Objectivism explained...

"The absence of noise is not the presence of music." HP

Great topic and interesting examples Tim. To me though, this does not fit the classic definition of an "objectivist" and by definition, objectivisim. Take your DAC example. The Navguy talks about how 98 db was not where it was at but better than 110 db. I have never seen vocal, self-appointed objectivists take that position. Indeed, I always talk about how we should get up to 96 db as to clear most of the hurdle for 16 bits and that a device that does say, 85 db is not good enough. I get strong pushback, in the form of months of posting and complaining in different threads that I am wrong. And that the only thing that matter is an ABX test showing any of that matters. If not, then 85 db or even lower would be sufficient. And importantly, absence of any double blind tests is proof that such specs are good enough!

In addition they absolutely believe chip specs unlike the good point NWavguy mentions that actual designs rarely match chip specs. When I put forward articles that show how challenging it is to get those numbers and how they are created in lab environment, they roll their eyes and ignore that information.

So having taken even a lower position than your example person here, and gotten more grief than I ever have on audio forums, I say this is not an accepted bar. If it were, we would be much closer to where we need to be. So question to you is how close to you think your definitional views are in this regard to the masses that call themselves objectivists in audio forums?

Me? I am an equal opportunity abuser . I look at the good work of this individual and can still find reason to criticize . He uses the ESS 9023 DAC in his design and produces very good specs. I go and look up the specs for the chip and there is marketing phrase about a circuit in there, "time domain jitter eliminator." No mention on their site as to what it does. Nothing! That is like saying you have produced a perpetual motion machine. You can't eliminate jitter. So I go and google some and see that my guess is right that it is some sort of asynchronous sample rate conversion. This is a process by which you generate your own clock for the DAC. To keep it from getting out of sync to the input, you resample. So let's say the input rate is 47,995 samples/sec instead of 48,000. You run your clock at 48,000 and perform a sample rate conversion from 47,995 to 48,000. If the input rate goes up to 47,997, that becomes the ratio to 48,000 in the sample rate conversion. What this means is that you eliminated input jitter since variations in input, did not result in variations in output. But that does NOT mean that you have eliminated distortion. You have traded jitter distortion for your sample rate distortion. Further, there is logic that modifies that conversion ratio. It has to then track the input rate and make the appropriate sample rate conversion ratio. That process actually becomes a form of tracking error not unlike jitter!

Here is the kicker. Once you have such a circuit, you no longer can test it using jitter test. His J-Test signal he used is not appropriate to detect errors in this circuit. Special tests need to be used based on knowing the design of the sample rate conversion circuit to detect its flaws. Such a test is much more complicated than jitter test especially since ESS does not seem to document what the block does. In addition, whether the circuit does its job right or not depends on what the PC is doing on the USB side, adding incredible complexity. So it is not done.

Even simpler is the issue of PCs vary hugely as to how clean, powerful and stable their USB power is. I hear that he has done a lot to isolate it but the testing needs to include many PCs as to give us confidence that he has conquered the variations there. Or at least a simulation of the same.

Long way of saying, this audio thing can be quite complex. Positions we take are only as valid as our knowledge level. A person who has convinced themselves of being an objectivist (and I don't mean you), based on just reading forum posts and their gut feeling, many indeed not realize that they are taking too simplistic view here. Guys in car hobbies become experts at what they are messing with. I know my two boys are. In audio though, I find a tendency to not want to learn what the engine does, yet have strong opinion about its performance specs not mattering. I think that is a mistake on the part of objectivists. If anyone needs to become an expert in technology, it should be them. The other camp is going by what they hear so they don't have that requirement. Yet, I find self-appointed forum objectivists to be dismissive of technology and its design to incredible levels of dumbing it down to, "do you have an audio test? If not, I don't believe it. And no, I don't need to be an engineer or know what the design is to be right." That doesn't sit well with me. I think it gives the whole position a bad name. It really does.

Phelonious Ponk said:

WellThere are articulate, highly-knowledgeable objectivists out there on the net running tests, designing gear and reporting on it regularly. Sean Olive's blog, The Well-tempered computer, Doug Self and a few others come to mind.

Click to expand...

