ClaudeJ1 Posted October 28, 2021 Share Posted October 28, 2021 On 8/13/2021 at 2:14 AM, Langston said: I used a balanced switched resistor passive attenuator to bring down the NC400's within 0.29dB of the AHB2. I recall from the inventors of AB/X testing (44 years ago) that the REQUIREMENT was 0.1 db or less to make the test valid. Otherwise the louder one will be more detectable. Also, as a reminder, the test is NOT designed to say which one sounds better or is perceived to sound better, or worse, but ONLY to detect the existence of a DIFFERENCE between 2 things. Nothing more. Quote Link to comment Share on other sites More sharing options...
Edgar Posted October 28, 2021 Share Posted October 28, 2021 3 minutes ago, ClaudeJ1 said: I recall from the inventors of AB/X testing (44 years ago) that the REQUIREMENT was 0.1 db or less to make the test valid. I have always struggled with that requirement. If the two items under test have different frequency responses, then should the levels be matched at one frequency, at a set of frequencies, using band-limited pink noise (what should the limits be?), using full-bandwidth pink noise, white noise, etc.? 1 Quote Link to comment Share on other sites More sharing options...
ClaudeJ1 Posted October 28, 2021 Share Posted October 28, 2021 2 hours ago, Edgar said: I have always struggled with that requirement. If the two items under test have different frequency responses, then should the levels be matched at one frequency, at a set of frequencies, using band-limited pink noise (what should the limits be?), using full-bandwidth pink noise, white noise, etc.? I have no struggle with it for electronics because they are supposed to be ruler FLAT from 20Hz-20Khz. Sine Wave are what music is all about. Pink and white noise jump all over the place. Many wideband designs are flat to 100 Khz. Beyond this, it's how they INTERACT with the non linear Loudspeaker LOAD that will cause a difference, if any. One variable at a time is what we use to get to the truth. So, yes, 1 Khz. is a good place to start, or 440 if you prefer to hit square in the center of all music. 2 Quote Link to comment Share on other sites More sharing options...
Edgar Posted October 28, 2021 Share Posted October 28, 2021 9 hours ago, ClaudeJ1 said: Beyond this, it's how they INTERACT with the non linear Loudspeaker LOAD that will cause a difference, if any. That is the crux of the problem, and the basis for my struggle. 1 Quote Link to comment Share on other sites More sharing options...
Langston Posted October 28, 2021 Author Share Posted October 28, 2021 Level matching is a big deal if the amps being compared sound similar, which will almost always be the case unless something is broken. As Greg infers, it's impossible to meet Claude's spec using a single frequency tone - or any other stimulus type for that matter. What would have to be done is to place a fancy 2in/4out EQ in front of the amps to remove these linear distortion differences. Non-linear distortion differences, if any, are helpful differentiators and shouldn't (and can't) be removed. The proper measurements would be at the loudspeaker input terminals. The difference in level and magnitude response (phase ignored) would then be equalized throughout the spectrum of interest within the tolerance of interest using standard minimum phase filters and gain. Then you run the ABX test on a different day when you've recovered from doing all the prep! : ) The following plot isn't an illustration of the above, but it makes the point that different amps are going to have different responses into a variable impedance load. After restoring a MC275CE tube amp I measured the output into an 8Ω resistor and then into a Danley SM100 loudspeaker. The measurement was taken at the amp output terminals. God bless you and your precious family - Langston Quote Link to comment Share on other sites More sharing options...
Edgar Posted October 28, 2021 Share Posted October 28, 2021 12 minutes ago, Langston said: The following plot isn't an illustration of the above, but it makes the point that different amps are going to have different responses into a variable impedance load. It rapidly becomes a circular argument. If the frequency response differences are so great that it matters, then you are tacitly admitting that the devices under test sound different ... which is the objective of the ABX test. And if you equalize them so that the responses are similar, then you are removing the very differences that the ABX is designed to expose. 1 2 Quote Link to comment Share on other sites More sharing options...
Langston Posted October 28, 2021 Author Share Posted October 28, 2021 19 minutes ago, Edgar said: if you equalize them so that the responses are similar, then you are removing the very differences that the ABX is designed to expose. Very true, but IMO it's a good thing to eliminate the loudspeaker impedance variable from the amp shootout so that the difference between the sound of the amps alone remains, if any. Not to eliminate the loudspeaker magnitude response differences (caused by amp output impedance differences) would make the test results applicable only to the amps under test into the specific loudspeakers used in the test. IMO. : ) Edit: I think I see your point now. If one were doing an amp shootout in their home with their loudspeakers, neither of which they planed to replace or modify, then you're absolutely right. I'm always thinking about other folks using the results of my foolishness measurements. : ) Quote Link to comment Share on other sites More sharing options...
Edgar Posted October 28, 2021 Share Posted October 28, 2021 @Langston It's a systems view vs. a component view. It has long been asserted that some perfectly good components just don't work well in some perfectly good systems. I tend to agree with that idea. 2 Quote Link to comment Share on other sites More sharing options...
