If you are biased, how can you tell the quality of your decision? What is the difference with my 5 year-old kid doing the vote?
In my DBLTs, ....
This would get better valid data out of the test. Please feel free to slip in some repeat presentations to weed out the deaf Golden Pinnae
You should set the objective first before defining the process. What do you expect from the result?
Member
Joined 2006
I was thinking of doing the big unveiling later today as interest seems to be waning now.
I think you'll find some surpises
Looking forward... I like surprises
Sure there will be lots of debates and discussions afterwards
That's why I'm not doing the test. Everyone who has read your comments will have been biased too.
- I'd hoped to see if there were people who could reliably tell the difference between OPAs.
- If 'yes', I wanted to see if there was a consistent preference for particular OPAs from these true golden pinnae
The test can't provide an answer to the first question so can't provide good data for the second. But given that, if there was a consistent (?!??) preference for certain OPAs, this is still data ... though much less valid than if there was a 'yes' answer to the first question.
But with rankings posted by several people, all validity is lost.
I'm still interested in the results cos I can extract carefully selected bits to support my own prejudices and evil attempts to influence people.
But I wouldn't use the 'data' to design better sounding stuff.
Last edited:
I put quite a lot of listening effort into this but only on my mp3 player (colorfly c3+B&W c5 earbuds).
I find it very hard to tell a difference between the files. I have occasionally looked at the player to see what track was playing when I found the quality to be less bad than average and found kiwi or banana on. I doubt I could pick these out with statistical significance though. I noticed a lot of distortion in the vocals at the end of the first sample on most (all?) of the tracks and in general the sound is warm, noisy and lacking in resolution. The violin section is more or less awful on all of the samples. I didn't get any significant kind of imaging or sense of an instrument or voice jumping out of the darkness - there is no darkness. The complex sections are pretty blurred.
Is the poor quality of the sample the elephant in the room?
I think my overall result is that I think the difference between the tracks is smaller than the difference between the average quality of the tracks and what I would consider to be a good recording to do this kind of test with.
So I'm not going to vote for any of the op amps because I don't like any of the samples.
I was tempted not to comment because saying this raises the embarrassing possibility that I might be going deaf but I do have some music on my mp3 player that does actually sound good...
I find it very hard to tell a difference between the files. I have occasionally looked at the player to see what track was playing when I found the quality to be less bad than average and found kiwi or banana on. I doubt I could pick these out with statistical significance though. I noticed a lot of distortion in the vocals at the end of the first sample on most (all?) of the tracks and in general the sound is warm, noisy and lacking in resolution. The violin section is more or less awful on all of the samples. I didn't get any significant kind of imaging or sense of an instrument or voice jumping out of the darkness - there is no darkness. The complex sections are pretty blurred.
Is the poor quality of the sample the elephant in the room?
I think my overall result is that I think the difference between the tracks is smaller than the difference between the average quality of the tracks and what I would consider to be a good recording to do this kind of test with.
So I'm not going to vote for any of the op amps because I don't like any of the samples.
I was tempted not to comment because saying this raises the embarrassing possibility that I might be going deaf but I do have some music on my mp3 player that does actually sound good...
You are correct here. Conrad Hoffman has mentioned it. And I have guessed even before I downloaded the files. I didnt even find a need to download the wav files.
Yes I'm curious too why the files are so terrible.
Today all being well. When I can relax and devote the time to re-read all your replies which in some cases are bang on the money
I think to do a good job of this test you need a loopback path with low noise and low distortion, ideally no op-amps in the path and you really, really need the DAC and ADC to be running from the same master clock.
Starting with a truly great sample would be good too. Something from linn maybe - a lot of their studio masters really stand out as good low noise recordings with good imaging.
I don't have a dog in this race, but I am amused to see that there appears to already be some attempts to discredit the test before the unveiling. Sadly, at least one serious attempt to ruin the test. I suppose excuses may carry more weight if they appear early?
One thing puzzles me: why bother with MP3? Surely even a good MP3 coder will fall seriously short of an old opamp in sound terms? The results ought to include knowledge of which format was used. Oops - have I just discredited the test too?
One thing puzzles me: why bother with MP3? Surely even a good MP3 coder will fall seriously short of an old opamp in sound terms? The results ought to include knowledge of which format was used. Oops - have I just discredited the test too?
Let us begin
Where is Jay
Right from the start you kept finding the "Banana" musical... and that is in fact the "direct" file with no opamps. Its just a pass through the A-D conversion process.
I think you liked the "Orange" at first but noted it was different, and in fact it came to light later that the L and R were reversed due to an error on my part.
Then here was a bit of dithering
The Orange was the OPA134. You liked that one I think.
Where is Jay
Right from the start you kept finding the "Banana" musical... and that is in fact the "direct" file with no opamps. Its just a pass through the A-D conversion process.
I think you liked the "Orange" at first but noted it was different, and in fact it came to light later that the L and R were reversed due to an error on my part.
Then here was a bit of dithering
The Orange was the OPA134. You liked that one I think.