Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: two hi rez vs redbook train wrecks in progress (Read 91325 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

two hi rez vs redbook train wrecks in progress

Reply #50 –
Quote from: Wombat on
Quote from: 2Bdecided on
It would be nice to re-run the test with unquestionably good down conversion, and a secure virtually tamper-proof ABX program. That would solve two problems.



Agreed. I'd like to see a audio file checksum feature added to foobar2000, for example.  Anybody can change a filename.

Quote
I was reading a bit on the AVS thread and to be honest i doubt a bit in armirm abx logs. That of course may only be me old fart loosing the faith in humanity more every day.


Don't feel pregnant. ;-)

Quote
Since even encrypted abx solutions can be cheated by some real time analysis i only see one solution.
Arny has to pilgrimage to armirm and do some live testing.


It's a big expensive trip from Detroit to Bellingham or wherever all the old Microsofties hide out.  Is someone going to raise a fund for me? ;-)

Quote
I'd happily donate some bucks if needed.


I think JJ lives around there, too so it wouldn't be all a waste.

two hi rez vs redbook train wrecks in progress

Reply #52 –
Quote from: Arnold B. Krueger on
It's a big expensive trip from Detroit to Bellingham or wherever all the old Microsofties hide out.  Is someone going to raise a fund for me? ;-)
We're up by the Canadian border in Bellingham, those Microsofties hang out around Redmond, I believe. It's just east, on the other side of the lake from Seattle. I'd love to drive down and participate though!

two hi rez vs redbook train wrecks in progress

Reply #53 –
It's impossible to make a bulletproof test you can't cheat on without a proctor watching your every move, in person.

As soon as the signal gets delivered to the test subject, the listener, through an analog headphone jack, they could then surreptitiously split the signal with a y-cord and feed half of it in to an ultra high frequency real time analyzer [there are even cell phone apps which would work].

"My wife is expecting so I need to leave my phone out in case she calls. You don't mind if I leave my phone out, within reach, do you?"

https://itunes.apple.com/us/app/ultrasonic-...d851069926?mt=8

two hi rez vs redbook train wrecks in progress

Reply #54 –
Quote from: Arnold B. Krueger on
I think JJ lives around there..

And knows exactly how to conduct a valid test, which is precisely why it will never happen. Shysters know the game.
Loudspeaker manufacturer

two hi rez vs redbook train wrecks in progress

Reply #55 –
You would need a cracker jack team of at least 3 people:

A. scientist to run the test
B. trained expert listener, a teenager, with 20-20k hearing, to listen for accidental tells left in the signal
C. magician to make sure no slight of hand was used to sneak in a shirt pocket spectrum analyzer, or see a reflection off a window seeing the scientist's control panel

two hi rez vs redbook train wrecks in progress

Reply #56 –
Quote from: Arnold B. Krueger on
I think JJ lives around there, too so it wouldn't be all a waste.

I stand on my words and will at least make a neat meal possible for him then with 50 bucks
What makes me a bit skeptical is that JJ once was employee of the proband at MS.

Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

two hi rez vs redbook train wrecks in progress

Reply #57 –
Quote from: Wombat on
Quote from: Arnold B. Krueger on
I think JJ lives around there, too so it wouldn't be all a waste.

I stand on my words and will at least make a neat meal possible for him then with 50 bucks
What makes me a bit skeptical is that JJ once was employee of the proband at MS.


Amusingly enough, the biggest antagonist at AVS used to be JJ's boss, as the malefactor is prone to say early and often.

two hi rez vs redbook train wrecks in progress

Reply #58 –
The good news is, looks like JA is now a big fan of (gameable) blind tests.
Hmmm, seems like cognitive dissonance doesn't ruin things, when the desired result is achieved. File corruption and/or cheating is ok too, no need to consider.

cheers,

AJ
Loudspeaker manufacturer

two hi rez vs redbook train wrecks in progress

Reply #59 –
Linking to amirm's ABX logs on avsforums. Jeeez...

Btw, anyone know what DAC he is using?
"I hear it when I see it."

two hi rez vs redbook train wrecks in progress

Reply #60 –
Quote from: xnor on
Linking to amirm's ABX logs on avsforums. Jeeez...

Btw, anyone know what DAC he is using?



I believe that many of his tests were run using the internal audio interface in a laptop.  Others may have been run on his home system.

two hi rez vs redbook train wrecks in progress

Reply #61 –
Quote from: xnor on
Linking to amirm's ABX logs on avsforums. Jeeez...

