I think if we don't get too personal, this thread is really having a healthy discussion. Regardless of what kind of horn/guide is used. If we review the various data, it seem the way the thoat expands, the overall angle of the device, and the lip of a device all have some impact in the way the horn/guide responds. Not having heard all types of horns/guides, there are things that I like about what I have heard, and things that I don't. But I think it's worth exploring various configurations to see what comes out.jipihorn said:
Looking at the data from my experiments, the latest set seems to come out somewhere between the two types being compared. The biggest problem I see with this discussion is that lots of people are trying to determine who is right or wrong rather than trying to find if anything can be better and understanding the tradeoffs.
Personally I think directivity of long throated horns are not ideal. Whether the OS waveguide of 90deg is ideal or not, I don't know, but I think that some degree of controlled directivity is desireable. Earl thinks constant directivity over a certain angle is ideal, but I don't think anyone has actually described how directivity effects what we hear in comparison with a different directifity pattern. Earl indicated that he had done the work of mathematically exploring how directivity effects image, but it was never put into the book he originally thought he did. So really, none of us have anything to hang out hats on.
HOMs is a really annoying issue, the honkyness is what I personally dislike about horns, but there is currently no data that can clearly show which device has what level of HOMs. I have previously posted impulse data for raw driver and the hornned/guided driver, but have not done this with the second wave guide because there are too many people just doing nothing.
The subject of this thread gives me the impresson that any type of such device is welcome into discussion just kickstarted with the OS wave guide that Earl had developed. If this is not the case, then we need a different title.
gedlee said:
That's an intriguing observation. In my experience I've found that the foam plug makes a dramatic improvement. The graphs help to explain why this is; there's a fair amount of diffraction in the waveguide, and the plug "soaks it up."
The illustrations also prove your argument about directivity, as they clearly demonstrate that the LeCleach horn is beaming at 5khz. (you can see the narrowing get more acute as the frequency rises.)
Very interesting!
I'm sorry, but there was absolutely nothing in his post that could have even remotely constituted an ad hominem attack.Jmmlc said:
Anyway, I am in agreement with soongsc. There is good information in this thread and I'm sure more will be forthcoming. It would be a pity if the exchange of ideas was halted...
Patrick Bateman said:
Lets face it, a straight tube with a radiused mouth would be diffraction free, very smooth, but it won't be CD. It simply makes no sense to compare a waveguide designed specifically to be CD (and works well as such) with one that isn't. When Jean-Michel can show me a CD horn that is better than mine THEN I WILL BE IMPRESSED. But for now all that he is trying to do is confuse the facts with data that is beside the point.
And John, would you comment on the "harshness" or "honkiness" of the waveguides in your speakers? Do they sound like horns or not? Is there any evidence to say that the waveguide is worse than a horn in respect to what Jean-Michel claims?
Jmmlc said:
You have to make some huge progresses about the art of sophism and the fact that you was a moderator is irrelevant.
Is it too difficult to do a few lines copy/paste of your affirmations on [son-qc] where you write where Geddes was wrong about his waveguides ? I can even translate them for you if you don't have the time. I don't obviously speak about the ad-hominem (read the definition) remarks you made about him on other posts, but the one where you make a list of critisms about his theory and results.
I would LOVE to see his answers on this very topic, all in the same place, in the right thread (i.e. this one).
Jipi.
I don't know if anyone else is interested, but I would be interested in seeing an impulse for just your wave guide part. The impulse you had shown claiming to be a system impulse was not really impressive from a technical point of view, and looking at it's shape, even without the woofer, I don't see how it can be improved.gedlee said:
I would also like to see the frequency family curves without any smoothing.
soongsc said:
Soongsc
The problem with your requests - almost all of them - is that they have already been done, but you either haven't read them of forgot. And your English is contradictory when you state "The impulse you had shown ... was not really impressive ... I don't see how it can be improved". It's "not impressive" but "can't be improved"? That makes no sense.
I smooth the data because I use noise and some smoothing is always required for noise because its not a deterministic signal. I can show 1/6 octave data, but below that the curves tend to be too noisey from the input signal noise.
Jêrôme,
Your hate of me is known here in France since long time, more than 15 years I guess. All along that time you refuted You refuted what I brought in Hifi about crossovers, quasi 2nd order bass-reflex, amplifiers and horns...
I don't know why your are so obsessed about me, may be a way to compensate your own lack of creativity.
Eevryone will be interested in the obsessional way you treated me as "an ennemy of science" on your blog:
"http://jipihorn.wordpress.com/2008/12/14/les-ennemis-de-la-science/"
Jean-Michel Le Cléac'h.
Your hate of me is known here in France since long time, more than 15 years I guess. All along that time you refuted You refuted what I brought in Hifi about crossovers, quasi 2nd order bass-reflex, amplifiers and horns...
I don't know why your are so obsessed about me, may be a way to compensate your own lack of creativity.
Eevryone will be interested in the obsessional way you treated me as "an ennemy of science" on your blog:
"http://jipihorn.wordpress.com/2008/12/14/les-ennemis-de-la-science/"
Jean-Michel Le Cléac'h.
jipihorn said:
gedlee said:
Okay, so since you always smooth data, which I would assume is also done with impulse data. I have tried to find impulse data mentioned in this thread and try to get them to about the same time scale as follows:
An externally hosted image should be here but it was not working when we last tested it.
From this, if we smoothed all the impulses in our minds, we can see the differences between these. I hope we can here what people think about what pros and cons can be seen in each.
