Pan said:
Show me what you have that can tell me you're equally or more experienced than Zaph. I am truly sorry that I didn't know you or remember you, if you're a person to be respected.
Why crude? You will see. I wouldn't reply again till I get back with data.
I didn't believe that mic distortion doesn't exist. The problem is how that relates to driver distortion and results in driver distortion plots we see. Again, let's see.
-jAy
simon5 said:
Exactly. Which is why I asked for evidence that has been published in a peer-reviewed journal. Stereophile is not a peer-reviewed journal. It is a magazine. I hope you understand the difference.
Then, another example. Zaph tested the mighty Thiel and Partners, Accuton D20-6 Pure Diamond dome tweeter at 2600$ per tweeter, or 5200$ per pair. The owner gave it to him for testing only if he had an embargo on the data produced. He never lifted the embargo. Strange, no ?
You can draw zero conclusions from this information. The terms of the tweeter exchange are between Zaph and the owner. If the owner does not wish the results published, that's his prerogative. In fact, I would not be surprised if the $2600 tweeters sounded absolutely amazing but had poor measurements in a couple of places. Zaph may be concerned that perhaps his measurements don't correlate with sound quality and choose not to publish. The manufacturer may be concerned that the measurements that Zaph does don't reflect the driver's true sound quality.
I'm glad you bring this up. Double-blind studies that show that differences in speaker cables and interconnects are generally not audible are found all over a) scientific websites that publish their methodology and analysis and b) reputable journals. On the other hand, it's on INTERNET FORUMS that people talk about how exotic cables improve imaging/detail/whatever and they dismiss anything published in a reputable journal, etc. Does this sound familiar? When I asked for support in the form of ANY published reputable information, I was poo-poo'd by you and jAy. I was instead directed to the pervasive opinion of non-professionals on internet forums. Ask the average person on an internet forum, and he'll tell you how much better his stereo sounds with Monster Cable.
You're asking why scientists and engineers write and publish papers??? Are you really that unfamiliar with...science?
And this is based on.....?
SG
Jay_WJ said:
I really don't get this. Is it a contest now??
We have a disussion and in my book arguments and science can stand on their own feets without a "name".
A little advice, try to show all people respect and stop the day they don't deserve it.
Fine, the data you pointed me too so far suported what I say and goes against the thing you first said.. which I find a little strange.
/Peter
Jay_WJ said:
Exactly. This is why I will no longer reply to his post. I don't understand what he's looking for in this thread. It seems that he wants subjective opinions on tweeters. And he keeps asking to show scientific basis on a related issue at the same time.
No. I believe that differences in tweeter sound quality can only partially be explained by measurements. I also believe that after listening to two tweeters extensively, you can tell which one sounds better -- and this makes measurements irrelevant. This is why I asked (repeatedly) for people who, specifically were familiar with the 7000 and another tweeter, to offer a comparison.
Instead, my thread was hijacked by you, who insisted that tweeter sound quality is solely determined by a SINGLE measurements - the HD products of the 1-2.5kHz freq range specifically. This type of BOLD scientific claim requires support. You offered zero evidence, other than what amounted to your interpretation of what non-professional semi-anonymous people posted on another internet forum. You also insisted that all tweeters crossed over >4kHz have the exact same sound quality. I find that absolutely preposterous and completely without merit.
The big irony is that you claim that you don't understand what I'm looking for in this thread - which is probably why 100% of your responses so far have been completely unrelated to my question in the original post (you did read it, right?) as well as my 3-4 repetitions of the original question. Here it is again:
Original Post:
Hi, just wondering - does anyone know of a tweeter (of any type) that sounds better than the SS R2904/7000? I'm curious if anyone has any first-hand experience with something better.
SG
smellygas said:
You think a peer-reviewed journal is different/better ? Everybody take a side in that too. You also can easily publish crap that will fly below the radar, if you peers are powerful enough.
Sorry but Zaph really wanted to release the results, but he also really wanted to test that tweeter. He said he will never accept that condition again ever, because he hated the experience. I don't know why you bash him, he's far from perfect, but he's reliable and honest.