Sean Olive is a Harman colleague of mine. I'm respectful of and impressed with his accomplishments here, including (but not limited to) the construction of identical audio reference rooms at all facilities and the development of a listener training program so subjective audio phenomena can be described by test subjects using a defined objective vocabulary. The combination facilitates A/B product tests whose results and findings can be rigorously analyzed.

And I like his blog.

Tim,

I know about this site for long. I appreciate their technical skills and organization, but their beliefs are clearly stated at:

http://nwavguy.blogspot.pt/2012/04/what-we-hear.html

and links referred in it. Not my cup of tea.

IMHO, they are not objectivists, I would call them measurativists.
But their real target seems the Benchmark and headphone listening, nothing else. Quoted from the page you refer:

BLIND EVIDENCE: So far I’ve run two relatively informal blind tests with the ODAC. The latest one used special software on the PC to play the same track on both my Benchmark DAC1 Pre and simultaneously on the ODAC plugged into the same PC (both connected via USB and running at 24/44). The ODAC was connected to an O2 headphone amp, and a switchbox allowed the headphones to be rapidly switched between the DAC1 and the O2+ODAC. The two sources were carefully level matched (using their respective volume controls) using a test signal and wideband DMM. I tried both my Sennheiser HD650 and Denon AH-D2000 headphones with a variety of well recorded favorite tracks. One other listener and I could not reliably tell which was playing.

IMHO, not enough for a true objectivist.
BTW, are you going to sell your Benchmark soon?

Thomas.Dennehy said:

Sean Olive is a Harman colleague of mine. I'm respectful of and impressed with his accomplishments here, including (but not limited to) the construction of identical audio reference rooms at all facilities and the development of a listener training program so subjective audio phenomena can be described by test subjects using a defined objective vocabulary. The combination facilitates A/B product tests whose results and findings can be rigorously analyzed.

And I like his blog.

Click to expand...

Sean Olive has developed a methodology to correlate subjective findings with objective data. IMHO, this method has some features that make it a very useful tool for Harman developments and audio science research. However, sometimes his findings are careless used for marketing purposes and fundamentalist groups use and abuse of his findings.

microstrip

Would you venture a guess as to what an "objectivist" is?

@everyone

NWAV person did post in this forum some time ago . I ,also, like his positions

here is what you get with Objectivism taken all the way....

http://www.hydrogenaudio.org/forums/index.php?showtopic=3974

read #8. those guys must be lots of fun. please just shoot me next time i even think about looking at that forum.

as far as explaining it, well, why would you even care to do that?

Phelonious Ponk said:

Excellent points, Amir. Perhaps we need some new labels (or perhaps we need to eliminate labels altogether). I know the folks you're referring to as classic objectivists. I used to be one of them. Then I started hanging around here and reading Sean Olive's blog and NWAV Guy and a few other sites. I'm not an engineer, and I don't understand all of it, but I get enough of it to understand, as you've said, that it's not as simple, as black and white, as some "objectivists" would make it. I also understand your point that "objectivists" have a greater responsibility for technical knowledge. I don't know if I'll ever live up to mine, but I'm learning, and having fun doing it.

In the end, I'm just as subjective as anybody. I like what I like. But what I like seems to correlate to measurable transparency/neutrality/accuracy pretty well, so I find the kind of experimenting and writing found in places like NWAV Guy interesting and informative. For everyone whose mileage varies...enjoy what you enjoy and read Michael Fremer or whatever subjectivist audio writer seems to fall most in line with your personal tastes. Unless you want them challenged, which I personally think is more fun.

There is one place where I remain a "classic objectivist," though. I've run enough personal, informal AB/X testing here at home to conclude that what I think I hear and what I can actually differentiate unsighted can be two very different things. And if I can't hear it, if it is below my personal noise floor, it is irrelevant to me. And while others may very well be able to hear what I cannot, if they are unwilling to do even the simplest, most informal blind listening to confirm that they are hearing real differences, not their expecatations...well, they may trust their ears, but I will not.

YMMV, etc.

Tim

Click to expand...

I can go along with that .

Mike Lavigne said:

here is what you get with Objectivism taken all the way....

http://www.hydrogenaudio.org/forums/index.php?showtopic=3974

read #8. those guys must be lots of fun.

as far as explaining it, well, why would you even care to do that?

