Quote:

sphinxvc said:

Quote:

+1
 
How many have actually done the ABX, volume matched tests?  Or even seen results from a decent sample?  I would think that's important before you give yourself up to straddling V's "position."  

Click to expand...

 
LOL. Funny how the scientific method gets thrown out the window with one view but needs to be applied rigorously with the other. "Objectively", nothing can be said either way at this time.
 
 

 
Quote:

purrin said:

 
 
LOL. Funny how the scientific method gets thrown out the window with one view but needs to be applied rigorously with the other. "Objectively", nothing can be said either way at this time.
 
 

Click to expand...

No, not at all. Claiming that nothing can be said either way is simply bollocks and a horrendous distortion of the burden of proof.
 
There are recognised numbers for the audibility of measurable imperfections. The O2 keeps everything at least 86db below the signal and the distortion is generally of a relatively benign nature. 
There are established thresholds of audibility. The O2 takes the stricter ones, makes them a bit more paranoid for good measure and then makes sure everything is well under them.
When everything that could possibly affect what you hear to the best of our current knowledge is at the very least 86db below the music and generally considerably more (THD at 1khz is about 20% of that) there is absolutely no reason to expect them to be audible, by established science. NwAvGuy hardly broke new ground here: this stuff has been spelled out in the AES archives years ago. 
 
The reason why blind listening is asked for when people declare other amps that it is likely measure well sound different. How does the burden of proof constitute double standards: one claim makes sense, the other less. You don't demand science proves itself simply because you reckon it might be wrong with absolutely no evidence at all. When someone suggests something that clashes directly with current scientific thought, however, it is considered a good idea to present at least a shred of semi-credible proof and casual listening impressions don't even get that far.

 
Quote:

willakan said:

The reason why blind listening is asked for when people declare other amps that it is likely measure well sound different. How does the burden of proof constitute double standards: one claim makes sense, the other less. You don't demand science proves itself simply because you reckon it might be wrong with absolutely no evidence at all. When someone suggests something that clashes directly with current scientific thought, however, it is considered a good idea to present at least a shred of semi-credible proof and casual listening impressions don't even get that far.

Click to expand...

Exactly, so where is the proof - objective measurements showing the O2 sounds the same as the discrete component amps such as the B22, Dynalo, etc? Keep in mind we haven't seen the more interesting measurements, i.e. full spectrum non-linear distortion measurements at specific frequency sets (or multiple frequency sets) for the B22 and Dynalo (or the O2 for that matter.)
 
All we've seen so far is that the O2 measures beyond what humans can hear in tests A, B, C, etc. This is hardly proof that the O2 sounds the same as discrete component amps (where no measurements have even been applied.) The only thing that can be inferred so far is that the O2 measures beyond what humans can hear in tests A, B, C, etc. - just that - nothing more (other than that it measures better than the mini3 in many of these tests.)
 
And finally, where's the science that says that the set of measurement tests applied to the O2 explains all aspects of human perception? In other words, are we measuring the right things?
 

Quote:

full spectrum non-linear distortion measurements at specific frequency sets

Click to expand...

 
Do you have a link to this test procedure, evidence this test will show meaningful data that other tests do not, and a quantity in regards to this test that will show an audible impact?  Preferably from a recognized source please.

 
Quote:

shike said:

 
Do you have a link to this test procedure, evidence this test will show meaningful data that other tests do not, and a quantity in regards to this test that will show an audible impact?  Preferably from a recognized source please.

Click to expand...

 
I love these tests (and full spectrum visualizations) because they show more and are better differentiator of non-linear distortion. In other words, as I've been saying, what the heck does 0.3% THD at 100Hz mean? With these types of visualizations, you see a lot more of what's actually going on. The disadvantage is that you need to run a lot of them and that can be a pain.

http://www.head-fi.org/t/566929/headphone-csd-waterfall-plots/120#post_7743752 (there are a few sets of these in the CSD thread)
 
http://www.audioheuristics.org/measurements/Testing/mdt33/mdt33_supreme110.htm (scroll down a bit to see the tri-tone tests)
 
I swear there was something similar on V's website, but I can't seem to find it. (EDIT: see the full spectrum THD and IMD graphs.)
 
