I actually don't own the 830983 yet, but I had heard it's slightly bigger cousin the 2.5" NE85 in a studio monitor before. The 830983 looked very interesting to me since it has a slightly flatter response in the published graph and was quite cheap. I will definitely be buying some now.
Yeah I picked H which was the SB65, but I found D/H/F all very close to each other in this test. It was very hard to pick them apart. I knew the 830983 was in that collection (DFH) but just could not find it. D was my second choice in the blind.
I think it's a mix of vacation time and maybe that people are putting it off because it's too much work to go through so many samples, if you are going to be even slightly serious about it. I know that is why I was too late.
It's too good an idea to let die though. Maybe fewer drivers next time X?
It's too good an idea to let die though. Maybe fewer drivers next time X?
Interesting that I selected the TB (F). I own several TB drivers and wonder if I selected what I was accustomed to enjoying? Also interesting that it appears to have the worst frequency response of the bunch. My top three were in the top three (F, D, B).
Additionally (and more importantly), I noticed I preferred the drivers that all dipped between 5kHz and 10kHz (I preferred them in the order of how deeply they dipped!).
I started building speakers because I was unhappy with excessive sibilance and brightness. This test reinforces my preference for drivers that are less bright.
Thanks for another great thread XRK!
I just listened again to the test knowing which driver was which and picked the one that 'took the edge off - the TB'. Anything too bright was eliminated.
Additionally (and more importantly), I noticed I preferred the drivers that all dipped between 5kHz and 10kHz (I preferred them in the order of how deeply they dipped!).
I started building speakers because I was unhappy with excessive sibilance and brightness. This test reinforces my preference for drivers that are less bright.
Thanks for another great thread XRK!
I just listened again to the test knowing which driver was which and picked the one that 'took the edge off - the TB'. Anything too bright was eliminated.
Last edited:
Odd that measured responses are so removed from datasheet graphs. In A the response is largely similar to published, except the weird dip.
Not criticising the measurements, but what would cause this?
Interestingly I think all the drivers show a discontinuity at one frq, albeit some peak, others dip.
Could some of this be artifact of the nautiloss baffle? Room refection? Diffraction?
More than a touch surprised that all the drivers responses look very different to.published, except maybe the FRS5X, and even that shows a hole in the rising response that the published graph shows
Even more strange considering the blind consensus that most voters blindly preferred the flatter frequency response (previous tests), is why did driver E fare so badly in this poll???
Not criticising the measurements, but what would cause this?
Interestingly I think all the drivers show a discontinuity at one frq, albeit some peak, others dip.
Could some of this be artifact of the nautiloss baffle? Room refection? Diffraction?
More than a touch surprised that all the drivers responses look very different to.published, except maybe the FRS5X, and even that shows a hole in the rising response that the published graph shows
Even more strange considering the blind consensus that most voters blindly preferred the flatter frequency response (previous tests), is why did driver E fare so badly in this poll???
Last edited:
Now that you know D is the poor Peerless, does it still matter? Do you still want driver D? 😀
While I've been generally following the thread, I didn't vote this time because of the time investment needed to provide a fair assessment of this many drivers. I agree that fewer drivers might make it easier for people to participate (and easier to make comparisons between the various drivers as well).
It's still a great idea (and good fun) but the more complicated and involved it gets the less likely people will be to invest the time and effort needed to fully participate.
Great thanks xrk971 and contributors round three, look forward if we get a finale with top two drivers from each round as you had talked about.
Thinking loud:
Okay winner B was the B80 a relative high cost driver as it was for winner in round 2, before revealed had hoped it was one of the cheaper ones as in round 1.
Nothing wrong with wriggled response if one prefer that or repair with electric/acoustic EQ somewhere in chain, but some of the worst wriggled ones below seems three MA drivers, interesting they at same is some of the best on paper to extended low end frq response as was it a bigger diameter 8" FR, seems to cost a prize this low end extension in that MF/HF area then looks like it was a 8" FR plot.
Round 1:
Round 2:
Round 3:
Thinking loud:
Okay winner B was the B80 a relative high cost driver as it was for winner in round 2, before revealed had hoped it was one of the cheaper ones as in round 1.
Nothing wrong with wriggled response if one prefer that or repair with electric/acoustic EQ somewhere in chain, but some of the worst wriggled ones below seems three MA drivers, interesting they at same is some of the best on paper to extended low end frq response as was it a bigger diameter 8" FR, seems to cost a prize this low end extension in that MF/HF area then looks like it was a 8" FR plot.
Round 1:
Round 2:
Round 3:
Attachments
Last edited:
Well yes and no. I wouldn't buy the Peerless based on published specs, the SB I perhaps would. Either way, for me it was a very close thing and I did alternate the top 3, BD and H, eliminated B. It was a bit of pot luck deciding between the last two.
Also I am kind of shocked how bad an impression A left with me, since I own them, and how far removed the sound was from how I hear them at home.