Jean-Michel
I have heard of your attacks on my work from more than one source and have always wondered why you will not present these arguments to me directly. This is by far the lowest form of criticism, one for which every court of law in the world will not abide. If you are so convinced of your beliefs then present them here, now. Stop hiding behind the fact that you speak two languages and I speak only the one and hence cannot respond to criticism in French.
I have heard of your attacks on my work from more than one source and have always wondered why you will not present these arguments to me directly. This is by far the lowest form of criticism, one for which every court of law in the world will not abide. If you are so convinced of your beliefs then present them here, now. Stop hiding behind the fact that you speak two languages and I speak only the one and hence cannot respond to criticism in French.
Without getting into general claims as to which is good or bad, and is just hand waiving statement.soongsc said:
First we notice that 3 curves show that display of data is probably limited by the display resolution due to screen capture, and you can see exactly not so mooth transition due to pixels. 1 curve does not show this, I assume some sort of soomthing occured somewhere that causes this.
In the time scale, we can see that the width of the first impulse are about the same in all impulses, so we can assume that the high frequency probably will be not too far apart, maybe 16K~20K range or so.
The first two impulses have a very large inverse impulse. In my own experience, impulses that show this characteristic seem to have softer sounding cymbals. The latter two impulses do not have this, and generally cymbal impact will have a crisp impact sound.
Notice the first, third, and last impulses have some high frequency ringing. This kind of ringing will let cymbal trailing rings sound a bit messed up, the latter two being worse, and can cause a bit harsh sound in general. The first one might give a bit impression of glare. What the second impulse will sound like is unpredictable because it just seems to have been smoothed out either due to original generation or in the graphics format conversion process.
Notice the third impulse has a secondary peak that is quite high. There is something very close to the source AC that causes this.
Notice in the second impulse, after the second peak, the response does not go negative, but rather seem to be effected by somthing a bit further away from the AC source. If one calculates the timing, it is possible to determine what causes this. But since this seems to be a system response impulse, we do not know what the real wave guide impulse looks like.
That's it for now.
Soongsc
None of that is correct. I've discussed many time why the impulse response MUST BE symetrical about zero, since there can be no DC. If it is not then its an impossible impulse response. Only the one that I did meets this requirement and so all of the others are not valid impulse responses. Three and four are giberish.
None of that is correct. I've discussed many time why the impulse response MUST BE symetrical about zero, since there can be no DC. If it is not then its an impossible impulse response. Only the one that I did meets this requirement and so all of the others are not valid impulse responses. Three and four are giberish.
Hello Earl,
Your work is not in question and most often the purpose of my interventions about your work or opinions, on french language forums wanted to be informative and eventually a source of discussion with their members, most of them don't read technical English.
So eventually we could discuss about :
- the inaudibility of compression drivers distortion (what pertinent
measurement should we perform to explain the differences we hear);
- the inaudibility of phase distortion (we did experiment with members of different forums on that...),
- the inaudibility of harmonic distortion at even large level (we did a group experiment on that on forum Audax...),
- the inaudibility of differences of sound between amplifiers (we did also many experiments at Mélaudia association),
- the useless of the loading of the loudspeaker by the horn...
- about loading largely not dependant on the horn profile...
- an irregular soundfield inside the horn doesnt mean bad sound at the listening position...
- directivity as the only important characteristics of a horn...
That, from memory, those subjects - that I am not alone to consider as very controversial - we used to develop the discussion of which starting eventually from a quote from you.
But attack of your work, "hidden in a foreign forum" they are not it is a lie to say that .
Even, often I translate in French some of your ideas about which I am OK with you, by example "multiple subs", "a new metric"...
I think that if you are somewhat known in France, a large part of this is probably due to the translations I did of large excerpts of your messages in English. You should thank me. ;-)
Best regards from Paris, France
Jean-Michel Le Cléac'h
Your work is not in question and most often the purpose of my interventions about your work or opinions, on french language forums wanted to be informative and eventually a source of discussion with their members, most of them don't read technical English.
So eventually we could discuss about :
- the inaudibility of compression drivers distortion (what pertinent
measurement should we perform to explain the differences we hear);
- the inaudibility of phase distortion (we did experiment with members of different forums on that...),
- the inaudibility of harmonic distortion at even large level (we did a group experiment on that on forum Audax...),
- the inaudibility of differences of sound between amplifiers (we did also many experiments at Mélaudia association),
- the useless of the loading of the loudspeaker by the horn...
- about loading largely not dependant on the horn profile...
- an irregular soundfield inside the horn doesnt mean bad sound at the listening position...
- directivity as the only important characteristics of a horn...
That, from memory, those subjects - that I am not alone to consider as very controversial - we used to develop the discussion of which starting eventually from a quote from you.
But attack of your work, "hidden in a foreign forum" they are not it is a lie to say that .
Even, often I translate in French some of your ideas about which I am OK with you, by example "multiple subs", "a new metric"...
I think that if you are somewhat known in France, a large part of this is probably due to the translations I did of large excerpts of your messages in English. You should thank me. ;-)
Best regards from Paris, France
Jean-Michel Le Cléac'h
gedlee said:
soongsc said:
You are missing the point - you said
This is completely wrong because there HAS TO BE a large negative impulse when there is a positive one because the average must be zero. And since the impulse response goes to zero very fast this average must be zero over the short term. And you say yourself that "The latter two impulses do not have this" and they must have!! I'm not going to argue this point with you its too fundamental.