Unfamiliar with science ? Sorry, I'm recently graduated in computer engineering. I'm familiar with science, thank you. I just tend to keep both feet on the ground and be objective. I must tell you that you asked for subjective measurements, where's science in this ? Then you said measurements are irrelevant... you aren't familiar with science maybe ? That's a personal attack, I didn't attack you. Well you seem to have no argument left, since your on personal attacks mode now. I guess I won this argument...
This is based on : It is super easy to do a tweeter that's good over 4 kHz. Basic physics to me. Look so evident, sorry...
To your original post, I will still answer the same thing to the first part of your question. Seas 27TBFC/G. For the second part, I'll let someone else answer.
simon5 said:
You think a peer-reviewed journal is different/better ? Everybody take a side in that too. You also can easily publish crap that will fly below the radar, if you peers are powerful enough.
Some peer-reviewed journals are better than others. That's why you also read the methodology and evaluate the paper yourself. Or perhaps you think that the opinions expressed on a semi-anonymous internet message board are somehow more authoritative?
For the second time, you cannot draw any conclusions about the sound quality of a tweeter based on the owner's reluctance to have measurements published. And honestly, if someone were using a cheapo ECM8000 mic + SoundEasy, which is a non-professional software package sold by Parts Express, I wouldn't want that stuff published either...but I digressed.
Then you shouldn't have to ask why scientists publish papers.
I've already addressed this exact issue in my response to Jay. There is no contradiction.
you aren't familiar with science maybe ? That's a personal attack, I didn't attack you. Well you seem to have no argument left, since your on personal attacks mode now. I guess I won this argument...
Regardless of your degree status, if you honestly do not know why scientists publish papers in peer-reviewed journals, and you do now know that Stereophile is a magazine and not a peer-reviewed journal, then you clearly are not familiar with science. There's no attack. It's a pretty obvious observation.
So your argument is that it's "super easy" to do a tweeter that's good over 4kHz, and this is basic physics to you -- therefore, there are a lot of tweeters that are better than the R2904/7000!? Wow. That is just incredibly convincing.
Oh, so you actually read my original post!? Great! So you are familiar and have experience using the R2904/7000 AND the Seas27TBFC/G, and you preferred the sound of the Seas. Can you tell us more about your experience with the drivers?
SG
smellygas said:
Why in the world someone would be reluctant to publish good results ? That is beyond me, sorry. To me, it's clear, no results means he wasn't satisfied and the tweeter is hyped.
Now you bash SoundEasy and say it's an unprofessionnal software package, and you said also because it's sold at PartsExpress... You are pushing your luck a bit far, me thinks. PartsExpress sell most of the big brands, it's not only for cheap things. Why SoundEasy is unprofessional ? Because it's not sold for 10 000$ ? You better have something to back up your statement.
Behringer mic is probably not the best mic, but you can easily work with it anyway and publish useful results. Just like old car mechanics can still fix a car with a wrench, even if it's not the best tool.
There is contradiction, you argue about me about science, then you say well just listen to tweeters and then you can ditch all objective measurements, since they are useless. That's what the cable crowd says about cables.
Where did I ask why scientists publish papers ? I just said scientists may avoid to publish contradictory papers if it's not worth it. Stop reading between the lines, there's nothing to be read.
As if I don't know that Stereophile is not a magazine, I don't know where you are going with that. Pointless tangent. I gave only an example, where did I state it was a peer reviewed journal ? Also it's usually a well regarded publication.
Now, I said it's easy to do a tweeter that's good beyond 4 kHz but I said nowhere that's there's a lot of tweeters better than the 7000. You again read between the lines and deform what I say.
I said I'll let someone else answer the second part of the question, your question was in two parts. You didn't read what I said did you ?
Now to be devil's advocate... let find us a good peer reviewed journal articles about the advantages of NOT crossing a tweeter low and using an overpriced tweeter to cover the last two octaves ?
Jay_WJ said:Exactly. This is why I will no longer reply to his post. I don't understand what he's looking for in this thread. It seems that he wants subjective opinions on tweeters. And he keeps asking to show scientific basis on a related issue at the same time.
So true ... well said !
He need subjective opinions, yet when he disagrees with our subjective opinions, he ask for objective data to back it up... let ask him the same in return.
