested to make sure that they do exactly what they're designed to do.
Nate Dort wrote:I'm inclined to agree with Scott. Sounds like BLA is trying to spin the tiny amount of error you would get on the die of the CPU as being something you could hear and pick out of a lineup.
Bingo.

Nate - to be clear, within a CPU (and surrounding circuitry), any errors are checked for and corrected. That's the glory of digital systems. If some voltage is a little off, it doesn't matter, because it's still read as a 0 or 1. If some voltage is a lot off, it is caught and fixed with error checking routines.

In other words, while the actual voltages in the circuit may vary and the data lines might have all kinds of noise on them, on the level of bits digital summing is 100% accurate.

Or, as Bob Katz says, "there is absolutely nothing wrong with digital summing, it is essentially perfect, especially since adding numbers is the easiest thing you can ask a DSP to do".

Leigh

Locked

Who is online

Users browsing this forum: No registered users and 51 guests