Click to expand...

Indeed, that is as "classic" as it gets. To me, it is a form of censorship that says no data can be shared if it is not accompanied with a formal listening test. The people who say that, are usually the people who don't appreciate the work and difficulty of conducting such. And that it is unreasonable to expect customers of products to routinely conduct such tests.

At the same time, there is something to be said as Tim mentioned about once in a while, calibrating what you think you are hearing, against what you may be hearing. I don't know how we can avoid doing that and yet, fully trust our judgments.

+100, amirm

amirm said:

Great topic and interesting examples Tim. .........
Me? I am an equal opportunity abuser . I look at the good work of this individual and can still find reason to criticize . He uses the ESS 9023 DAC in his design and produces very good specs. I go and look up the specs for the chip and there is marketing phrase about a circuit in there, "time domain jitter eliminator." No mention on their site as to what it does. Nothing! That is like saying you have produced a perpetual motion machine. You can't eliminate jitter. So I go and google some and see that my guess is right that it is some sort of asynchronous sample rate conversion. This is a process by which you generate your own clock for the DAC. To keep it from getting out of sync to the input, you resample. So let's say the input rate is 47,995 samples/sec instead of 48,000. You run your clock at 48,000 and perform a sample rate conversion from 47,995 to 48,000. If the input rate goes up to 47,997, that becomes the ratio to 48,000 in the sample rate conversion. What this means is that you eliminated input jitter since variations in input, did not result in variations in output. But that does NOT mean that you have eliminated distortion. You have traded jitter distortion for your sample rate distortion. Further, there is logic that modifies that conversion ratio. It has to then track the input rate and make the appropriate sample rate conversion ratio. That process actually becomes a form of tracking error not unlike jitter!

Click to expand...

Correct, amirm, the ES9023 uses a form of ASRC. I have worked on this chip, use it in one of my DACs & introduced it to both DIYA & the SDR Widget guys, one of whom is a partner with NWAguy in this project. The interesting thing about this ASRC is that it operates on a different basis from the normal PLL based ASRC solution. ESS keep the details fairly secret but it seems to essentially use 3 samples to determine the clock speed of the incoming clock & upsamples the signal to 864KHz speed & does all it's work here. It is a different approach to ASRC but is it a "jitter eliminator"? No! Just try sources with different jitter feeding it to hear the difference. The really interesting bit is that you can bypass this ASRC & guess what, it sounds better (if you signal is a low jitter source ). So if NWAguy is an objectivist he should check out some details of th eDAC chip he is using & at least get the claims correct.

Here is the kicker. Once you have such a circuit, you no longer can test it using jitter test. His J-Test signal he used is not appropriate to detect errors in this circuit. Special tests need to be used based on knowing the design of the sample rate conversion circuit to detect its flaws. Such a test is much more complicated than jitter test especially since ESS does not seem to document what the block does. In addition, whether the circuit does its job right or not depends on what the PC is doing on the USB side, adding incredible complexity. So it is not done.

Click to expand...

Yes, I will be interested to see his jitter measurements when he does it correctly!

Even simpler is the issue of PCs vary hugely as to how clean, powerful and stable their USB power is. I hear that he has done a lot to isolate it but the testing needs to include many PCs as to give us confidence that he has conquered the variations there. Or at least a simulation of the same.

Long way of saying, this audio thing can be quite complex. Positions we take are only as valid as our knowledge level. A person who has convinced themselves of being an objectivist (and I don't mean you), based on just reading forum posts and their gut feeling, many indeed not realize that they are taking too simplistic view here. Guys in car hobbies become experts at what they are messing with. I know my two boys are. In audio though, I find a tendency to not want to learn what the engine does, yet have strong opinion about its performance specs not mattering. I think that is a mistake on the part of objectivists. If anyone needs to become an expert in technology, it should be them. The other camp is going by what they hear so they don't have that requirement. Yet, I find self-appointed forum objectivists to be dismissive of technology and its design to incredible levels of dumbing it down to, "do you have an audio test? If not, I don't believe it. And no, I don't need to be an engineer or know what the design is to be right." That doesn't sit well with me. I think it gives the whole position a bad name. It really does.

