You cannot *equalize* those broad peaks & valleys as they are alterations given by cone break-ups.
To extend the concept, you cannot EQ anything....!
And, oh yes, you can alter the original sound before the speaker but it's not supposed to work like that
To extend the concept, you cannot EQ anything....!
And, oh yes, you can alter the original sound before the speaker but it's not supposed to work like that
How did you get these percentage..
The price difference is huge.. but that difference is clearly measured and can be heard.. also we are talking about one hundred dollars which is not so much for the audible difference..
I believe I can answer an honest: "Yes" and "Yes" to that question for the driver central in this thread. I needed extra measurements to get me to really believe in them though. Found that data published in "Klang und Ton" February 2010, 2 distortion plots, one at 85 dB and one at 95 dB and a waterfall plot. Zaph's measurements agreed with those measurements but I had already made up my mind at that time.
Granted, I use EQ, but still needed the numbers and graphs to believe in my build that took me over a year to complete. I wouldn't have done it without the graphs and numbers on this particular driver.
EDIT: Just realised: maybe the first "Yes" isn't a complete "Yes" as I needed extra measurements to convince me... The factory data did however help me to investigate further.
Last edited:
A wild subjective guess based on how close the data look. One could do a differential analysis between the two curves relative to a standard and assign an RMS variation to a "goodness" of fit to quantify how close.
Your scaling is wrong, once again. The two breakup peaks in Zaph's and my data for the TC9FD align if you do it right, you really need to stop scaling stuff via photoshop - how accurate do you think that is, especially given the fact that you don't show tickmarks for the data that you scaled from? Do you think photoshopping data is any less accurate than the so-called inaccuracies of my data as measured and plotted using real data on a real graphing program?
For a guy who says measurements are not important, and won't take his own measurements or share them, you sure spend a lot of time photoshopping other people's data.
I am just going to ignore your posts with photoshopped plots because as usual, they are unreliable, non quantitative, and inaccurate.
If you want to have an objective debate with quantitative data, either take your own (you have a TC9FD) or get the real data files from Zaph or whoever you feel like quoting in your quest to debunk the XRK data. I think this is a sentiment probably shared by others on this thread: just please stop photoshop scaling other people's plots - not accurate and not useful!
I can make the TC9FD REW data file available to anyone by request via PM so you can plot it for real.
Last edited:
Quickly I'm running awaaaay from the "Loudspeakers" forum. This is by far the worst and ugliest part of Diyaudio.
Guys I'm posting this as a member NOT as a moderator. I think that this has all been blown out of proportion, and I think every one needs to take a step backwards, take a deep breath, put their preconceptions behind them and get back to discussing, building and enjoying speakers!
I was going to upload a trace of zaph's TC9fd18 measurement and suggest that XRK also upload his, for anyone to compare the two in whatever program they like. I actually traced the mark audio datasheet one back when all this blew up and traced xrk's I did tricky things like apply a 4ms gate to the MA data as well... but in the end I decided it wasn't worth uploading it because it was not likely to change anyone's mind on anything that they had already decided.
I think it is time to accept that everyone has their own opinions, and that is in the end what they are. When we can't agree, we should agree to disagree.
Provided everyone puts their point of view across as their opinion, rather than as a statement of fact, then everyone should be able to get along. We are (hopefully) all adults, lets behave like them!
Tony.
I was going to upload a trace of zaph's TC9fd18 measurement and suggest that XRK also upload his, for anyone to compare the two in whatever program they like. I actually traced the mark audio datasheet one back when all this blew up and traced xrk's I did tricky things like apply a 4ms gate to the MA data as well... but in the end I decided it wasn't worth uploading it because it was not likely to change anyone's mind on anything that they had already decided.
I think it is time to accept that everyone has their own opinions, and that is in the end what they are. When we can't agree, we should agree to disagree.
Provided everyone puts their point of view across as their opinion, rather than as a statement of fact, then everyone should be able to get along. We are (hopefully) all adults, lets behave like them!
Tony.
Thank you Tony. The post above yours is exceedingly alarming for the FR forum, and I gather it is far from being an isolated example. Not good at all.
