-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Research Project: Infinity IL10 Speaker Review & Measurements
- Thread starter amirm
- Start date
I think what this comes down to is the need to implement the Gedlee metric in speaker ratings. Speaker design people can have productive conversations about distortion since it is their job to sniff out nonlinearities and optimize them, but the task of ranking speakers in terms of sound quality is another matter entirely. We see that there is very little consensus on this issue, largely due to the statements from Toole about distortion 'not being a factor'. Well, something besides DI and FR is at play here. As someone with a personal interest in designing good speakers, I am not HD plots show what we are hearing (I'm not a trained listener; I don't even like Dire Straits.) The Gedlee metric, on the other hand, seems to have greater explanatory power, but would be difficult to implement. Difficult for me at least.
I would encourage everyone to read @hardisj 's new review of the Jamo speaker, which has some interesting measurements showing power compression vs. frequency. I think this is a really useful way of showing how a speaker shows strain at high amplitude.
At times it seems Amir is overstating the value of the professional listener (he sometimes seems to almost suggest they are immune to bias), but then he restates his argument, and he clearly is not suggesting they are infallible. Blumlein captures Amir's claim quite well, the pro is not infallible, but she or he is more likely to identify an issue and in a shorter amount of time than the layperson. Just like one cannot say with absolute certainty that a professional's sighted evaluation is free of bias, one cannot conclude that it is of no value.
QMuse, it is ironic that you are the one actually committing the "sin" you attribute to Amir...you are the one suggesting an universal claim; i.e. all sighted tests are tainted by bias, thus of no value. In contrast Amir's argument has not been that his evaluation is without question because he has been professionally trained, but rather his evaluation has value because of his training (please note the statement of "value" as opposed to "truth") and is worthy of examination, thus his suggestion that others should listen to the speaker and see if they share his perception. Now, you may argue that others will be biased by his findings and be affected by said bias when they listen, and you may be right, but they nonetheless would provide another data point. Your repeated references to bias in sighted tests suggests that there is absolutely no value in Amir's evaluation, the total dismissal of that data point is questionable stance at best, and a corruption of scientific inquiry at worse.
PS Amir and others have already stated this, but it seems to have been forgotten. Amir did not compare this speaker to the other speakers tested by Harman, and there is nothing to suggest from his testing that he would not come to the same conclusions they did...it was the best speaker from the pack.
OP
- Thread Starter
- #224
For near field testing, I have tracks converted to mono. For far field testing, I just play the same channel, same music and same location for all speakers under test (give or take an inch or two). I am not trying to enjoy the tracks so conversion to mono is not necessary. It is just a test signal in this regard.
OP
- Thread Starter
- #225
Adding on, in near field testing I switch left and right speakers on and off and hence the reason the conversion to mono is necessary (I am comparing speakers there).
Vuki
Senior Member
At least here in Europe doctors have legal obligation to report side effects noted by them or patient to the drug agency. Collected data is then used to control and correct data in producer's drug documentation. Based on that data, every few years drug documentation is revised. So the story of the doctor that builds his own database...I would ask for second opinion
martijn86
Active Member
Uh oh. Someone forgot to plugin a noise harvester!
Vuki
Senior Member
I don't get this thread or review though, there seems to be so much energy put into irrelevancies, I don't understand the point of this thread, particularly if it's main purpose is to find out if Amir's Klippel rig is measuring accurately....there's simpler ways of doing it, I really don't understand all the detailed discussions about this speaker and how it relates. Like it was said in another thread, just test the same 'controversial' speaker that produces dubious results on Amir's Klippel....just test that speaker on Amir's rig & then send it to Klippel for independant testing to see if it marries. All the calorie burning in this thread about god knows what and the review (if that is the aim), just seems to be a lot of people spinning their wheels and burning calories for no reason. There's gotta be better ways of concluding if Amir's Klippel rig is accurate....and to be honest I can hardly believe that is the purpose of this thread (I still doubt it)...Amir didn't really say what the purpose of this thread was beyond comparing Klippel with anechoic on the same speaker model....but that was 20 yrs ago and the speaker is a different unit and the speaker has aged and anechoic does not equal Klippel....too many factors & uncertainties if the main reason for this review & thread is to establish Amir Klippel accuracy....there's gotta be a simpler way of doing it....I really don't get this thread, feel like I'm in a parallel universe! I still don't know the intended purpose of this thread and can't believe it's purpose is to establish Amir Klippel Accuracy like some have said....hopefully Amir can clarify the overall purpose of this review & thread.
Last edited:
Afaik, Amir is already working with Neumann on verifying some earlier results with that speaker.
I know Amir said he wanted to replicate the anechoic measurements, but I think the reason this speaker is interesting is because it is the only speaker that has been tested here that we know was used in one of Olive's preference studies.
Moreover, it was the highest bookshelf speaker in Olive's preference study part 1. So one goal was likely to see how amir's listening impressions correlated to that measured data.