ClaudeJ1 Posted October 28, 2021 Share Posted October 28, 2021 3 hours ago, Langston said: Very true, but IMO it's a good thing to eliminate the loudspeaker impedance variable from the amp shootout so that the difference between the sound of the amps alone remains, if any. Not to eliminate the loudspeaker magnitude response differences (caused by amp output impedance differences) would make the test results applicable only to the amps under test into the specific loudspeakers used in the test. IMO. : ) tith their loudspeakers, neither of which they planed to replace or modify, then you're absolutely right. I'm always thinking about other folks using the results of my foolishness measurements. : ) this makes the exercise more real world when you drive the same speakers with 2 amps under test. Load interaction is part of the puzzle being solved. You should not try to eliminate it because it introduces another variable. Again, the goal is to find out: "Does and Audible DIFFERENCE exist," no more no less. It is a PRIMARY test after all. The RED curve is the Winner, the BLUE Curve needs to be for the B amplifier!! Maybe the RESISTOR should be used to set the Gains of the 2 amps under test to that +/-0.1 db level on the ends of the speaker wires, then just the test loose between the 2 amplifiers with whichever speakers you have. When I attended the Original Demo for the AES, my much younger ears were involved. They took 2 different speakers as A and B and with the same amplifiers driving them. Proving that an ABX test of speakers is just too easy, only requiring a few trials since the differences in speakers are 1,000 times greater than the 2 Pre Amps we also tested for the big DEMO. Those were really tough to tell apart. Quote Link to comment Share on other sites More sharing options...
CoryGillmore Posted October 28, 2021 Share Posted October 28, 2021 22 hours ago, ClaudeJ1 said: No you don't. Honestly I was surprised he got as many correct as he did even if it wasn't the proper control amount. Quote Link to comment Share on other sites More sharing options...
ClaudeJ1 Posted October 28, 2021 Share Posted October 28, 2021 4 minutes ago, CoryGillmore said: Honestly I was surprised he got as many correct as he did even if it wasn't the proper control amount. Now you know why. The NC-400 by his own admission is LOAD INDEPENDENT. The other must not be. 1 Quote Link to comment Share on other sites More sharing options...
CoryGillmore Posted October 28, 2021 Share Posted October 28, 2021 2 minutes ago, ClaudeJ1 said: Now you know why. The NC-400 by his own admission is LOAD INDEPENDENT. The other must not be. Does that mean it likely wasn't properly volume matched? Quote Link to comment Share on other sites More sharing options...
ClaudeJ1 Posted October 29, 2021 Share Posted October 29, 2021 1 hour ago, CoryGillmore said: Does that mean it likely wasn't properly volume matched? No it just means that it might explains why he could tell them apart on the ABX test. And why BEFORE the ABX test, he still preferred the Benchmark on his Khorns. ABX is not about preference, it's about repeatable statistical probability that a DIFFERENCE does exist when level differences have been minimized. Otherwise it's too easy to pick the louder one. 1 Quote Link to comment Share on other sites More sharing options...
ClaudeJ1 Posted October 29, 2021 Share Posted October 29, 2021 9 hours ago, Langston said: Edit: I think I see your point now. If one were doing an amp shootout in their home with their loudspeakers, neither of which they planed to replace or modify, then you're absolutely right. I'm always thinking about other folks using the results of my foolishness measurements. : ) I think your modification, which lowered the gain AND increased the signal to noise ratio on the Hypex NC-400, proves out Bruno Putzeys Motto about that amp: "Neither dirt nor fairy dust." A Laboratory Grade amplifier that is deemed load independent should perform exactly as claimed, but not necessarily the sonically preferred during long term listening in a particular room with particular speakers. Quote Link to comment Share on other sites More sharing options...
CECAA850 Posted January 13, 2022 Share Posted January 13, 2022 Well, I'm late to the party but that's nothing new. I've pretty much been a class AB guy all my life or at least since I've been into audio which has been a while. I can't comment on the original ncore amps but i recently purchased an amp with the Purifi modules. Im absolutely in love with it. The way it controls the woofers is nothing short of amazing. I wasn't expecting that. The micro detail it reveals it the recordings was a nice surprise as well along with the dead silent background. I'm impressed to say the least. No way I'd characterize the sound as dry or sterile. The amp came with the option to return for a refund. No way I'm sending it back. 3 Quote Link to comment Share on other sites More sharing options...
NBPK402 Posted January 13, 2022 Share Posted January 13, 2022 20 minutes ago, CECAA850 said: Well, I'm late to the party but that's nothing new. I've pretty much been a class AB guy all my life or at least since I've been into audio which has been a while. I can't comment on the original ncore amps but i recently purchased an amp with the Purifi modules. Im absolutely in love with it. The way it controls the woofers is nothing short of amazing. I wasn't expecting that. The micro detail it reveals it the recordings was a nice surprise as well along with the dead silent background. I'm impressed to say the least. No way I'd characterize the sound as dry or sterile. The amp came with the option to return for a refund. No way I'm sending it back. From what I have heard the class D amps are best at bass. Quote Link to comment Share on other sites More sharing options...
CECAA850 Posted January 13, 2022 Share Posted January 13, 2022 Just now, NBPK402 said: From what I have heard the class D amps are best at bass. I had no idea that was the case but there was seemingly more output from the bass bins of my LSIs. 1 Quote Link to comment Share on other sites More sharing options...
babadono Posted January 13, 2022 Share Posted January 13, 2022 Link to what you bought? Please. @CECAA850 Quote Link to comment Share on other sites More sharing options...
ClaudeJ1 Posted January 13, 2022 Share Posted January 13, 2022 18 minutes ago, NBPK402 said: From what I have heard the class D amps are best at bass. The mids and highs are pretty great too! 2 Quote Link to comment Share on other sites More sharing options...
Guest Vahorns Posted January 13, 2022 Share Posted January 13, 2022 " ...an amp with the Purifi modules...." - is that one of the VTV units? Stereo or mono block? Any info. on what pre pro used with this unit ? Any further info. on this unit and how it is used would be appreciated. I am in the market for a power amp and pre pro. Thanks. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.