Hey smellygas, can you find a paper that shows the superiority of the 7000 tweeter over all other tweeters made to date ?
Let make assumptions like he always does with what I say.
He already got the 7000, paid alot of it, and want us to congratulate him on what an awesome purchase he just made, the best tweeter ever.
simon5 said:
No results means he kept his promise not to publish them. Any conclusion you draw other than that is 100% speculation.
Sound Easy is not a mainstream professional software package, just like Windows Paint is not a professional graphics software like Adobe Photoshop. Now you're just arguing for the sake of disagreeing - if you actually were familiar with the software, you would know where Sound Easy fits with the packages out there. It may also interest you to know that Parts Express has a very limited selection of higher end drivers. Compare what Madisound offers and you'll see.
For the third time now, if you do careful controlled listening tests of two tweeters, you should be able to determine which one sounds better. Measurements, on the other hand, are SURROGATES for listening tests. While data may predict which one would sound better, ultimately it is with a listening test that you would verify. This is why I am asking for people who have done such listening comparisons - because regardless of the measurements, if one sounds better it is irrelevant that one has 2dB higher F2 HD, for example. Please read each of the 3 explanations again if you still have trouble.
You brought up how Stereophile was paid off by Bose not to publish negative articles about their products, and you used this example to argue that this is why publications cannot be trusted. This is why I pointed out that I had specifically requested peer-reviewed journals, not magazines like Stereophile. ...and now you call Stereophile a "well-regarded publications" when just one post again, you were "bashing" it because they were paid off by Bose. That is a textbook contradiction.
Now, I said it's easy to do a tweeter that's good beyond 4 kHz but I said nowhere that's there's a lot of tweeters better than the 7000. You again read between the lines and deform what I say.
You wrote "you can probably easily find one that will beat the 7000 over 4 kHz." If you can "easily find" such a tweeter, then there must be a lot of them out there. Or are you suggesting that it is very easy to find better tweeters, but they're very rare?
Now to be devil's advocate... let find us a good peer reviewed journal articles about the advantages of NOT crossing a tweeter low and using an overpriced tweeter to cover the last two octaves ?
simon5 said:
Yeah, I'm having a lot of trouble following your grammar and syntax...but I asked for people with experience with the 7000 and another tweeter to share their observations. There's nothing wrong with that. However, what I did NOT ask for was for people to make a scientific statement that a specific distortion measurement is solely responsible for sound quality and offer absolutely no support. That is NOT ok. This is probably the 4th time I've explained this. Instead of just typing a response, please try to understand it.
Hey smellygas, can you find a paper that shows the superiority of the 7000 tweeter over all other tweeters made to date ?
Once again, I never made that claim. In fact, I was hoping that people could tell me what tweeter they have HEARD and COMPARED that were BETTER than the 7000. Instead, I received responses from you, which failed to address my original question.
Let make assumptions like he always does with what I say. He already got the 7000, paid alot of it, and want us to congratulate him on what an awesome purchase he just made, the best tweeter ever.
I think part of the problem is that your English isn't very proficient, and you don't quite get across what you want to say in your writing. That's why you think people are making assumptions. On that note, I sense a lot of hostility here from you and Jay and in Zaph's writing towards higher priced drivers. It is as though this group seems to look for any justification (other than actual listening) in order to conclude that a less expensive driver is better. However, when pressed, you and your colleagues offer zero support for your bold statements. For instance, you tell me that the it is easy to find a tweeter better than the 7000 for less than $50. And when asked why, you say it's because it's easy to design a tweeter that sounds better at >4kHz. And your next statement is "basic physics." I mean maybe that's sufficient to convince you of something, but it just doesn't cut it for me.
SG
smellygas said:
Again, you need to resort to personal attacks
My english is far from perfect I agree, I'm a native french from Canada and english is my second language. On the other hand, I manage to get my point across fairly good. I think you can understand me well enough. I don't know what this have to do with this argumentation. I'm sure you've read worse english in your life.Now you say it's my fault that you received scientific statements. Let see how it started. It started with some people giving you some good advice and some good comments, then it was an healthy discussion. After a few posts you said :
After that, it went downhill from there and with that kind of post, you were kinda asking for it.
