Here is HD measurement for B80 at equivalent to 90dB at 1m drive (96dB at 0.5m). I get the same bump at 1.2kHz, otherwise seems cleaner. Sometimes the mic adds noise and HD. This UMIK-1 is cleaner than the UMM-6 when it comes to HD data.
Attachments
So X is going to kill me as it is off topic for him, buy why not the 5" from PRV Audio for more spl you need ?
Byrtt, X, why not includ this high efficienty wide band ??? instead let him alone eslwhere ?? Ok the test 5 is full and with also littlier cones...but...
list is long, why not add a driver and divde in two tests : one for paper cones and the other for the aluminium and plastic cones ?
Last edited:
Eldam,
sound clips at round one include PRV 5MR450NDY driver and it got relative prefered by many (link: http://www.diyaudio.com/forums/full...n-3in-5in-full-range-drivers.html#post4242856).
sound clips at round one include PRV 5MR450NDY driver and it got relative prefered by many (link: http://www.diyaudio.com/forums/full...n-3in-5in-full-range-drivers.html#post4242856).
X can you do the harmonic distortion for the 10f and B80 from 40-20k hz?
If it is not too much work could you do a 40-20k with both the RS225 included and then just the driver with its 350hz crossover curve applied (mute the RS225)?
Maybe at the louder level (96dB at 0.5m)
I am curious why myself and others heard an increased "fat" bass in the B80 compared to the other drivers.
thank you
If it is not too much work could you do a 40-20k with both the RS225 included and then just the driver with its 350hz crossover curve applied (mute the RS225)?
Maybe at the louder level (96dB at 0.5m)
I am curious why myself and others heard an increased "fat" bass in the B80 compared to the other drivers.
thank you
Last edited:
It would be interesting if it's possible to measure distortion at 85db and 95db and record at the same time. Then calibrate them to same loudness and make a blind test.Here is HD measurement for B80 at equivalent to 90dB at 1m drive (96dB at 0.5m). I get the same bump at 1.2kHz, otherwise seems cleaner. Sometimes the mic adds noise and HD. This UMIK-1 is cleaner than the UMM-6 when it comes to HD data.
Peter
Sent from my D5803 using Tapatalk
3rd harmonic still biger than 2nd..
Other tested drivers have different distribution of harmonics, that may be reason for victory..
Last edited:
Do you mean a test to listen for the distortion? Very few will be able to "hear" it. And to proof that it is really audible instead of just imagination, it must be done for example using ABX. But then the test must be controlled [similar starting time, similar sound level, and so on and so on] so that the difference heard is not because of something other than the distortion alone.
Then the electronics [of the recording chain] should have very low noise so that the distortion difference is not buried under the noise.
Then the speaker [and electronics] at home should be low distortion also
I think most people [experts especially] don't believe that small non linear distortion difference of drivers is audible
Last edited:
I missed that, thanks a lot Byrtt for the link
The setup has changed a lot since then and I may put a 5MR450NDY into the same setup here for comparison. It will be a Nautaloss though and not a Dagger, as it is too big to fit Dagger. This way you will get a more current comparison vs the top drivers.
Last edited:
I think a proper objective controlled test is almost impossible to setup. But I believe in subjectively low threshold of audible non linear distortion. It is unproven of course but I believe in it.
Imagine you were unable to hear differences between these drivers in this round test. Wouldn't it better if you just purchase the cheapest driver, instead of looking for objectively good measurement results? Don't you find it ironic? Objectively good measurement results should only be more relevant for those with good ears.
Earl Geddes has interesting things to say about that here:
Perception
and of course here:
http://www.diyaudio.com/forums/multi-way/121253-geddes-distortion-perception.html
and also here:
http://www.dagogo.com/an-interview-with-dr-earl-geddes-of-gedlee-llc
Yes, and I disagree with many of the points.
Saying that non linear distortion doesn't matter is wrong. You have to specify the threshold, and you have to mention/specify the listener. Logical reason is: if the listener is deaf of course it doesn't matter.
And I'm more interested with my own ears than how many hundred ears any expert have used to test his hypothesis.
And if you want to say that it is not critical for a good speaker design, of course it is only correct as long as the bottleneck or weakest link is NOT on the non linear distortion itself.
Last edited:
X can you do the harmonic distortion for the 10f and B80 from 40-20k hz?
If it is not too much work could you do a 40-20k with both the RS225 included and then just the driver with its 350hz crossover curve applied (mute the RS225)?
Maybe at the louder level (96dB at 0.5m)
I am curious why myself and others heard an increased "fat" bass in the B80 compared to the other drivers.
thank you
I will see what I can do, time permitting.
Geddes doesn't say that, he says that the importance of nonlinear distortion is overestimated, and that there is no clear correlation between THD/IMD and perceived sound quality up to a point where the nonlinearity simply becomes too much to be ignored. He actually agrees with your assessment that you have to specify a threshold.
Other qualities of the loudspeaker (most importantly, directivity) which do not appear in the "standard" list of specs and graphs have a greater impact on the reported subjective sound quality. In that sense, Floyd Toole gives exactly the same message. When I read Geddes' articles and watched Toole's presentation on Youtube, I had to think of the room in a way acting like an integrating sphere, albeit with low frequency modes and some damping added. After a certain reverberant time, the spectrum of the reflections becomes a weighted average of what the loudspeaker has spewed out in all directions. If this diffuse field has an obnoxious tonal balance, your speaker will subjectively suck, regardless of a ruler flat on-axis anechoic response and vanishingly low distortion.
I think I can agree with that position. It is a bit of a shift in design priorities, unfortunately towards things that are harder to get right. Instead of aiming at a flat on axis response and low distortion, the speaker can only be "good" when it has a smooth integrated polar power response under the conditions that the on-axis response is flat AND it has sufficiently low distortion at its intended output level. If you reverse this coin, it means that tweaking the on-axis response to flat within 0.5 dB and aiming for ultra-low distortion is fruitless as long as the power response is sub-optimal.
Even that statement is not precise. Who has overestimated the importance of non linear distortion? The majority of ? May be he is the one who had over estimated it's importance, using his own assumptions, then he used other people ears to "prove" that he was wrong.
This is what he said: I am putting this topic out so that people can become aware of what is being recognized as the truth about distortion. Its not what we thought it was. Pursuing a loudspeaker design to lower the distortion is a waste of time if its nonlinear distortion that you are trying to lower. It simply doesn't matter.
We have to learn from this driver test. People prefer 10F, B80 and even that bloody Dayton. Do you think those ears do not matter? Experts should learn and understand what people prefer, not to force his own preference. If you know precisely that directivity is the key to great speaker, then you know precisely that you will get rich because everyone will purchase your speaker.
Now how many years do you think does Toole and Harman International need to put M1 in the group of most wanted speakers in history?
Doubtful at the low listening levels that the test was done at all drivers exhibited very low distortion. It's only under duress that the cracks start to appear.