AVS Forum banner

601 - 620 of 733 Posts

·
Registered
Joined
·
15,938 Posts
Discussion Starter #601 (Edited)
With no other changes, it's obvious to these 60+ year old ears, Su-8 delivers more imaging, soundstage and detail than SHD, opposite of my expectations.
Since you claim to hear differences in DACs' imaging, soundstage, and details I wonder if you might demonstrate that to us by reading the first post (which outlines the purpose of this thread) and performing the double blind test described, so you can then post your score sheet for us. Thanks.
 
  • Like
Reactions: WoodyTobiasJr

·
Premium Member
Joined
·
6,546 Posts
That's too much work, would rather shop for a second ES dac. Thanks for a thought provoking thread, truly wanted to believe they all sound the same, though it could be something other than dac chip that provides obvious perceived differences.
 

·
Registered
Joined
·
15,938 Posts
Discussion Starter #603 (Edited)
The whole point of this thread's test is that it is:
  • level matched precisely by instrumentation since small loudness differentials we don't even consciously perceive are known to be misconstrued as qualitative differences and nobody is "immune" to this despite their claims otherwise
  • synchronized so instant switching between A and B is possible for on the fly comparisons without fear small timing errors will give away the identities
  • allows the user to repeat any specific passage of the music they wish to focus on
  • has an optional training mode they can use pre-test to practice and refine their skills
  • the test is "open book" so the exact sounds of both A and B are fully disclosed to the listener at any time, [they just press a button to instantly hear A or B whenever they want] so they can constantly refresh their memory for the target sounds and don't have to rely on their memory of hearing it minutes ago
  • automatic statistical analysis is provided rather than just meaningless "I heard a difference" type results
  • double blind so we know that possible subconscious expectation bias can't influence the results
 

·
Registered
Joined
·
1,543 Posts
The whole point of this thread's test is that it is:
  • level matched precisely by instrumentation since small loudness differentials we don't even consciously perceive are known to be misconstrued as qualitative differences and nobody is "immune" to this despite their claims otherwise
  • synchronized so instant switching between A and B is possible for on the fly comparisons without fear small timing errors will give away the identities
  • allows the user to repeat any specific passage of the music they wish to focus on
  • has an optional training mode they can use pre-test to practice and refine their skills
  • the test is "open book" so the exact sounds of both A and B are fully disclosed to the listener at any time, [they just press a button to instantly hear A or B whenever they want] so they can constantly refresh their memory for the target sounds and don't have to rely on their memory of hearing it minutes ago
  • automatic statistical analysis is provided rather than just meaningless "I heard a difference" type results
  • double blind so we know that possible subconscious expectation bias can't influence the results
You are so fixated on your test you just cannot believe others testimony on there actual truth...you being a salesman most of your career I can see how this is hard to grasp...but your test isn't the right way about it as it in it it's self is bias...towards your own ambition...if not publish it and let all take it...not just us lowly forum surrogates on a simple forum to prove your simple minded bias that diffences do not exist.Have the test done by a professional then take it yourself...I would almost gander a guess even you might hear subtitle's in the audio...agree or is your test the end all be all because you came up with it and control it?

Sent from my MHA-L29 using Tapatalk
 

·
Registered
Joined
·
15,938 Posts
Discussion Starter #605
Have the test done by a professional then take it yourself...I would almost gander a guess even you might hear subtitle's in the audio..
I took the test myself and posted my results. You however have not.
 

·
Registered
Joined
·
1,543 Posts
I took the test myself and posted my results. You however have not.
A test done by your own hand...not a professional...test in a proper way if you want results and evidence...or publish your test for all...not just a handful of forum visitors...you seem to be after a personal result...not a professional one...my opinion of course.

Sent from my MHA-L29 using Tapatalk
 

·
Registered
Joined
·
15,938 Posts
Discussion Starter #607 (Edited)
A test done by your own hand.
Incorrect. The test administrator at hand is one's computer so there can be no administrator bias, which effectively makes the test "double blind" since the computer can't inadvertently reveal the identity of X by, say, smiling, sighing, etc., and the listener is under blinded conditions as well. When both parties are blinded (test subject and test administrator) we call it "double blind".


"In other words, if you want to obtain an accurate and reliable measure of how the audio product truly sounds, the listening test must be done blind. It’s time the audio industry grow up and acknowledge this fact, if it wants to retain the trust and respect of consumers."

- Dr. Sean Olive
 
  • Like
Reactions: WoodyTobiasJr

·
Registered
Joined
·
1,543 Posts
Incorrect. The test administrator at hand is one's computer so there can be no administrator bias effectively making the test "double blind" since the computer can't inadvertently reveal the identity of X by, say, smiling, etc., and the listener is under blinded conditions as well. When both parties are blinded, test subject and test administrator, we call it "double blind".