At the beginning of the thread, when I saw this :
I was like, well a new member that needs help and no one want to answer his thread. Maybe because it's a subject that has been beaten to death here, another best tweeter thread. He's getting slightly pissed because no one answers and using sarcasm to get attention.
Then, I decided, let answer something, I will suggest him a good alternative that might surprise him ! Then we might have a nice discussion and argumentation. I guess I was wrong, and as I said earlier, you were converted already to the mighty 7000 god.
Now, you didn't make the claim that specific distortion measurement is solely responsible for sound quality but after that claim was made, you claimed otherwise. You are no better than the one who made that claim since you didn't bring proof either.
I will quote you even :
Anyway, I may be a bit hostile against high priced drivers. On the other hand, as you would say, don't make assumptions for Jay or Zaph.
Jay doesn't have anything against high priced drivers as far as I know with what I read here. He has something against high priced drivers crossed at 4 kHz, if I read what he said correctly. If I understand him correctly, when he purchases a high priced tweeter, he prefer to cross it low, so he can enjoy it more, since to his opinion, high priced tweeters should be crossed low, since the price is high and improvements in high priced tweeters is often the low range.
Jay correct me if I'm wrong on this.
On the other hand, Zaph is a value oriented speaker builder, but again, he has nothing against high priced tweeters when the performance is there, except he doesn't like the excessive cost. To him, high price and high performance is okay because it is still high value.
He even use some higher priced drivers in one of his latest design.
Let me introduce the ZRT :
http://www.zaphaudio.com/ZRT.html
It uses the ( 220$ ) Scan Speak 6600 AirCirc tweeter and two ( 2x 225$ ) Scan Speak 18W8531G Revelator woofers in the more elaborate enclosure. So about 700$ per speaker, only in drivers, no crossovers included.
Here's the full part list ( 1430$ ), sold as a kit by Madisound that helps you save a bit of money, which you prefer over PartsExpress because it doesn't sell SoundEasy :
http://www.madisound.com/catalog/product_info.php?manufacturers_id=148&products_id=8448
Now, is Zaph that bad ? He's even a recognized designer on Madisound.
Now, Zaph tested the 7000 two months before he made the ZRT. 7000 was tested in march, ZRT was out in may. Why he used the 6600 ? Because, maybe to him, the 6600 sounds better than the 7000, or the difference is negligible.
It's true that I may not have the best arguments, but to me if you want the best sounding tweeter to cover the last two octaves, you use a 3/4" tweeter. Better off-axis response, easier to have a flat response to 20 kHz since it's smaller and will act as a point source longer. Mms is often lower, lower moving mass, a better piston-like performance.
Take for example the Dayton ND20FB at 8$. Flat to about 32 kHz, flatter than the 7000 which have a dip at 15 kHz and a peak at 20 kHz. The CSD is better for the Dayton ND20FB if you look over 4 kHz than the 7000, way better, because the 7000 have nasty resonances between 10 kHz and 20 kHz. Now, for distortion measurements over 4 kHz... Slightly lower 2nd harmonic for Dayton ND20FB. Slightly lower 3rd harmonic for the 7000. 4th harmonic it's even. 5th harmonic is very slightly better for 7000.
That's a comparison between an 8$ tweeter versus a 400$ tweeter, that if crossed correctly over 4 kHz, could sound about the same, in my humble opinion. The 8$ tweeter might even sound better because of the flatter frequency response.
smellygas said:
You are right, speculation, but in my opinion it looks like that.
I think it's a mainstream professional software package.
First, since the price is affordable, there's probably much more people using SoundEasy than any other package, so it's mainstream. Maybe not mainstream at Bose or EgglestonWorks but mainstream anyway. As I said earlier, those are more marketing companies than anything, so I don't put much faith in what they believe is good. Since they sell high priced loudspeakers, they need to use high priced software, so it looks good to the customer. They see that money goes toward building better speaker (in theory), in practice the money goes in their pockets...
Then, I think it's professional enough. To compare Windows Paint and Adobe Photoshop is really harsh. A better comparison would be The Gimp and Adobe Photoshop. Even if The Gimp doesn't cost 1000$ it is still a professional piece of software to me. It can do anything that Adobe Photoshop can do.