Yep, no chance an ex-Microsoft exec would figure out how to game computer audio files after several trials...excuse me, a good nights rest.
Especially one with a penchant for (and getting caught!) blatantly fabricating listening tests to bolster some woo belief. 

Quote from: xnor on
Btw, anyone know what DAC he is using?

Is it relevant to the question?
Better one would be, since both he and JA friends with someone (jj) quite capable of setting up a valid, robust test of the subject, why have neither pursued such an avenue? Hmmm, let me guess...
Instead, what we have is:
When one fails to identify an amp during a controlled ears only test, but become dissatisfied with it during long term uncontrolled viewing, the preferred optional explanation is "cognitive dissonance", the bane of blind tests.
However, when an ex-Microsoft exec/known fabricator seemingly passes a amateur online computer file test....well, there's your proof, see blind tests do work! 
The pseudo-subjectivist position in a nutshell. (Objective) Blind tests are useless...unless useful for the sales/woo agenda.

cheers,

AJ
Loudspeaker manufacturer

two hi rez vs redbook train wrecks in progress

Reply #62 –
Oh you're right, I wrongly remembered the files being in different formats. In that case, if the test was done honestly, he could still be one of those audiophiles that use broken DACs that exhibit big differences in frequency response with different sampling rates ..
Nevermind!
"I hear it when I see it."

two hi rez vs redbook train wrecks in progress

Reply #63 –
Here's is my undeniable proof:

Quote
foo_abx 1.3.4 report
foobar2000 v1.3.4 beta 4
2014/10/20 15:34:34

File A: D:\temp\Just_My_Imagination_A2.wav
File B: D:\temp\Just_My_Imagination_B2.wav

15:34:34 : Test started.
15:34:43 : 01/01  50.0%
15:34:47 : 02/02  25.0%
15:34:51 : 03/03  12.5%
15:34:54 : 04/04  6.3%
15:34:56 : 05/05  3.1%
15:35:00 : 06/06  1.6%
15:35:05 : 07/07  0.8%
15:35:08 : 08/08  0.4%
15:35:11 : 09/09  0.2%
15:35:14 : 09/10  1.1%
15:35:16 : 10/11  0.6%
15:35:19 : 11/12  0.3%
15:35:22 : 12/13  0.2%
15:35:24 : 13/14  0.1%
15:35:28 : 14/15  0.0%
15:35:30 : 15/16  0.0%
15:35:33 : 16/17  0.0%
15:35:35 : 17/18  0.0%
15:35:37 : 18/19  0.0%
15:35:39 : 19/20  0.0%
15:35:43 : Test finished.

----------
Total: 19/20 (0.0%)


... and I didn't even touch the audio data. 


Btw, since Peter doesn't seem to be interested in "improving" fb2k ABX logs, is ABC/HR for Java more tamper-proof?
"I hear it when I see it."

two hi rez vs redbook train wrecks in progress

Reply #64 –
Quote from: ajinfla on
Quote from: xnor on
Linking to amirm's ABX logs on avsforums. Jeeez...

Yep, no chance an ex-Microsoft exec would figure out how to game computer audio files after several trials...excuse me, a good nights rest.
Especially one with a penchant for (and getting caught!) blatantly fabricating listening tests to bolster some woo belief. 

Quote from: xnor on
Btw, anyone know what DAC he is using?

Is it relevant to the question?


Since some of the tests involved recordings with strong content in the 20-40 KHz range, the DACs freedom of IM in this range can be an issue.

Quote
Better one would be, since both he and JA friends with someone (jj) quite capable of setting up a valid, robust test of the subject, why have neither pursued such an avenue? Hmmm, let me guess...


I'd like to see that, but...


Quote
Instead, what we have is:
When one fails to identify an amp during a controlled ears only test, but become dissatisfied with it during long term uncontrolled viewing, the preferred optional explanation is "cognitive dissonance", the bane of blind tests.


I beg to disagree.

When one fails to identify either amps during a DBT amp comparison, but later on develops a preference for one or the other, the honest experimenter returns to the DBT to see if there are insights about the operational environment that have been developed or changed since the original DBT that may be relevant. 

IOW long term listening may have led to the discovery of recordings, speakers, playback levels, or listening rooms etc. that may possibly make a difference. 

Otherwise its just another duel between an evaluation with bias controls versus one that does not have bias controls.  Not exactly an apples-to-apples comparison.




two hi rez vs redbook train wrecks in progress

Reply #65 –
Quote from: xnor on
Here's is my undeniable proof:

Congrats, you win!