"In other words, if you want to obtain an accurate and reliable measure of how the audio product truly sounds, the listening test must be done blind. It’s time the audio industry grow up and acknowledge this fact, if it wants to retain the trust and respect of consumers."

- Dr. Sean Olive
Quoting olive is not a way to show your test has merit... publish it for all then I'll take you serious...until then this test is for your own ambitions and nothing more...prove me wrong.

Sent from my MHA-L29 using Tapatalk
 

·
Registered
Joined
·
15,938 Posts
Discussion Starter #609 (Edited)
Quiting olive is not a way to show your test has merit... publish it for all then I'll take you serious...until then this test is for your own ambitions and nothing more...prove me wrong.
As I understand it you think my test has no merit because it is not "published for all" yet the same could be said of nearly every other post in our forum.

If you don't like my test, that's fine, I don't care.

I'm here to discuss audio science. Not bicker. If you want to discuss specifics of the test you don't understand or have questions about, that I'll be glad to discuss or elaborate on, but arguing about my test being "published", which I've never claimed, is of no concern to me and I will not respond to that red herring.

As I mentioned earlier, the founder of Stereophile Magazine, J. Gordon Holt, published articles discussing the test protocol generically called "straight wire with gain/bypass test", usually a test applied to amplifiers, to see if when set to unity gain [signal level in equals signal level out, no boost] there is an audible change or if the unit is "perfect" to the listener's ear, aka "transparent". I did not invent this test.

"bypass test Directly comparing the output signal from a device with the input signal being fed to it, by putting the device into and then out of the signal path and observing the difference."
-J. Gordon Holt, Stereophile Glossary

The principle has even been used in some advertising, including this one for Quad amplifiers I posted. Read the copy:
3024681


P.S. I quoted Dr. Olive simply because of his great truism regarding blind testing. J. Gordon Holt in his later years came to understand it and agree as well, it would seem:

"Audio as a hobby is dying, largely by its own hand. As far as the real world is concerned, high-end audio lost its credibility during the 1980s, when it flatly refused to submit to the kind of basic honesty controls (double-blind testing, for example) that had legitimized every other serious scientific endeavor since Pascal. [This refusal] is a source of endless derisive amusement among rational people and of perpetual embarrassment for me..”

- J. Gordon Holt
 

·
Registered
Joined
·
15,938 Posts
Discussion Starter #610 (Edited)
@Ethan Winer, if I recall correctly, has done similar tests to the Quad test but with DACs, not amps! He did a whole big chain of conversions:
D to A to D to A to D to A to D to A to D to A to D to A to D to. . . etc. etc.

His test was to see how many conversions into and out of the digital domain it took before the degradation became audible.
 
  • Like
Reactions: CruelInventions

·
Registered
Joined
·
211 Posts
A test done by your own hand...not a professional...test in a proper way if you want results and evidence...or publish your test for all...not just a handful of forum visitors...you seem to be after a personal result...not a professional one...my opinion of course.

Sent from my MHA-L29 using Tapatalk
Quoting olive is not a way to show your test has merit... publish it for all then I'll take you serious...until then this test is for your own ambitions and nothing more...prove me wrong.

Sent from my MHA-L29 using Tapatalk
... and by not taking the test (the protocols for which are clearly laid out in the first post - reading is fundamental by the way) your comments bashing the test have no merit. I took the test, posted the results afterwards and failed, just as I expected. Oh yeah, I am also a professional. Just sayin’.
 

·
Registered
Joined
·
835 Posts
The MOTU name brings back many memories of recording stuff for college radio. Are you getting an M2/M4? I've been looking at those myself.
 

·
Registered
Joined
·
15,938 Posts
Discussion Starter #613 (Edited)
The MOTU name brings back many memories of recording stuff for college radio. Are you getting an M2/M4? I've been looking at those myself.
M4. As you may already know, it uses the same chips for both DAC and ADC but it has extra inputs and a few more niceties over the MOTU M2. Even if it turns out to be no better for my application than my Behringer UMC204HD used in this thread's test, I still like that its headphone jack is getting stellar objective reviews (the only kind that matters to me) including perfect channel balance tracking at all levels, high output with low noise into a wide variety of loads, and a state of the art output impedance of just .06 ohms :love:so one has no fears of any frequency response variations per load.


It isn't perfect and seems to have a slight "ESS hump" (as ASR calls it) but I concur with the reviewer in that I doubt it's audible and for the price the M2/M4 come amazingly close to price-no-object USB interfaces, even beating them in some regards.
 

·
Registered
Joined
·
15,938 Posts
Discussion Starter #615 (Edited)
How is this test not limited by the system you are using to play the test on?
It is, just like with all audio tests, which is why I keep recommending people use the very best system they can to listen to the test files, A and B.
 

