No cliff it is not. That green is smoother than blue, it is not really critical. But if you want to find out the effect of the fatiguing sound, you can try this: Play green as your background music for 2 days.
I have slept 2m from my speaker playing blue. All day long. Full volume. No complaint. Green? 🙂
But I'm just trying to transfer what for me is obvious. I'm not on the lost. I feel blessed with my knowledge. You must know that feeling when you think what is the thing that is illogical here such that it can't be understood? Hehe I will try to let it go and mind my own DIY activity 🙂
For those who use the on-board sound chip, this is what you most probably get. Please compare to the files posted in #post3707076
http://www.diyaudio.com/forums/ever...est-your-ears-audio-chain-13.html#post3707076
http://www.diyaudio.com/forums/ever...est-your-ears-audio-chain-13.html#post3707076
Member
Joined 2006
I feel relieved after seeing the results 😀
There are only 12 opamps employed in my Pluto speaker system...@@"
There are only 12 opamps employed in my Pluto speaker system...@@"
@Jay
Hmmmm.. Some of that sounds like mad religious euphoria to me - but whatever turns you on!
Listening to the same snippet for two days 😱😱
Hmmmm.. Some of that sounds like mad religious euphoria to me - but whatever turns you on!
Listening to the same snippet for two days 😱😱
SY. Yes, I did do the Foobar ABX. On headphones I did not even proceed to X-Y because I simply could not tell the difference between A&B. But on speakers I could tell the difference - it was even more striking after I remembered to turn on the tweeters! 
At first they didn't seem much different, but after comparing both to my rip of the same song, it was easier. Funny about that - I needed to compare them to something else to hear the difference between them. The difference was slight, but Green was more rolled off, a little smoother. Blue was closer to my rip, but more ragged on top. I didn't like it. Once I got a handle on that, I could pick them ABX with about 75% certainty. Before comparing them to another rip, I'm not sure I could really tell them apart.
Heard without a direct A/B I doubt I could hear the difference. However, given the choice between the two system sounds, I'd take Green because it was smoother. Even tho I now know that Blue is direct, I still like Green.
Would be interesting to hear the master file, resampled to 48kHz. Would I be able to tell it from Blue? Maybe not.
At first they didn't seem much different, but after comparing both to my rip of the same song, it was easier. Funny about that - I needed to compare them to something else to hear the difference between them. The difference was slight, but Green was more rolled off, a little smoother. Blue was closer to my rip, but more ragged on top. I didn't like it. Once I got a handle on that, I could pick them ABX with about 75% certainty. Before comparing them to another rip, I'm not sure I could really tell them apart.
Heard without a direct A/B I doubt I could hear the difference. However, given the choice between the two system sounds, I'd take Green because it was smoother. Even tho I now know that Blue is direct, I still like Green.
Would be interesting to hear the master file, resampled to 48kHz. Would I be able to tell it from Blue? Maybe not.
@ pano,
how many trials did you do in the ABX, the usual 16 or another number?
@ PMA,
over the years we have discussed a lot of topics wrt controlled listening tests.
DBTs do not produce better results (or more correct/reliable results) per se, as there is a lot more to consider than "blinding", but if you want to know what is really going on, you can´t avoid scientific methods.
A/B - tests are a good choice; using AB BA and AA BB is called a same/different test, using AB BA only would be a discrimination test based on preference.
If there is no "no difference" answer it is called a 2-AFC (two alternatives forced choice paired comparison), if a "no difference" choice is possible (usually called a "tie" ) it is known as a 2-AC test.
As there might exist a bias for presentation order or even color, that should be randomized.
If you are not testing groupwise, but individuals instead, you must have at least 6 trials per listener in the case of an undirectional hypothesis (means "A better than B" and "B better than A" are both considered) on SL=0.05 .
how many trials did you do in the ABX, the usual 16 or another number?
@ PMA,
over the years we have discussed a lot of topics wrt controlled listening tests.
DBTs do not produce better results (or more correct/reliable results) per se, as there is a lot more to consider than "blinding", but if you want to know what is really going on, you can´t avoid scientific methods.
A/B - tests are a good choice; using AB BA and AA BB is called a same/different test, using AB BA only would be a discrimination test based on preference.
If there is no "no difference" answer it is called a 2-AFC (two alternatives forced choice paired comparison), if a "no difference" choice is possible (usually called a "tie" ) it is known as a 2-AC test.
As there might exist a bias for presentation order or even color, that should be randomized.