But basically I'd like to see a full suite of these at different sets of frequencies with single tones, tri-tones (grouped closer together), or even more tones, i.e. ~5
 

Quote:

purrin said:

I swear there was something similar on V's website, but I can't seem to find it.
 
But basically I'd like to see a full suite of these at different sets of frequencies with single tones, tri-tones, or even more.

Click to expand...

There are graphs like that into different loads for single tone THD and SMPTE and CCIF IMD in the first O2 article.

There have definitely been plenty of multi-tone tests developed other than SMPTE and CCIF, many of which probably tell a little more interesting things about the performance than SMPTE or CCIF and may be more interesting.  But since those two are more established, that's more often than not what we get.  At least those cover kind of low frequencies as well as high frequencies.
 
But again, for reasons stated before, I would be pretty surprised if one device could ace SMPTE and CCIF tests and fall flat on its face with something else.  If you can find an example of a device that does, I'd be interested to see that.

 
Quote:

mikeaj said:

There have definitely been plenty of multi-tone tests developed other than SMPTE and CCIF, many of which probably tell a little more interesting things about the performance than SMPTE or CCIF and may be more interesting.  But since those two are more established, that's more often than not what we get.  At least those cover kind of low frequencies as well as high frequencies.
 
But again, for reasons stated before, I would be pretty surprised if one device could ace SMPTE and CCIF tests and fall flat on its face with something else.  If you can find an example of a device that does, I'd be interested to see that.

Click to expand...

I'm curious and I feel it's worth trying more than the two-tone SMPTE and CCIF - don't you think? Maybe something along the lines to the closely grouped tri-tone and penta-tone tests on Mark K's website.

Quote:

purrin said:

 

Exactly, so where is the proof - objective measurements showing the O2 sounds the same as the discrete component amps such as the B22, Dynalo, etc? Keep in mind we haven't seen the more interesting measurements, i.e. full spectrum non-linear distortion measurements at specific frequency sets (or multiple frequency sets) for the B22 and Dynalo (or the O2 for that matter.)
 
All we've seen so far is that the O2 measures beyond what humans can hear in tests A, B, C, etc. This is hardly proof that the O2 sounds the same as discrete component amps (where no measurements have even been applied.) The only thing that can be inferred so far is that the O2 measures beyond what humans can hear in tests A, B, C, etc. - just that - nothing more (other than that it measures better than the mini3 in many of these tests.)
 
And finally, where's the science that says that the set of measurement tests applied to the O2 explains all aspects of human perception? In other words, are we measuring the right things?
 

Click to expand...

The reason they *should* sound the same is because these amps have supposedly been designed around being wonderful at reproducing audio, not simply adding some pleasant colouration. The O2 is provably transparent to the best of our knowledge by virtue of the fact it excels in every known parameter. As for more complex tests, why would you expect they are needed to characterise performance? NwAvGuy has written moderately extensively about why he feels absolutely fine armed with relatively standard test-signals and what he has said is hardly new or revolutionary. That said, I'm sure he would at least discuss any test you would care to suggest on his blog that the O2 would not excel in.
 
As it stands, the current tests will be insufficient the day someone hears something that shouldn't be there under controlled conditions, in common with how the rest of science works.
 
Anyway, to the crux of the point, it is maintained that the uber-expensive discrete amps are transparent and wonderful, conveying soundstage and other such things with superb fidelity - try suggesting on the high-end forums that really high-end amps just colour the sound in a nice way. So far I'll admit I have been slightly inconsistently granting them the benefit of the doubt. I strongly suspect that some of them do not measurably excel.
 
Anyway, there are two possibilities:
 
1) The expensive discrete amps measure as well or better than the O2 and should (should in the context that I would happily bet lots on the outcome: if I lose I'll be too excited about the revolutionary new audio parameters discovered to care about my loss!) sound the same as it into loads they are designed to drive and volume-matched. If they are held to sound different this could easily be verified with a blind test.
 
2) The expensive discrete amps measure decidedly dubiously and it is entirely possible they sound different from the O2.
 
As I said, I have, so far, been assuming that these other amps measure reasonably well. It is entirely possible they do not, in which case we should ask the question of why people spend a fortune to accomplish what could be done with ease in the digital domain.
 

