Mostly, but again context, which seems to still an issue.
If you just EQ a few dB's, that's fine, totally different story when EQ'ing more than 6-12dB
Maybe some people say that's obvious, but I have seen it more than once, even with professional products.
Also important to add, is that dips are far less noticeable than peaks (sometimes even close to impossible to hear)
Last edited:
I'm rather surprised that someone can hear 0.3dB difference in your very noisy test room.
He must have "GOLDEN EAR".
He must have "GOLDEN EAR".
Last edited:
What an amazing waste of time!? Why does anyone even post a single thing here? The OP hasn't conducted anything like a real double blind test, he can't give any genuine details, & his "results" and objectives are so transparently biased. He simply loves watching you waste your time -- like I did scanning through this thread. Best thing you can do for DIYaudio is to ignore this whole thread. Unsubscribe like I am doing. Amazing the mods don't shut him down.
There are billions of posts with similar content?
Can't say I agree with everything here, but why would you shut someone down?
That's called freedom of speech. If you don't like it, just don't read it.
1. Goal? Where did you read that anything was oriented? 😕
Quite the opposite, the very start of that project was to conduct an appreciative blind test, not an identification one. There is a thread with something like 100,000 views about it... Sorry but you are wrong on that one.
2. Room. The room was not noisy. And the RT60 was very acceptable. More importantly, I wanted to have a test, therefore a room, that reflects real-life conditions. And that was the case, except it was a large room... But used in nearfield. I see no problem with that room whatsoever, quite the opposite.
The very concept of FLAT is open for discussion, anyway.
Equal-loudness contour - Wikipedia
...it's an average. The best ''flat'' FR would be one that is tailored for each of us. And even for our left and right ears (!)
Depends on the variations. Tones & sweeps are known to be of limited (though not devoid of) practical value in assessing such things. Human hearing being primarily amplitude based, we 'key off the peaks', so narrow-band nulls tend not to be especially audible in ordinary listening. And a very high Q peak may not be a major issue providing (significant caveat) it occurs at a point less likely to be excited with most programme material, and the driver's distortion performance is sufficiently accomplished that any HD excitation isn't an issue. You can certainly hear a peak if you hit it with the right frequency though.
Well, seriously, you should do the test in quiet environment, not in 52dB noise floor room as I explained already.
Why do you think those HeadFi heads love DAC comparison? It's not because the headphones sound better than speakers, but headphones offer lower noise floor.
Why do you think those HeadFi heads love DAC comparison? It's not because the headphones sound better than speakers, but headphones offer lower noise floor.
Answered the user silent_mike in PM.
I don't think he deserves any public attention here...
It's possible that my AC RTA is not the best tool to measure noise floor. Will test later my living room to see what is the noisefloor and we'll see how it compares.
But for sure, that audio lab room was not noisy at all. Quad 5/8'' drywall, rockwool, soundproof door, A/C off.
My friend has a private music studio in a building in Brooklyn. I told him his room is somewhat noisy, but he denied it. Once you accustomed to noise floor, you can't hear it.
A desktop computer can add +20dB noise floor. Extreme care is needed to obtain reference 20-30dB noise floor.
It's pretty obvious and completely logical.
If you don't conduct an identical 'before' test you can't be sure what factor is responsible for the results obtained from the listening subjects. You assume it's the bandpass filter implemented in the miniDSP. But you don't know that for sure
Maybe it's cabinets the drivers are mounted in, or the cheap D-class amp with poor harmonic distortion specs that you are using, or lousy MP3 source files, or something else. The point is that you don't know, because you haven't eliminated all of the other possible causes. The only way to do that is to run an identical 'before' test, which you apparently you did not do.
You also should have run and provided for us full FR graphs with a microphone located at the listeners head position. Those also should have been done both before and after the filtering was added with miniDSP.
You apparently did not do any of that, or if you did, you haven't reported it.
Without that data and the 'before' ABX test results your conclusions lack any merit and remain useless.
- Status
- Not open for further replies.