If you are not testing groupwise, but individuals instead, you must have at least 6 trials per listener in the case of an undirectional hypothesis (means "A better than B" and "B better than A" are both considered) on SL=0.05 .
No disagreement here. And, this is a diy/hobby forum, so probably we are not 100% rigorous. To me, it has been an interesting test, and once again I thank to all who have participated and wrote their impressions.
I probably sound really lazy but its hard to sort through all of the big words to find what I'm looking for.
And what difference did that produce in the sound?
I remember in the PM I said I first had a hunch the bass was slightly different but then later concluded that I couldn't tell the difference. I was listening with some average consumer grade sony headphones.
Since so many liked Green maybe you should give the people what they apparently want: 😉
All of PMA's OPA + BUF PCBs can be reconfigured to ... well, anything!
It's just the good bypassing and layout, what you put between the 627/637 and BUF634/LT1010 is up to you ...
😎
I think the phono PCB is the best bang-for-buck!🙂
Hey PMA, maybe it's time for another offering since so many like the actual sound 😀
PS Good to see the workload has dropped off and you can contribute again.
All of PMA's OPA + BUF PCBs can be reconfigured to ... well, anything!
It's just the good bypassing and layout, what you put between the 627/637 and BUF634/LT1010 is up to you ...
😎
I think the phono PCB is the best bang-for-buck!🙂
Hey PMA, maybe it's time for another offering since so many like the actual sound 😀
PS Good to see the workload has dropped off and you can contribute again.
Last edited:
What's confusing about giving marks out of 10 or ranking 3 'presentations' that MAY have one of them essentially repeated? They may all 3 be the same 😱
It's only 'confusing' if you THINK one of them is repeated and you are trying to guess which one. As I said, guessing is fatal when you are trying for the Zen state where your auditory faculties are most acute.
I have been awed by the perceptive powers of some of my true golden pinnae. But of course when you do these tests, you find loadsa deaf Golden Pinnae too.
Mooly, thanks for posting the comments. Most interesting but at present, they have NO significance. Their true value is when we repeat the test with ABC.
If he makes the same or similar comments, the listener is probably a true golden pinnae. If his preference changes, he's DEFINITELY a deaf Golden Pinnae.
Only when we sort this out, are the comments significant.
_______________
Mooly, PMA, SY, could we get together via email to plan the next evil series of tests? We are so close to getting significant and useful data .. even if it might be that our whole listening test panel is deaf 🙂
But we have to drop green/blue. In da last Millenium, I did an extensive series of DBLTs with a music college to tap into their large pool of interested listeners. They were told we were assessing Sound Quality but the research was actually about the Influence of Colour on Perceived Sound Quality.
As one of the 2 colours which showed significant advantage is in your names, we should drop using colour names .. though I'm not sure just using names of colours is similarly effective. AFAIK, ABC or 123 are sufficiently neutral.
And yes. This is NOT BS.
And fruit names too. I hate damsons. 🙂
Both the names and the order of presentation could influence perception. To average out the effect, the order and names of the files need to be randomized for each test, which is difficult to organize on a forum like this.
Hi kgrlee, I just realized that the whole idea of the BT is too advanced for some people. We are engineers and probably scientists (sometimes I work professionally too in the lab with oscilloscopes and other measurement devices).
When we do a research, we usually set the objective, a question that we will try to answer through series of experimentations.
Take one question example: “Who are here in DIYAudio the true GP and who are the deaf GP?”. It is very very easy to set the blind test process/protocol to solve the question. But of course, it is NOT what a scientist want to know the answer for. Scientists are after something more useful and usually more complex.
You might have been professionally hired to do a research that will help a company to decide on the product that they want to produce. It is part of quality assurance to predict/ensure the success of the product if it is released to the market.
Take simple example, if it is the green preamp that I want to produce, is it good enough to ensure BEP or success in the market from sound quality perspective? How can I answer that through a blind test? Who should be the listeners, and how to determine it? This is still simple, and you know it.
Harder question (may have nothing to do with blind test) is: how critical is the “fatiguing” phenomenon? What will be the symptoms? Will and how this could possibly affect the deaf listeners? Can I make a circuit that is not fatiguing? Can this fatiguing aspect be quantified? What physical parameters are in effect?
Why, is it not even in your mind?
Hehehe this is funny. I have mentioned previously that even SPL difference do not affect my decision. And now color? If the color issue is still a big problem, the whole idea could be too big to solve.
- Status
- Not open for further replies.