Actually, I think I agree with you and said it is not a black art, but many say it is. I do think that many speakers just are not that great, maybe because some make the wrong compromises? And what makes a compromise good or bad? And maybe costs have little to do with it?
For example, about two months ago I heard a comparison between two speakers and about equal size, both very thoroughly engineered I am sure. The cost difference was $82.000 versus about $5000 (a DIY published design that has been around for some years) and I was one of four persons in a decent room. The two pairs came from two friends of mine.
The only thing that was changed was the speakers and the location in the room was the same, the other speaker not in the room when the in-system speaker was. The difference was not small, all of us were of the same opinion, even the owner of the very expensive speakers (they are now for sale).
Bottom line, the difference was not small. By all accounts the more expensive (highly reviewed) speaker should have been better, and the owner should be insisting his expensive speakers were better, but knew better. His best mate was there and just could stop chuckling. But the question remains, what is really going on?
For example, about two months ago I heard a comparison between two speakers and about equal size, both very thoroughly engineered I am sure. The cost difference was $82.000 versus about $5000 (a DIY published design that has been around for some years) and I was one of four persons in a decent room. The two pairs came from two friends of mine.
The only thing that was changed was the speakers and the location in the room was the same, the other speaker not in the room when the in-system speaker was. The difference was not small, all of us were of the same opinion, even the owner of the very expensive speakers (they are now for sale).
Bottom line, the difference was not small. By all accounts the more expensive (highly reviewed) speaker should have been better, and the owner should be insisting his expensive speakers were better, but knew better. His best mate was there and just could stop chuckling. But the question remains, what is really going on?
Ohh how some wishes there is something magic involved. Because otherwise they could not meet the competition?
//
Black magic is almost a euphemism for insufficient scientific understanding. You’ll see some engineers speak of things like RF as black magic, too. It’s not, of course, but it can require a lot of work to figure out what’s actually going on. Some of the tools available now for simulation have really taken the magic out of it for many applications.
If you’re working on the fringes of physics then maybe you get a pass to use black magic
. Everyone else is just lazy.
If you’re working on the fringes of physics then maybe you get a pass to use black magic
. Everyone else is just lazy.
Because he was honest, and made no extraordinary claims, and made no allusion to science.
That hypothesis could be relatively easily tested, and fixed if found true.
Once again, what's the point of the question? That I am anti-measurement? I think not. I am very much pro-measurement.
I use measurements for development, can't we have a discussion about that and not constantly go down the ABX road? Is that where you are trying to aim this? Not interested, haven't that been beaten to death. Consider me somewhere in the middle, OK?
Speaker development is yet to be finished, maybe some might think that true about amplifiers (have my doubts about that), but it is not true about speakers. You said "compromises" - are you aware of some of them? Not a trick question, just want to know your extent of interest in speakers?
Speaker development is yet to be finished, maybe some might think that true about amplifiers (have my doubts about that), but it is not true about speakers. You said "compromises" - are you aware of some of them? Not a trick question, just want to know your extent of interest in speakers?
Joe, there are so many successful speaker designs, including conventional two way, that had probably millions of happy customers even repeat customers that I don't see the point of your first comment.
You've been asked before about some measurements to substantiate your claims, not hard to make, but none are forthcoming. First that adding conjugate shunt networks to make a speaker have a constant real impedance has anything substantial to do with the actual distortion of its output. Second, that class A/B amplifiers can't drive a complex impedance (this is no longer 1970).
Last edited:
You are not alone Joe,
I made some speaker designers blow a gasket on another forum with some of the things I was doing in my 10” two way build.....I posted results there and have also posted a couple here and was told my tweeter was disconnected! (I assure you it was not). Textbook design (I’m not talking about online calcs) is not going to work unless ones listening room is an anechoic chamber......things must be tuned to the lp with music at preferred volume afaic.
Audibility of low-level noises is dependent on the music being listened to, and with some (especially early) DACs the shaping of the noise floor does not hide lowest-bit granularity which dither should have smoothed. With the best (and likely most) DACs these days the noise floor is smooth even if a recording is listened to at -70 dB. It just gains a high smooth-sounding, analog tape-like noise floor.
I have only heard it when intentionally probing noise floor sonic character, or listening in headphones to low-level orchestral passages or quiet ambient electronica.
Cheers,
Howie
Last edited:
Random modulation creates sidebands too.mmerrill99 said:
How do you propose distinguishing between random noise in your test signal and random noise in the DUT output? You will need some clever autocorrelation stuff to sort out the mess. As you are proposing this test I assume you have some ideas about how you will analyse the output?
To some extent yes it is, Harbeth is an example of a lightweight loudspeaker with high cabinet radiation and has high regard whose trick is the subjective tuning of these cabinet behaviours and this is reinforced by measurements and experience - Factory Visit. So there are cabinet dimensional and ratios rules and driver placement rules in addition to actual drivers and crossover implementation rules and some loudspeakers come together and the result is powerful clean and musical, lesser speakers get things slightly out of tune somewhere and the subjective enjoyment goes down fast.
The electrical load that the loudspeaker presents to the cable connected unknown amplifier is another set of critical dependencies and important part of design considerations and usually essentially overlooked. If combining all of these considerations and achieving a pleasantly transparent and musical systems is a black art, then so be it. Impedence compensation across the loudspeaker input terminals although a costly way to remove amplifier/cable dependency out of the equation and ensure optimal amplifier operation and hence loudspeaker performance is another design decision, and in the case of Joe's speaker design the user feedback is very positive.
Dan.
Last edited:
But it doesn't, a 70's era amplifier had crossover distortion into a perfectly resistive load.
Right & if it is time series recording of a dynamic mutltitone test signal then it is possible to make it pseudo random & possibly see some pattern emerging of noise floor modulation in correlation with a pseudo-random multitone pattern
As I said maybe a pseudo random dynamic muititone test signal could be used?
- Status
- Not open for further replies.