Yes, Madisound carry more expensive drivers, I agree with you on this, but PartsExpress have some expensive brands too, like Aurum Cantus, B&C, JBL, Morel, Radian Audio, Usher...
I disagree on this, most of the time, the ear is fooled by what seems to sound best. Tube amps is a good example where people take 2nd order distortion as pleasant. On the other hand, it may depend on what is your definition of the best sound. To me, the best sound is accurate sound.
For this very thing, measurements shouldn't be ditched because of a listening test, they should be used together, and personally I will always favor measurements. Maybe it's my "scientific" side.
So you don't know the story, sorry for this example then. Here is some background information. Stereophile wasn't paid by Bose. They were brought to court, they received cease and desist letters. In face of high legal cost and maybe bankrupcy versus the Bose legal and marketing monster machine, they dropped the ball and retracted their article then never reviewed a Bose system after that. That's why I said they shut them up.
In that same paragraph, I was talking about manufacturers who can blackmail anyone to get their point across and keep their reputation intact. I used Stereophile as a reputable publication to show that even them got shut up and backed out instead of fighting and losing energy and money for nothing. They knew it wasn't worth it to fight against Bose converts.
I also said that to further support my point that big manufacturers are there for the money first, and sound is last. So you get a big marketing machine instead of a big engineering department.
As you can see, no contradiction there on my part. Never said Stereophile was a bad publication. They weren't paid by Bose or anyone as far as I know.
Again, that could happen in peer reviewed journals also. In fact, I'm sure it happened countless times already. World isn't perfect and corruption is everywhere.
Look at my comparison above, I think it's quite fair to say that the Dayton ND20FB at 8$ "could be" a 7000 beater, when used exclusively over 4 kHz. I'm sure we can find others.
You never made the assertion, but you took the opposite point. No better than me on this.
simon5 said:
He may have his reason but your assumption is not reflecting things as they are. I posted some of the performance figures for these tweeters. Performance is top notch but of course the step up in price does not match the improvement in performance, but that's unfortunately the way things use to be in SOTA technology.
And we have not started to look at thermal behaviour yet. Well I have but Zaph doesn't seem to do such tests. The diamond tweeters have less thermal compression than most other dome tweeter as well.
Bad analogy. The jury is still out on ECM8000 but when you measure things, the measuring device needs to have some margin agianst the things being measured, otherwise as I have mentioned earlier, you can not trust the results.
Serious measurements always start with measuring the measuring equipment to verify that the set up is up to the job. Not doing that is amateurish. A friend I use to "work" with is a professional testing engineer measuring industrial and medical equipment before they get the governmental approvement to go out on the market.
/Peter
simon5 said:Again, you need to resort to personal attacks
Again, it's not an attack. It's a statement of fact, which you just acknowledged. I bring it up because you think I don't always understand you. I am explaining why.
As I have already stated explicitly, this is NOT a best tweeter thread. And I have mentioned at least 4 times that I was looking for a) people who have experience with the 7000, and b) for a comparison with a better tweeter.
Again, speculation.
I'm glad that you acknowledge that making a bold scientified statement without a shred of proof is unacceptable.
Thank you for acknowledging this.
[Discussion re: Jay]
If Jay has followup comments, I will be happy to address them directly with him, not you.
[Speculation about what Zaph thinks]
Jay made a negative remark about the high price of the R2904/7000, which is where my comment came from. However, your interpretation into the mind of Zaph is 100% speculation.
Let me introduce the ZRT :
http://www.zaphaudio.com/ZRT.html
[Zaph designed a kit for Madisound using high-end drivers]
Fair enough. Perhaps it is only you and Jay that despise expensive drivers. However, Zaph's decision could have been driven by any number of things. For instance, the 6600 may have sounded ALMOST as good as the 7000, and therefore not worth the big price jump. It is also possible that Madisound told him that he would be required to use the 6600 in the design, and to just make the crossover for it, etc. You don't know this. This is why speculating is useless, yet you keep doing it as if it has some merit.
Thank you, but that is a completely different topic.