Wintermute,
I will happy to post my TC9FD REW file for 0 deg case. Hopefully it fits under size upload limit - might have to put .asc extension on it. I get vocal when p10 continually keeps making erroneous comparisons based on photoshop scaling without including axis and tickmarks of original data. You know the term "photoshopping" means to distort the truth or not portray the truth from an original image. That is exactly what is happening here. Without a reference coordinate system, you can photoshop any result you wish. I am just defending my data when this keeps happening. Count how many times p10 has photoshopped data vs his own real data. Zero real data so far from him for frequency response.
I will happy to post my TC9FD REW file for 0 deg case. Hopefully it fits under size upload limit - might have to put .asc extension on it. I get vocal when p10 continually keeps making erroneous comparisons based on photoshop scaling without including axis and tickmarks of original data. You know the term "photoshopping" means to distort the truth or not portray the truth from an original image. That is exactly what is happening here. Without a reference coordinate system, you can photoshop any result you wish. I am just defending my data when this keeps happening. Count how many times p10 has photoshopped data vs his own real data. Zero real data so far from him for frequency response.
Last edited:
I have to agree and this is NOT a preconception.
X's data is real, true and objective. Anyone not agreeing needs to present evidence. Doing acoustical measurements that hold up to professional standards is really easy these days. Get REW for free and a $70 UMIK-1 and you're good to go. This will get you started http://www.minidsp.com/applications/acoustic-measurements/umik-1-setup-with-rew
Last edited:
If those peaks and dips are minimum phase then you can equalize them perfectly with a standard parametric EQ. One would want to keep an eye on not making the response off axis worse though.
Everyone can have their own opinion.. but everyone opinion can not be right..
It is obvious that prices of drivers are not random, more expensive are better..
Almost all manufacturers provide false or incomplete informations about their products.. that is why they are trying to discourage any attempt of objective comparasion..
Hearing tests certainly makes sense .. but only when they come from well-known people in whom we trust.. on the forum only measurements can be objective..
I would dispute that the $12 TC9FD, the topic of this thread, has a better competitor in the under $60 price range.
That's something I can fully understand. At the end of the day they want to have made some money. What I don't understand is fellow enthusiasts discouraging any attempt of objective comparison.
I think the data if fair enough and representative of a specific set of conditions that may be met to differing levels by other people. I sure like to post measurements I've done, with the same caveat. Both poster and reader need to understand this IMO. I don't think this makes it reference-level data, but can certainly be useful if you know what you're getting into. IMO, this can still not let one know fully how a speaker sounds, yet can give important clues with correct interpretation.
As far as photoshoped graphs, perhaps a few of them were done wrong initially, but I don't see any issue if all scales end-up matching. The Photoshop process should not introduce any error when done right.
xkcd: Connoisseur
If that's not bordering on you spilling trade secrets, approximately what value of Qp would you say your vents have? I like to tinker with the various Q-parameters when simulating in WinISD, but have yet to implement such design more seriously.
I still do not agree with this. In other words our opinion differs.
A good while back I searched for and read a lot of tests, looked at a lot of data on these smaller full range speakers and saw pretty expensive speakers loose out to the TC9 in objective tests. Price is absolutely no guarantee of performance.
The reason for the low price on a driver like this, that has been around for quite a while has more to do with the numbers in which it has been produced than a lack of quality of materials or otherwise. Test a bunch of these and you will notice they are very close in performance to each other. To me that means the quality control is working as it should.
Wesayso probably has more experience here than most of us. The only other people I know of on this forum who have tested 50+ drivers of the same make and model is Melotheory and Koldby. I have about 16 of them, and yes, very close in performance indicative of well controlled process of manufacturing.
Just because they cost less due to efficiencies of mass production doesn't mean they are not as good. Automobiles are a good example of this, as well as examples of amplifiers. Take a $70 boxed TPA3116D2 amp and compare to some $1000 to even $2000 boutique amps.
Just because they cost less due to efficiencies of mass production doesn't mean they are not as good. Automobiles are a good example of this, as well as examples of amplifiers. Take a $70 boxed TPA3116D2 amp and compare to some $1000 to even $2000 boutique amps.
Last edited:
I fully agree with that..
But these drivers are not the only cost in speakers building..
You need a good woofer, well designed crossover with quality components, box itself can be the biggest expense..
Complete project will cost between 500 and 1000 dollars.. then difference between 50 for a couple of average or 150-250 for top drivers is no longer huge..
ScanSpeak 10F and Visaton B80 is far expensive and far better than TC9..
In this case price is absolutely guarantee of performance..