Please find the original disc sitting on the top of my opamp test box. It is not for sale, though.
But the box might be after all of this
It has all been very interesting. We all voted, we all had our say, and it is really everything that happened before the actual devices were disclosed that is the important part.
I wonder if a mono file would prove anything. I'm thinking also of two buffers in series, perhaps each with a modest loading of say 3k3 and just two devices, say 1458 vs 4562.
That could be interesting.
I love British customs and traditions
Go and have a cuppa Pavel
(this test makes you wonder on all the 'hi end' stuff that has device type numbers sanded off them
)
First, I like more the music with the "new" J River Media 22.0.110 than foobar2000. With the old JRM 21 not.
I used the foobar2000 1.3.15 in the all tests.
Now, with JRM 22.0.110 and my first an second choice (uu - TL072, ss - OPA2134) and the new ww (original file) the best is the ss - OPA2134.
Why? I hear HF distorsion in the uu - TL072.
PS: the best, in the sense that it is the most similar to the original.
I used the foobar2000 1.3.15 in the all tests.
Now, with JRM 22.0.110 and my first an second choice (uu - TL072, ss - OPA2134) and the new ww (original file) the best is the ss - OPA2134.
Why? I hear HF distorsion in the uu - TL072.
PS: the best, in the sense that it is the most similar to the original.
Hi Mark,
Has it not crossed your mind that Karl and Pavel may have been acquaintances or even friends for years/decades. There are people on this forum who have known each other for over 40 years. The fact that this then leads to friendly banter does not mean anything other than they know each other
you posting the link to the usage
I actually find offensive.
Did you run any diff maker tests on these, and if so did thay show any differences on this latest test ?
I've always felt with things such as opamp swaps that you have to listen for an extended time, as you would for serious listening for pleasure. I wonder whether differences would show under those conditions.
I've always felt with things such as opamp swaps that you have to listen for an extended time, as you would for serious listening for pleasure. I wonder whether differences would show under those conditions.
rr - LM4562 versus ss - OPA2134 and ww - Original
I like more the sound of ss - OPA2134. The rr - LM 4562 has a little more LF and the ss - OPA2134 more HF, more detail.
And yes, the ss - OPA 2134 sounds more like ww - Original.
The rr - LM4562 has minor "life".
My conclusions:
I like more the sound of ss - OPA2134. The rr - LM 4562 has a little more LF and the ss - OPA2134 more HF, more detail.
And yes, the ss - OPA 2134 sounds more like ww - Original.
The rr - LM4562 has minor "life".
My conclusions:
- The original SMPS of my iFi iCAN is a bottleneck. Maybe the iFi iPower or a good linear PSU.
- JRMC 22.* is more "musical" and detailed than foobar2000.
- I love my tweaked second system: ODAC + AV Marantz SR4500 + KEF Q100 with JRMC 22.* and with Bug head player.
Last edited:
Talking about PSU alternatives:
Power Supply (8+) Group Test, LPS and SMPS
-> https://www.computeraudiophile.com/...ply-8-group-test-lps-and-smps/#comment-639437
Power Supply (8+) Group Test, LPS and SMPS
-> https://www.computeraudiophile.com/...ply-8-group-test-lps-and-smps/#comment-639437
Yes, I did. I did both direct subtractions and diff maker extractions. Direct subtraction may be tricky because time differences between the files are from 4us to 17us. So I am posting the statistics from diff maker files.
In the first row there are opamps vs. original file. You can see that statistical differences are very similar for all opamps (072 not shown but no difference).
In the second row there are opamps vs. opamps. You can see that 4562-1458 is a little bit below the opamp-original, but not very much. The last image is 2134 1st take - 2134 2nd take. This is a very important result and it shows very good repeatability of recordings. Though the timing, time difference between the samples was about 4us between 1st and 2nd take files, the average RMS difference is about -90dB. No one will hear -90dB signal of any kind below the basic signal.
A comparison with the previous test would not be precise. In the first test, I did run DAC and ADC in the same audio device. The advantage was synchronous timing, disadvantage worse DAC in that test. In this test, DAC and ADC were separated instruments, higher output level (2Vrms), better DAC, but asynchronous sampling. However, repeatability is very good as you can see from ss-vv.
Attachments
Last edited:
Thanks for the diff maker files. I've never studied those before (as in used the program).
Hmmm ! I can see they are all close, and as you say particularly so for the 2134vs 2134 and yet I have doubts
These are taken over the whole length of the track I am assuming. So an average ? If so then a lot could be happening at the miniscule level that may not register.
Or have I not grasped the idea correctly ?
Hmmm ! I can see they are all close, and as you say particularly so for the 2134vs 2134 and yet I have doubts
These are taken over the whole length of the track I am assuming. So an average ? If so then a lot could be happening at the miniscule level that may not register.
Or have I not grasped the idea correctly ?
Yes, it is a 50ms window average through the whole track. But the images are not from Diff Maker. Diff Maker was just used to prepare the difference files. The evaluation windows are from another software, similar to Cool Edit 2000. But you can also see the instantaneous maximum deviations in those window images.
Regarding "sound difference" between vv and ss, or other files, this is exactly the stage where I would ask for the proof like valid ABX result. Otherwise, I am afraid, we may be subjects of a self-bias constructed be repeating of attempts and self-assuring us we hear a difference. Let's go Olive or Mason, Brookes, Pike. I understand that now I am starting to be unpopular.
Regarding "sound difference" between vv and ss, or other files, this is exactly the stage where I would ask for the proof like valid ABX result. Otherwise, I am afraid, we may be subjects of a self-bias constructed be repeating of attempts and self-assuring us we hear a difference. Let's go Olive or Mason, Brookes, Pike. I understand that now I am starting to be unpopular
.
Last edited:
No, but I have results of opampX - opampY which is more or less the same. You can see 4562 - 1458 and 2134(1) - 2134(2).
The most important in original vs. ad da chain comparison is a capacitive coupling and LF roll off resulting in a small phase shift at LF.
=============
Anyway, original file was posted yesterday and it is the ww file. If anyone feels it sounds different from rr, ss, tt, uu or vv, just show me your valid ABX protocol as a proof. Please let's stop speculations and start to act. Because speculations may lead to any level, audiophile speculations about golden fuses, wires etc. are well known but have never been confirmed by a scientific method.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.