Here we go again. How do you know which sounds better without LISTENING? A CSD plot, amplitude vs. freq plot, etc. can only help you PREDICT which one will sound better. That's it. Rarely are you ever going to be in a situation where one tweeter is substantially better in every single last aspect of the measurement routine. Ultimately, you have to LISTEN.
SG
simon5 said:
Speculation and your "opinion" really carries no weight. Sorry.
That's fine. If you want measurements done on a package that is not commonly used and may have not consistently comparable results with the commonly used packages used by professionals, that's fine. This is really a moot point.
Let me ask you something. How do you know if an abnormal measurement, say F3 distortion, results in good/bad sound quality? You have to LISTEN to varying amounts of F3 distortion until you HEAR a change. You have to conduct controlled LISTENING tests to determine if there is a differencewhe F3 distortion is introduced. In fact, every measured quantity should be correlated with perceived sound quality, which we LISTEN for. Some measured quantities are not audible. For example, every exotic cable brand can provide a list of measurements that make their cable sound different - but NONE Of them make a difference when you actually LISTEN. Which is why ultimately, if there is no audible difference between two entities, then the meaurements are irrelevant.
Pan said:
I would like to see those results please, it interests me, that diamond tweeter always ticked my curiosity.
And for the Behringer mic, the jury is still out on this one, since we don't know if Zaph modded his Behringer or not. You're speculating he didn't but maybe he did.
Thanks !
smellygas said:
The problem I have with that is even if you acknoledge that it's unacceptable to do it, you still do it, like us.
I couldn't find the post where Jay made a negative remark about 7000 price. I only found one where he made a negative remark against using an expensive tweeter for only the top two octaves, since it's a waste to him, and he would prefer to cross it low. I just re-read that whole thread again.
You also make speculation in there, that Madisound could push around Zaph and make him choose a tweeter over another. That's not the case at all. By the way, it would be in the interest of Madisound to use the 7000 in Zaph's kit because that would mean higher profit margin.
Zaph choose drivers, then design the crossovers, then Madisound follow with an exact replica, in accordance with Zaph, since publishing a complete Zaph kit means profit because alot of people will build them.
If you read Zaph regularly, you'll even see that he's often mad at Madisound, since they regularly fail to stock enough drivers for his kits, so he need to pull a design or suggest an alternative. So, Zaph relation with Madisound is not always perfect.
Maybe a different topic, but you wanted me to back up my earlier saying that it's easy to build a tweeter to cover the two top octaves, and that could be done on the cheap, how about 8$ ?
Well, to me, measurements can tell alot. I also said that it "could" sound the same, or even better. I didn't say that it will. I said it "could" be a fair comparison if crossed at 4 kHz. I didn't say that it will win. As you said yourself, I predict it "could" sound the same, or even win.
Then apply the same standard to you.
It is commonly used, did you visit the Yahoo user list ? Another user here compared SoundEasy to LSPCad Pro, since you can do the same thing in both software. Is LSPCad Pro an acceptable professional software to you ? It's expensive so I guess yes.
Now you speculate again that it may not have comparable results with the commonly used packages. First, it's commonly used, and second, you don't know. To speculate myself, I'm quite sure it has comparable results, thank you.
What's a loudspeaker professional designer again ? To me, it could be someone with a MBA from Havard with a Masters in marketing. As I said, what they use can't be used to prove the superiority of a software package. Also, some people here have more experience than those professionals. Not saying all loudspeakers professionals are bad, but you can't use that to prove your point.
Well, before you said to ditch measurements if measurements don't correlate because to you, a tweeter sounds better than the other, so measurements which say otherwise are worthless and should be ditched. I completely disagree with this, and you just got fooled by the tweeter and let your subjective opinion win.
Now, you say that measurements are irrelevant if when you compare two tweeters and you perceive no audible differences in a controlled listening test, since the measured differences are too small to be audible. I agree with this, our ears aren't perfect.
I agree, but remember that some amateur DIY speaker builders here, like you like to call them, are very experienced and talented. We even see professional loudspeaker builders here, for example, Tom Danley.