I wonder if any of these threads will be updated with your highly trained hearing results?
In any case, make sure to brag how you beat Amir with near perfect scoring on the first try. Be as smug and condescending as possible to the poor peasants who can't afford/hear $100k cable lifters. 

cheers,

AJ
Loudspeaker manufacturer

two hi rez vs redbook train wrecks in progress

Reply #67 –
I too have suggested cheatproofing the ABX (let equip fb2k with a key to sign it upon creation). But for hi-rez one should expect quite a few positive tests due to ultrasonics creating intermodulation distortion, right?

(What about "noising up" a 44.1? If the golden ears cannot tell that from 96/24 files, then perceptual noise substitution gives them the same effect for cheap ;-) )

two hi rez vs redbook train wrecks in progress

Reply #68 –
There's no reason to cheat to pass the ABX test on their test song Mosaic, at least; I know didn't. Simply pause the music at a particularly telling location and notice the alteration due to the time misalignment. I can clearly hear the guitar slap at the end of this phrase repeated loop of this riff I posted ends with a "cha" vs "chip" sound. Please tell me someone else here can hear it too! I could have made it even more obvious by selecting a different location but this was literally my first stab at it and I thought it's subtle nature was appropriate.

Did I make it too subtle?

https://www.youtube.com/watch?v=Eog82nV21_A

Gadzooks. I'm only just now thinking the reason nobody at AVS has said, "Yes, I hear it too" is because so many there, who have never used foobar in their life, let alone ABX, mistakenly think the guitar slapped strum sound at the end of the phrase repeated loop I posted  is a "glitch", NOISE that is, not actual music content. Argh!

two hi rez vs redbook train wrecks in progress

Reply #69 –
It's clearly audible and another problem.

I think the ABX component generally needs some reworking..
"I hear it when I see it."

two hi rez vs redbook train wrecks in progress

Reply #70 –
Quote from: ajinfla on
The good news is, looks like JA is now a big fan of (gameable) blind tests.
Hmmm, seems like cognitive dissonance doesn't ruin things, when the desired result is achieved. File corruption and/or cheating is ok too, no need to consider.

cheers,

AJ


My stars, who could have predicted *that* would happen?   

I'd be surprised if there wasn't already an editorial/article about it in Stereophile.

two hi rez vs redbook train wrecks in progress

Reply #71 –
Quote from: ajinfla on
Quote from: xnor on
Linking to amirm's ABX logs on avsforums. Jeeez...

Yep, no chance an ex-Microsoft exec would figure out how to game computer audio files after several trials...excuse me, a good nights rest.
Especially one with a penchant for (and getting caught!) blatantly fabricating listening tests to bolster some woo belief. 



Wow, nostalgia time for me.  What ever happened to Terry Montlick anyway?  He used to fight the good fight.

Quote
Instead, what we have is:
When one fails to identify an amp during a controlled ears only test, but become dissatisfied with it during long term uncontrolled viewing, the preferred optional explanation is "cognitive dissonance", the bane of blind tests.


Wait, was that a line being proffered by JA?  Or by Amir?  When I asked JA before an audience (of mostly confirmed 'audiophiles') why he didn't follow up his long-term amp acclimation with *another* DBT to nail its audibility down, his answer was something like, 'why would I do that'? 

I was, as they say where he comes from, gobsmacked.

two hi rez vs redbook train wrecks in progress

Reply #72 –
Quote from: Arnold B. Krueger on
Since some of the tests involved recordings with strong content in the 20-40 KHz range, the DACs freedom of IM in this range can be an issue.


But didn't he says he ran your IM tests, and his rig passed?  (I have trouble sorting through all the wreckage.) 

two hi rez vs redbook train wrecks in progress

Reply #74 –
To be fair, fb2k's ABX component is a tool for people who want to know what they can really hear. It's not designed as a bullet proof / tamper proof evidence generation tool.

We've been through this before. You can make it more secure (and for a test tool, someone probably should), but it still comes down to trust. If you can put a spectral analysis tool on the output, you can pass the test without even connecting any speakers or headphones.

Automatic time alignment, like automatic level matching, is a useful tool sometimes but not something I'd ever want mandating. It's only useful if people know what they're doing. I cringe when I see people enabling ReplayGain on near-identical files. It's more likely to hinder than help. (Especially EBU R128, which can "hear" ultrasonics perfectly - which is usually fine because they're usually too quiet to make any significant difference, and if they're loud you might want to turn them down to save your tweeters - but it'll mess things up in an ABX test.) Similarly automatic time alignment isn't going to help if the files drift out of sync, or have some complex phase issues. It'll help more often than it doesn't, but it's blindly applying it those times that it doesn't that will trip people up.

Cheers,
David.