The reasons there is so much discussion in this thread is because for the first time we have a speaker we know to be a blind test champion (back when the study was conducted), which has (at first glance) a good spin, which has extensive written listening impressions in the original study, which Amir's results correlate perfectly with Harman's, and yet which Amir could not like.
For all of the above I'd actually say this is the most interesting speaker Amir's measured yet; and perhaps it will end up being one of the most enlightening in terms of how well we can correlate measured data to listening tests.
GelbeMusik
Senior Member
- Joined
- Apr 19, 2020
- Messages
- 445
- Likes
- 290
Please don't do this. First, this is prescribed by Dr. Geddes. There are side effects to be expected. However, I like to see other papers than Harman's here.
Question: Is the model correct? Nope. It's incomplete. It's not specifically frequency dependent by amount and phase. It's not level dependent.
How much does a test person cost? Or ten for five days? I think a little less than a single REVEL flagship. You could always take students for the sake of science. And yet you can't design a study to get real, complete, conclusive results? My friend, that's not science, that's writing papers.
And Dr Geddes hasn't made much progress over the years, has he?
As I already explained above, completely ignored of course, THD cannot be a criterion, because every instrument, even electronic ones, needs harmonics. In general, their level is hardly determined to fractions of a percent. That's a matter of hundreds of percent (great translator, by the way--this is on you, you little AI mouse).
Well, that's not in the Toole, in THE book. But you should know that, right? I am definitely not into human sciences. Rather the opposite, and logically quite critical of their methods. But to ignore their experience with human decision making simply ruins the whole thing. It turns into a pretty funny quack.
After using the new translator, I think that the lack of understanding has little to do with the language barrier. Neither the reference to human sciences is of interest, nor technical facts that are not in THE book. Continue!
Admittedly I have found it much easier to (subjectively) assess performance with a single speaker playing -- rather than stereo!
The point of linking a Geddes paper was to show that we don't need to re-invent hot water when considering distortion. Geddes showed that not only THD as a single number is not indicative but he also showed that you cannot simply analyze each harmonic component in absolute term, like you did with 3rd harmonic. Instead, he developed a formula which introduces weighting to each harmonic component and calculates a new index for which he showed it correlates perfectly with listening impressions.
In the context of Geddes paper Amir's comparison of HD3 with sighted listening impression, in which he saw distortion measurement before listening, is of course irrelevant.
If it is to be done the right way I suggest a distortion index is calculated according to Geddes formula and than compared to listening impression obtained without seeing the distortion measurement before, as everything else is a joke.
Last edited:
tuga
Major Contributor
This is probably the biggest challenge of critical listening: trying to track down probable cause(s) for the issues that you have identified in your listening assessment.
I would add that this assessment should be performed over a long period (at least a week or two) with your system in your room with known and fit-for-specific-purpose material order to achieve the highest effectiveness possible.
Last edited:
tuga
Major Contributor
I thing we should start with a more descriptive characterisation of what "grunginess" means to @amirm , which frequencies affected, how does this translate to sound in the way that instruments and vocals are affected.
And there's still a chance that the unit might not be operating to spec...
restorer-john
Grand Contributor
One thing is for certain, the price of them on the secondary market will go up due to the attention their are receiving.
maty
Major Contributor
We have to use both possibilities.
When I modify some speakers, first I only do one and compare with the other without that modification. Sometimes, even so, the difference is very noticeable, surprising. Then you fix the other one and listen in stereo and you can hallucinate like it happened to me months ago. We hallucinate the two people, the same who have been present in the previous modifications. With the theme I always use to compare: Hotel California (Eagles).
tuga
Major Contributor
In regard to harmonic distortion here's a clip from a Neumann tech note:
restorer-john
Grand Contributor
With loudspeakers, our brains adjust to them over a period of time, make allowances for their deficiencies and essentially "fill in the gaps" in their responses. In a quick A/B comparison, a new speaker, as compared to a "multi year loyal friend" is at a distinct disadvantage, even if it is better in some parameters.
The classic one speaker demo gives no comparisons and given enough time, the speaker will become the new normal (assuming no gross faults or anomalies).
For punters walking into a HiFi store, the old comparator setup enabled an instant, unbiased and effective way of determining the best speakers in very short order.
The classic one speaker demo gives no comparisons and given enough time, the speaker will become the new normal (assuming no gross faults or anomalies).
For punters walking into a HiFi store, the old comparator setup enabled an instant, unbiased and effective way of determining the best speakers in very short order.
tuga
Major Contributor
On the other hand long term listening allows the listener to:
a) go over a more varied programme increasing the chance of exposing a wider range of faults (the BBC mentions in one of their papers that high-Q resonances can be audible if they hit a fundamental note; my experience corroborates this i.e. violins and mid-woofer resonances)
b) move focus from a new (to the listener) tonal balance and room interaction to other aspects of the transduction
The trick is to focus on sound and not the music. Listening for enjoyment and critical assessment are different tasks.
Last edited:
Similar threads
- Poll
