Of cource it's possible those Hobby Hifi informations aren't that accurate. To me it sounds a bit funny starting to build ultimate loudspeaker and choose 2-way configuration to do it. 
I'm not damning this driver from this information. Or maybe from its price but maybe there is a reason. Mostly expensive stuff is just low production number hand made stuff that doesn't correlate that well to absolute quality of single unit.
Sure stuff gets rare and expensive while R&D costs rise and this leaves smaller selled units more production costs to cover so it's understandable at some point.
Jussi
I'm not damning this driver from this information. Or maybe from its price but maybe there is a reason. Mostly expensive stuff is just low production number hand made stuff that doesn't correlate that well to absolute quality of single unit.
Sure stuff gets rare and expensive while R&D costs rise and this leaves smaller selled units more production costs to cover so it's understandable at some point.
Jussi
ScottG said:
One example I could find posted on the 'net from the same magazine:
http://www.usheraudio.com/news/Hobby%20hifi/HOBBY%20HIFI%202004-2/images/6_jpg.jpg
BTW, especially the low frequency region of any CSD is suspect because of the processing. I don't pay too much attention to those. The way CSD's are processed and the way some parameters are used also end up with different CSD's. For instance, from the way they look this magazine is using some "apodization" (a slight windowing to the left end of the window used) that smooths out the decay. With this processing, dips on FR usually show an initial fast decay, but the ringing of them show up later on as time increases. The ATD driver has two small in amplitude dips at 400Hz and 700Hz, and note the CSD has an initial high slope decay at those frequencies, but then as time increases they recover back and show as continued ringing. What I am trying to say is, CSD's are more of a marketing tool, since they are not that easy to interpret, but easy to tweak and present the data in a nice looking way. Since drivers are mostly min-phase devices, instead of trying to interpret how much decay one sees on a CSD in how many msecs, compare the FR's. The FR that is smoother and flatter will have better decay and rise characteristics, given that they are minphase, which some exceptions they all are. And the FR of the ATD driver below 1Khz is not necessarily exceptionally flat and free of dips and peaks, but it is not bad either.
The peak and following dip just after 1Khz, resembles a typical cone-surround edge resonance, but in this driver it is particularly severe wrt other drivers. Mostly paper or poly drivers show this resonance. For instance the above link to Usher 8945W driver shows it: it has a dip right after 1Khz, which is accompanied with an impedance wiggle at the same frequency. But note on the Usher the FR there is a lot smoother, and correspondingly the CSD is better looking there, and also the amplitude of the impedance wiggle there is lower than the ATD. Metal cones like Seas Excel's are very good in that region, they don't show signs of this cone-edge resonance. ATD is the severest I have seen so far in that aspect.
In the distortion graph, as far as I can make from the blurred image, the ATD shows high K3 and K5 wrt some other drivers that use Faraday rings. In my experience high K3 and K5 usually means the driver has high amounts of IMD, which is audiable.
In my previous post, I said the 90db sentivity is nice, but I now realized this is a 4ohm driver, which explains the high sensitivity. A 4ohm 90db is not very special either.
To sum it up, the available information on this driver doesn't contain anything that shows it as a superior driver. It actually shows some deficiencies of the driver wrt to some other less costly drivers. It may sound better than the other drivers, that is subjective and can only be determined after listening to them. If it didn't have that price tag, and such distinguished looking cone surface, would you be still interested in it?
Feyz said:
One example I could find posted on the 'net from the same magazine:
http://www.usheraudio.com/news/Hobby%20hifi/HOBBY%20HIFI%202004-2/images/6_jpg.jpg
BTW, especially the low frequency region of any CSD is suspect because of the processing. I don't pay too much attention to those. The way CSD's are processed and the way some parameters are used also end up with different CSD's. For instance, from the way they look this magazine is using some "apodization" (a slight windowing to the left end of the window used) that smooths out the decay. With this processing, dips on FR usually show an initial fast decay, but the ringing of them show up later on as time increases. The ATD driver has two small in amplitude dips at 400Hz and 700Hz, and note the CSD has an initial high slope decay at those frequencies, but then as time increases they recover back and show as continued ringing. What I am trying to say is, CSD's are more of a marketing tool, since they are not that easy to interpret, but easy to tweak and present the data in a nice looking way. Since drivers are mostly min-phase devices, instead of trying to interpret how much decay one sees on a CSD in how many msecs, compare the FR's. The FR that is smoother and flatter will have better decay and rise characteristics, given that they are minphase, which some exceptions they all are. And the FR of the ATD driver below 1Khz is not necessarily exceptionally flat and free of dips and peaks, but it is not bad either.
The peak and following dip just after 1Khz, resembles a typical cone-surround edge resonance, but in this driver it is particularly severe wrt other drivers. Mostly paper or poly drivers show this resonance. For instance the above link to Usher 8945W driver shows it: it has a dip right after 1Khz, which is accompanied with an impedance wiggle at the same frequency. But note on the Usher the FR there is a lot smoother, and correspondingly the CSD is better looking there, and also the amplitude of the impedance wiggle there is lower than the ATD. Metal cones like Seas Excel's are very good in that region, they don't show signs of this cone-edge resonance. ATD is the severest I have seen so far in that aspect.
In the distortion graph, as far as I can make from the blurred image, the ATD shows high K3 and K5 wrt some other drivers that use Faraday rings. In my experience high K3 and K5 usually means the driver has high amounts of IMD, which is audiable.
In my previous post, I said the 90db sentivity is nice, but I now realized this is a 4ohm driver, which explains the high sensitivity. A 4ohm 90db is not very special either.
To sum it up, the available information on this driver doesn't contain anything that shows it as a superior driver. It actually shows some deficiencies of the driver wrt to some other less costly drivers. It may sound better than the other drivers, that is subjective and can only be determined after listening to them. If it didn't have that price tag, and such distinguished looking cone surface, would you be still interested in it?
I agree completely. This is a "boutique" driver. That's all. On the surface, without doing detailed measurements, this driver is not any better than the cream of the other drivers commonly available to us at significantly better prices. It probably could be used very well in a high end system. But it does not appear to be objectively superior.
Sorry, that's just the way it is.
serengetiplains said:
Will I buy this driver to test? No way. Nothing in it's design or published data makes me think it's worth the asking price.
But, by all means, send me a driver to test. If it is able to distinguish itself in any way, I will post the data with a public apology.
Now, I have very clear and defined criteria for what constitutes a good driver and you can get that by reading my site. You may not buy into that. I'll accept that. Something may be worth $500 to you without having any objective evidence that it is better.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.