·
Registered
Joined
·
15,938 Posts
Discussion Starter #616 (Edited)
That's one of the advantages to me posting this test publicly in an international forum of people into audio as opposed to running it on just a small, private group of random people.
  • It reaches a broad audience, young and old, both rich and frugal, etc..
  • they'd probably tend to be audiophiles/audio hobbyists who focus on accuracy, sonic details, and high fidelity (and hence they may have better skills to discern subtle differences than random Joes)
  • there are no geographic limitations
  • in theory some people performing the test may have some extremely expensive setups no typical laboratory could amass quickly and affordably
 

·
Registered
Joined
·
15,938 Posts
Discussion Starter #617 (Edited)
That's one of the advantages to me posting this test publicly in an international forum of people into audio as opposed to running it on just a small, private group of random people.
  • It reaches a broad audience, young and old, both rich and frugal, etc..
  • they'd probably tend to be audiophiles/audio hobbyists who focus on accuracy, sonic details, and high fidelity (and hence they may have better skills to discern subtle differences than random Joes)
  • there are no geographic limitations
  • in theory some people performing the test may have some extremely expensive setups no typical laboratory could amass quickly and affordably
^Plus in my test, where people download the files from the web and are provided the SHA-1 check sums (digital fingerprints) so they can verify on their own the files they watch me create in real time, provided in the authentication video, are indeed the ones they've downloaded, there's a distinct time advantage:
  • people can take a nearly unlimited amount of time to perform their own pre-training before conducting the test, listening to the material for hours, days, weeks, months, or whatever they desire
  • they can go at their own leisurely pace and take as many breaks as they want while doing the actual test too. They can even do part of the test, put their PC/laptop in sleep mode, and then recommence the test another day, no problem!
In a formal lab test we have to consider that time is money. A test conductor doesn't have the time to sit there for months waiting for the listener to become familiar with the material, nor take hours/days/weeks to finish the 16 question test itself, if that's what the listener insists is necessary.

Contrary to lies spread on the web, Foobar ABX tests have no time limits and the listener is free to leisurely switch as quickly or as slowly as they choose. They can optionally focus on some specific snippet(s) of the music they personally pick, or, conversely, they can take in the complete, uninterrupted work (or do both). It's all entirely their choice and their pace.
 

·
Registered
Joined
·
15,938 Posts
Discussion Starter #618 (Edited)
M4. As you may already know, it uses the same chips for both DAC and ADC but it has extra inputs and a few more niceties over the MOTU M2. . . . the M2/M4 come amazingly close to price-no-object USB interfaces, even beating them in some regards.
So my preliminary results show "excellent" performance in all parameters I measured. Woo-hoo!

I'm testing its unbalanced RCA connections. Its balanced (XLR and TRS) jacks may yeild even better results but my main concern is RCA performance because that's what most gear in my setup uses.

Theses are the results for the line in on the back and used in its raw state (no external DAW connected at the moment). I can't seem to raise its level optimally so I'm taking a hit in SNR performance with these jacks but that may be offset by the fact that these bypass the mic preamp's noise. So it may be an even trade. Not sure.

The drivers are hard to configure and I might have better results using ASIO modes in both play and record but my measurement system does not seem friendly to it at the moment.

These are the results for the combined "loop back" trip:
3025741


This full trip is: D, to its DAC, to RCA wire, to its ADC, cheap USB cable to my laptop's analyzer software (RMAA). The individual DAC and ADC can be expected to have better measurements if measured individually (usually how ASR does it). My results above are for the combined noise and distortion of the DAC and ADC used together with the signal traveling unbalanced (RCA wire).

Overall I am quite happy.
 

·
Registered
Joined
·
15,938 Posts
Discussion Starter #619 (Edited)
The Motu M4's headphone out loopback (its DAC, to its headphone amp, to its ADC, again using RCA ins) is even better, just hitting the illustrious 110 dB signal-to-noise level! [I got the ASIO drivers to work finally but they didn't really seem to matter unless the latency has improved, which I didn't bother to measure.]
3026340

That's a very good combined score of the DAC noise, plus the headphone amp noise, plus the ADC noise used in series (added together). Individual measurements of the DAC alone should yield an even better result but unfortunately I don't have a hardware analyzer to test it.
 

·
Registered
Joined
·
7,570 Posts
You are so fixated on your test you just cannot believe others testimony on there actual truth
*Their

You're saying a subjective, anecdotal account is more "true" than a computer-controlled test?
Quoting olive is not a way to show your test has merit... publish it for all then I'll take you serious...until then this test is for your own ambitions and nothing more...prove me wrong.
Sounds like someone is trying to justify an expensive purchase.
 
601 - 620 of 733 Posts
Top