Again, I do think all the 2" were very compromised wrt SPL and THD performance, far less so with the SB driver however. I continue (perhaps blindly) to believe that the FRS5X had little chance with its SPL limitations and was probably over worked. Perhaps others too.
Perhaps it was too many drivers that kept people from taking the poll. I will try to keep is more manageable for final round. That may mean not being able to have a surprise driver. There was a request to put all three similar Vifa/SS drivers in final and that would mean adding the TG9FD. That would bring it to 7 drivers. Let's hear your thoughts on how many drivers you are all willing to do for the final round.
Thanks
Thanks
Which ones are you saying are different from their published specs (besides the MA drivers which are known not to resemble factory measurements)? The Nautaloss was shown to have same response as an open baffle as described in round 1 so it is not the enclosure that is causing the differences. Nautaloss is a very transparent open enclosure that doesn't add anything or take anything away which is why it was chosen as the test enclosure. Furthermore, the mic is now calibrated by a third party lab so should be good.
Id say 4 maybe 5 at most.
8 was a bit much, but understandably you want to include many options, despite that wish it probably clouds and assumptions one could make of the results.
The only assumption I would attempt to make is that (guessing) the SB probably has the best THD and FR compromise of the bunch.
Id love to see the THD of this lot.
8 was a bit much, but understandably you want to include many options, despite that wish it probably clouds and assumptions one could make of the results.
The only assumption I would attempt to make is that (guessing) the SB probably has the best THD and FR compromise of the bunch.
Id love to see the THD of this lot.
Well we crossed posts X.
Well there is a couple of wiggles in the FRS plot that aren't apparent (or perhaps just not as obvious) on their published data.
I am just a bit surprised how awful they sounded, when I deliberately went back and tried the original track on my own FRS5X to see if I could identify which letter driver they were. I didn't pick A...so in my own room, with my own amp and FRS5X, playing original track, I found a world of difference.
Just bewildered me a touch, as they sound quite reasonable in my room, even without EQ, but sounded awful.in this test.
This is why I'm.lead to believe that THD would show alot,and COULD explain this, but equally it could be that they just sound totally.different in your home as opposed to mine
Well there is a couple of wiggles in the FRS plot that aren't apparent (or perhaps just not as obvious) on their published data.
I am just a bit surprised how awful they sounded, when I deliberately went back and tried the original track on my own FRS5X to see if I could identify which letter driver they were. I didn't pick A...so in my own room, with my own amp and FRS5X, playing original track, I found a world of difference.
Just bewildered me a touch, as they sound quite reasonable in my room, even without EQ, but sounded awful.in this test.
This is why I'm.lead to believe that THD would show alot,and COULD explain this, but equally it could be that they just sound totally.different in your home as opposed to mine
Last edited:
Did you listen through headphone or speakers? Is there any possibility that your speaker has too much of what you don't want?
Actually the highs of the Tang Band (F) is fatiguing. This HF problem seems to be well known if you read some users' review of the driver. At first listening I like this driver the most. With more listening I found this driver has too much 2nd order distortion, which in further listening, as always, will show a hint of fatigue (found it on 10:5 of the Clapton clip).
Like my prediction, the preference would be shared between 3 or 4 drivers. That's because the other drivers are not in the same league (I scored them below average). One of the reason I voted for D is because I don't want to see D/F/H have the same score (4) because I sensed that Peerless tends to be preferred by many members in the blind listening.
And between these 3 drivers, each has their own strength, so it is just about subjective preference imo. But B is objectively the best. But I have compared it with 10F and there's no way it can win as their strengths are similar.
If I have the chance I will purchase the Peerless. I believe it can be modified to become a great midrange for a 3-way. The problem with it is imo, Peerless has tried too hard to boost it's low frequency performance.
I believe that many has listened to the files. It is just that they don't want to vote for several reasons. Many has expressed their difficulty to choose in this third round!
It took me quite some time, after cross comparing the big four, to find out that B is objectively the best. And even when I knew that B is objectively the best, I had and have no interest to purchase it, even if the price is lower.
I myself didn't want to vote, but I thought that it would be better to make a difference.
Now the results are in!
I was one of the people that chose D and looking a the measured responses I can see why.
First of all I am not surprised that I didn't vote for the SB driver. Although very nice it does have a slight rising response and with my objective head on I simply discarded any driver that had an added amount of treble lift that wasn't present in the original recording. In actual use the fact the SB has this though is somewhat perfect. Due to the fact that full rangers inherently beam quite early on the slight rise means you can move around a little off axis without things becoming too dull, or it gives you the option to listen directly on axis for a brighter balance if it would be desired.
On the Alicia keys track B was my favourite but I then eliminated it using the strings track because it added a forwardness and slightly nasal quality that's not present in the original recording. The original sounds soft and inviting and B unfortunately didn't convey this.