Quote:

willakan said:

The reason they *should* sound the same is because these amps have supposedly been designed around being wonderful at reproducing audio, not simply adding some pleasant colouration. The O2 is provably transparent to the best of our knowledge by virtue of the fact it excels in every known parameter. As for more complex tests, why would you expect they are needed to characterise performance? NwAvGuy has written moderately extensively about why he feels absolutely fine armed with relatively standard test-signals and what he has said is hardly new or revolutionary. That said, I'm sure he would at least discuss any test you would care to suggest on his blog that the O2 would not excel in.

Click to expand...

Well, I wouldn't say "every known parameter". Perhaps a more precise statement would be that the O2 excels with the RMAA suite, and SMPTE/CCIF distortion tests. I don't believe any of those standards ever state that performing well in those tests equates with transparency. I'm also pretty sure those standards were never designed for the purpose of measuring high-fidelity. The only thing that is proven is that the O2 measures well with those set of measurements.
 
I'm puzzled why you are against more complex tests that could possibly show more differences among different amps (there's even no guarantee of this.) As I've said, it doesn't hurt, and aren't you at least curious to see if we can push the boundaries of current measurements? The "Bah humbug! No need for such tests, V says that these existing ones are good enough!" attitude hardly sounds inquisitive and smacks of someone who wants to stay in his little box.

I want to know why you believe these tests will actually benefit finding transparency, and if these tests are needed to gauge it when ABX would be sufficient myself.
 
Would it not make sense to first prove audibility between them before assuming the tests even matter?

Can more complex tests be done with Voldemort's equipment? I think it's already been said that you could contact him.
 
Reversing the burden of proof and claiming the numbers don't mean anything doesn't seem like much of an argument to me, but I suppose it's the only way to be on the other side. . . 

 
Quote:

shike said:

I want to know why you believe these tests will actually benefit finding transparency, and if these tests are needed to gauge it when ABX would be sufficient myself.

Click to expand...

 
Measurements that provide more data and greater differentiation give us more opportunities to find correlations with human experience. If these tests are meaningful (they may not be), then they come in handy since ABX tests are much more difficult to conduct.
 
 

Would it not make sense to first prove audibility between them before assuming the tests even matter?

Click to expand...

About as much sense as for V to first ABX the O2 against the supposedly voodoo rip-off amp designs using discrete components before making claims and taking non-comparative isolated measurements of the O2.

Wait a second, the O2 was not tested with RMAA. It was tested with a $10k piece of specialised audio measuring equipment and he ran rather more tests than RMAA could ever hope to manage (including the aforementioned specialised IMD tests.) Have you looked at the blog? It measures distortion relative to frequency, phase, noise floor differences with batteries and AC, uber wide-band THD, slew rate...it is pretty comprehensive and the individual tests are run at a resolution far beyond anything RMAA could hope to pull off.
 
I'm still not quite sure what you're getting at with measurement correlation. The point of the O2 isn't to get numbers to specific ideal values, it is just to get rid of everything that isn't the audio signal as much as possible to the point where it is irrelevant to the listener regardless of hearing acuity. Who cares exactly how different distortion harmonics sound when you can't possibly hear any of them?
 
Also, I may be being quite astoundingly thick, but could you kindly rephrase the last line of your post for clarity?

 
Quote:

willakan said:

Wait a second, the O2 was not tested with RMAA. It was tested with a $10k piece of specialised audio measuring equipment and he ran rather more tests than RMAA could manage (including the aforementioned specialised IMD tests.) Have you looked at the blog? It measures distortion relative to frequency, phase, noise floor differences with batteries and AC, uber wide-band THD...it is pretty comprehensive.
 
I'm not quite sure what you're getting at with measurement correlation. The point of the O2 isn't to get numbers to specific ideal values, it is just to get rid of everything that isn't the audio signal as much as possible to the point where it is irrelevant to the listener regardless of hearing acuity. Who cares exactly how different distortion harmonics sound when you can't possibly hear any of them?

Click to expand...

I stand corrected - it's a Prismsound dscope running a suite of tests more comprehensive than RMAA.