Click to expand...

Absolutely, amirm, if somebody is going to take up the position of objectivist or measureist, then at least it is expected that they should not have a simplistic approach to matters but rather a more sophisticated, understanding of what the measurements can & cannot show. Dumbing things down to mantras & slogans is doing everybody a dis-service

Phelonious Ponk said:

I mean no offense, jkeny, but you are awfully demanding of data for someone who has, as far as I can tell, failed to show any of his own. Have you backed your product up with the kind of testing you're demanding here? Have you provided better evidence in support of your positions than Ethan has provided to support his? Let's see it.

Tim

Click to expand...

Tim, what you fail to comprehend with every post of mine, whether intentionally or not, is that what I'm essentially saying is that if you are going to use measurements to support your case, at least be comprehensive & accurate in your usage of them.

Maybe I haven't produced any results because:
- I don't believe we have the measurements yet that show what is really going on in some of the areas of audio that I'm concerned with?
- maybe I trust my ears & those of people I know (& lot's I don't) to evaluate what sounds best. Please don't call for a DBT to prove my lack of expectation bias - it'a another abused term & if you really knew how difficult & costly it is to set up a scientifically acceptable DBT, you wouldn't call for one. I've no problem about informal blind tests & yes I have done a number of them on myself & on friends!

Edit: Don't start with the Ethan thing again - I can show you a vast number of times where Ethan has ducked the perfectly valid & correct explanation given to him, ignored it & continued to make simplistic arguments. In these cases he does a dis-service to the scientific approach. You know just having measurements even though they are misguided or wrong does not help anybody, in fact one could argue it's setting things back.

Some of these posts were from amirm, some from Jneutron, some from me - I'm sure there are lots of other examples - please don't hold Ethan up as the model of what to aspire to! I find his approach less than scientific.

It's very hard to imagine that a DAC with noticeable pre-ringing (whether minimum-phase filtering or not) is truly audibly transparent.

I'm pretty sure the industry's level of knowledge regarding correlation of measurements with audible perception is incomplete (a position I know Sean Olive agrees with).

Meaningful audio DBT's are very difficult to design and perform, as has been repeatedly noted. In addition to the problems with adequate numbers (of both trials and listeners) common to scientific studies of any type, using an audio DBT to try to test for more than 1 or 2 specific sonic characteristics is difficult to the point of impossibility except for gross sonic differences (as anyone who has participated in one knows).

amir's first post is an excellent statement of most problems with the current "objectivist" position.

FrantzM said:

microstrip

Would you venture a guess as to what an "objectivist" is? (...)

Click to expand...

FrantZ,

Not easy to summarize. Formally it should be some one who makes his opinions and takes his decisions based on objective data, that means literally independent of mind. Confusion can arise immediately here, as most people will immediately say that perception of reproduced sound is dependent of mind. However, in my view, independent of mind in this case means data that everyone accepts because it is reliable, repeatable, verifiable and was collected and analyzed in a way people trust.

For me an objectivist is someone who finds a proper and science based way of dealing with the subjective aspects of sound reproduction, and using knowledge of perceptual processes, including surely a deep knowledge of statistics, manages to convert this information into objective data. This surely means knowing and accepting the constrains and limits of his method.

IMHO this is much beyond the capabilities of hi-end consumers. It is lifetime work and for me sound reproduction is supposed to be enjoyment, not research.

The easiest form of being objectivist is just reductionism - considering that all we can not embrace in a simple model does not matter and live with it. I could not find such an acceptable model or system for me. Some people pretend that successive states of disillusion can make it easier to accept in the long term. Not yet my time!

BTW, most of us disagree with the ultimate objective (meaning goal, in the different sense of the word) of sound reproduction. And this explains most of our debates, much more than being an objectivist or a subjectivist.

Apologies for the typos or mistakes, no time to re-read this posting before hitting Submit Reply!

The problem with all this is the subjective nature of science which isn't being admitted to. Let's face it everybody, science is not as objective as is being made out here. What to measure? How is this decided? What to research? How is this decided? What results are discarded? How is this decided? etc.