Well, to me, a tweeter that measures the same, should sound about the same. I agree that measurements aren't perfect and we didn't find all ways to measure a tweeter to always correlate with what we hear, but to me, we are not that bad.
I'm not concluding anything. I'm saying it "could be" a fair comparison.
simon5 said:
Your grasp of English is worse than I thought. When you draw a hard conclusion (i.e. the 6600 is better than the 7000) from your own made-up theory (i.e. otherwise, Zaph would have chosen the 7000 instead for his Madisound design), THAT is speculation. I provide equally likely alternate explanations (i.e. perhaps Madisound contracted Zaph to make a kit with the 6600, regardless of which driver was better), to illustrate that you cannot draw conclusions from speculation. I'm sure you're a very intelligent person, but I don't think we're discussing this at quite the same intellectual level. I really shouldn't have to explain what I just did. From this point on, with you specifically, I won't respond to anything that deals with your speculative theories. It's a waste of time because I don't think you understand.
I didn't understand any of that.
I don't agree that SoundEasy is a professional software package, and I do not wish to debate it with you here because, quite frankly, is 180 degrees off-topic.
Well you're entitled to your opinion. If you think you can determine which speaker sounds best based on evaluating a limited number of measurements published using low-end equipment on an internet website, that's fine. People have a lot of beliefs here, some of which may or may not be true.
simon5, thanks for trying to answer my original question. I think that it is very nice that you feel that a $20 tweeter could sound better than the 7000, and that you based this on zero personal experience, zero listening, and ONLY on a very narrow reading of a some measurements done on a website. Thanks, but that's not what I'm looking for.
SG
Pan said:
Here's my modeling result. The modeling scenario is as follows. First, suppose that a driver has a distortion profile of its 3rd order harmonic products that shelves from -50 dB at 1 kHz to - 70 dB at 10 kHz at a rate of -6 dB/oct, responding to fundamental tones at a reference SPL (scaled at 0 dB). We assumes three different cases where mic distortion levels are different. In all three cases, its distortion has an increasing profile at a rate of 6 dB/oct, but they differ in their overall distortion level with 20 dB increment. See the first column of the figure below.
Now, due to the nonlinearity, we can't assume that these two distortion products are summed in any fixed phase. In the model we may assume that their phase is completely random and independent across frequencies. But this is unrealistic since it doesn't take account any natural continuity of phase change. So, in this model we assume that phase at 1 kHz (initial point) is random, that is, randomly sampled uniformly from 0 to 360 degrees, but from 1 kHz and above, it changes at a rate of alpha / oct. Parameter alpha is also selected at random from 0 to 180 degrees. Thus, for each random sample, we have a random initial phase at 1 kHz and it changes over the frequency at a randomly selected rate. This simulation is repeated 1000 times with different, random initial phase and rate parameter values. In each session, the summed distortion profile is computed across all frequencies.
The results are shown in the figure below. Shown from second to sixth columns are 5 samples out of 1000 just to show what individual situations look like. The last column summarizes results of 1000 samples by showing 5th and 95th percentiles across frequencies. As you can see, in Case 1, the driver's distortion profile is recovered quite accurately despite the mic distortion. In Case 2, the mic distortion starts to cover the driver distortion in the higher frequency range, and in Case 3, the mic distortion almost completely swamps the driver's distortion profile.
What do these results mean? If the situation with Zaph's HD sweeps is like Case 3, we will definitely see a similar pattern almost all the time. Upon seeing Zaph's plots, we know that this is not the case. How about Case 2? Do we see a consistently similar pattern in a certain range of Zaph's sweep, which is suspected as being from his mic's distortion profile? I looked at all his tweeter mishmash HD plots again. I couldn't find any such symptom in the range of 1.5 kHz to 4 kHz, in which the tweeter's fundamentals are set to generate the same reference SPLs. We only see all different patterns in this range across different tweeters.
However, there are two possibilities that we might see these different patterns even if the situation is like Case 2 or 3. One is that Zaph uses different mics with different distortion profiles every time he measures a tweeter. Another is that Zaph made up the plots. I doubt these possibilities.
-jAy
An externally hosted image should be here but it was not working when we last tested it.
P.S. If anyone's interested, I can post my Matlab file I coded for this simulation.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.