Out of the others the Peerless, SB and TB remained laid back, which isn't a surprise. The dips associated with the 1-3kHz range no doubt were largely responsible. The SB was quickly eliminated due to the bump at 10khz adding a degree of air not present in the original and then it was down to the TB and Peerless. This was difficult to differentiate between but I managed to identify a touch of extra high frequency energy with the TB that was not present with the Peerless, so out went the TB and I voted for D.
I am not at all surprised by this result.
One thing that bothers me though is the dips in the response of some of the drivers around the 1-3kHz mark. Especially with the SB and Peerless. Both of these drivers are stiff, rigid metal coned drivers. They do not exhibit any resonances within this range as their cones are acting as pure pistons. Any response aberrations around this this frequency band will surely be due to either diffraction or possibly rear wave issues. Both of these drivers are of the same size too and have very restricted space behind them with which to breath from. I do wonder if this has anything to do with it.
Aside from that though, the test once again shows that people do prefer, oh the whole, drivers with the flatter frequency response.
Xrk thank you once again for carrying this out. As to any future tests I don't think shrinking the sample size is strictly necessary. What made this test take a lot more time was the fact that so many drivers actually sounded very nice. Previously it was easy to eliminate the majority, this time it was not. So maybe one thing to do would be to limit the number of nice driver in each batch! I can only imagine what would happen if we tried to determine our preference for the top 2 from each batch!
I was one of the people that chose D and looking a the measured responses I can see why.
First of all I am not surprised that I didn't vote for the SB driver. Although very nice it does have a slight rising response and with my objective head on I simply discarded any driver that had an added amount of treble lift that wasn't present in the original recording. In actual use the fact the SB has this though is somewhat perfect. Due to the fact that full rangers inherently beam quite early on the slight rise means you can move around a little off axis without things becoming too dull, or it gives you the option to listen directly on axis for a brighter balance if it would be desired.
On the Alicia keys track B was my favourite but I then eliminated it using the strings track because it added a forwardness and slightly nasal quality that's not present in the original recording. The original sounds soft and inviting and B unfortunately didn't convey this.
Out of the others the Peerless, SB and TB remained laid back, which isn't a surprise. The dips associated with the 1-3kHz range no doubt were largely responsible. The SB was quickly eliminated due to the bump at 10khz adding a degree of air not present in the original and then it was down to the TB and Peerless. This was difficult to differentiate between but I managed to identify a touch of extra high frequency energy with the TB that was not present with the Peerless, so out went the TB and I voted for D.
I am not at all surprised by this result.
One thing that bothers me though is the dips in the response of some of the drivers around the 1-3kHz mark. Especially with the SB and Peerless. Both of these drivers are stiff, rigid metal coned drivers. They do not exhibit any resonances within this range as their cones are acting as pure pistons. Any response aberrations around this this frequency band will surely be due to either diffraction or possibly rear wave issues. Both of these drivers are of the same size too and have very restricted space behind them with which to breath from. I do wonder if this has anything to do with it.
Aside from that though, the test once again shows that people do prefer, oh the whole, drivers with the flatter frequency response.
Xrk thank you once again for carrying this out. As to any future tests I don't think shrinking the sample size is strictly necessary. What made this test take a lot more time was the fact that so many drivers actually sounded very nice. Previously it was easy to eliminate the majority, this time it was not. So maybe one thing to do would be to limit the number of nice driver in each batch! I can only imagine what would happen if we tried to determine our preference for the top 2 from each batch!
Yes there were more drivers that sounded good and this made deciding much harder this time. When I get back to my computer I will post HD and impulse function plots. The drivers were really not stressed - I think about 85dB at 0.5m or 79dB at 1m. HD of most was not bad but one had a peak of 3 or 4% at 700Hz even at tbis low drive voltage.
Curious, the A7.3 appears to have measured the flattest yet it was at the bottom. Why? Alpair can never catch a break during listening tests.
Jay, I listened using cheap Sony earbuds. I think Sony makes among the best sounding earbuds from the cheapies I've owned. As with drivers, oftentimes there's great sound to be had for less money 🙂
Of course these listening tests are fraught with variables but they are excellent and fun, IMHO. Regardless, they are useful in determining the 'flavor' of each driver (too bright, too dull, etc.) and for that the tests are extremely useful.
Jay, I listened using cheap Sony earbuds. I think Sony makes among the best sounding earbuds from the cheapies I've owned. As with drivers, oftentimes there's great sound to be had for less money 🙂
Of course these listening tests are fraught with variables but they are excellent and fun, IMHO. Regardless, they are useful in determining the 'flavor' of each driver (too bright, too dull, etc.) and for that the tests are extremely useful.
The way I look at it, the A7.3 measures like a roller coaster, and the flattest ones got the highest scores. Which is no surprise, but nice to see confirmed also in this test.
- Status
- Not open for further replies.