A Subjective Blind Comparison of 3in to 5in Full Range Drivers

[xrk971]

The Big Reveal

Here is a snapshot of the results circa 16:00GMT.

Here is a cumulative frequency response plot of all the drivers as level-matched to the bass unit in the FAST. The legend shows the drivers tested.

Here is the high resolution 1/48th oct smoothed with 1ms gate before and 4ms gate after data:

A = PRV 5MR450NDY
B = Vifa TC9FD
C = Dayton PS95-8
D= Faital Pro 3FE22r16
E = Dayton Audio RS100P-4 (paper)
F = MA CHN-70
G = Dayton Audio RS100-4 (aluminum)

** more data to come **

Thanks for playing,

The winner is the TC9FD. No surprises there...

 

[OMNIFEX]

Cheers xrk971!

Now I can place my order at Parts Express on the driver which will suite my requirements. 😀

 

[Philosophil]

Very interesting. Thanks XRK.

 

[pnix]

A = PRV 5MR450NDY
B = Vifa TC9FD
C = Dayton PS95-8
D= Faital Pro 3FE22r16
E = Dayton Audio RS100P-4 (paper)
F = MA CHN-70
G = Dayton Audio RS100-4 (aluminum)

And again it has been demonstrated that drivers with a smooth frequency response are preferred. Thanks for putting this test together X.

Frequency response is NOT meaningless. It is still the number one criteria for every driver. No other performance characteristics can make up for a good frequency response.

If the driver isn't smooth it needs to be equalized -- even a TC9. This works for some drivers while others are beyond repair.

 

[sippy]

If the driver isn't smooth it needs to be equalized

Really?

 

[sippy]

Very interesting. Thanks XRK.

+1

 

[mt490]

Intriguingly the Daytons didn't rank anywhere near as well as their regard.

I'm still wondering how the PRV's got the bass sucked out of them (terrible snare presentation) despite the bass being apparently the same. I voted for the PRV on the first time around for its other parts that it reproduced properly that barely any others did.

Also when the reference clips became available apparently the driver that I thought showed merits despite the obvious lack of highs was the RS100P-4. It seemed to be doing everything but the highs right.

 

[wushuliu]

And again it has been demonstrated that drivers with a smooth frequency response are preferred. Thanks for putting this test together X.

Frequency response is NOT meaningless. It is still the number one criteria for every driver. No other performance characteristics can make up for a good frequency response.

If the driver isn't smooth it needs to be equalized -- even a TC9. This works for some drivers while others are beyond repair.

Easy now. No need to start passing out pamphlets and knocking on doors. We're just having fun here, not looking to be saved.

 

[ra7]

Indeed, thanks XRK. Thanks for Pnix for putting together the offset chart, though I would like to see the first chart in post 281 with 10 db offset for each driver and highpassed (through REW) to 200 Hz.

Great stuff. Yes, the test involved some compromises. It would have been good to get everyone to listen in the same room, or at the very least through the same set of headphones. But then those tests have already been done by the folks at Harman (Toole, Olive, and so on) and they reached pretty much the same conclusion with statistical significance.

 

[alwineden]

Thank you XRK! I listened to the tracks through headphones directly from my laptop and then through an external DAC feeding my home system. I feel like I can hear some pretty big differences between tracks, and my subjective decision on which ones I liked best was agreeable in both mediums.

I liked B the best, A worst, distortion in F and the others were ok...

I bought a pair of MA chr70 metal cones about a year ago and was pretty disappointed. It wasn't until I did some AB testing through REW that I made the correct decision to give up on them; I was wasting time on something that would never work for me. For what it's worth, I really appreciate what you're doing ...transducers don't grow on trees.

Keep up the good work.

TM

 

[Mooly]

I listened through headphones too. Never tried anything like this before, hadn't heard (as in heard of 😉) pretty much any of these names before either.

There were obvious differences just randomly sampling them but A sounded fairly reasonable compared to some others and so I gave that 5/10 as a basemark comparing everything to that. After a couple of runs through and re-adjustment of the scores I ended up with:

A = 4
B = 3
C = 3
D = 6
E = 2
F = 1
G = 3

So there we have it 😀 It was D for me.

 

[alwineden]

Cool. The above comments were made before I saw the release.

As I expected, the flatest response curve worked best for me.

TM

 

[5th element]

I prefered E followed by G. G for a slightly brighter balance if wanted. The others were too bright for my liking. And the downward upper tilt on the RS drivers helped to fill out the thin midrange, all of them need more BSC imo.

The one I ran away from was F. The MA driver, no surprise it's the least linear.

I don't know why frequency response flatness comes as a surprise, it is the no.1 parameter required for good sound quality.

 

[tuxedocivic]

Dang can't find my notes. I picked D. B, C, and D were my favorites. E and F were my least favorite. A and G were mediocre for me. In any case, I wasn't surprised by the results (ratings) at all except maybe A.

I called E and F crap. A was shouty and bright. D was clear and neutral with maybe a slightly raised top. B was very neutral but just not quite a transparent as D. That's all I can remember without my notes.

 

[ra7]

I prefered E followed by G. G for a slightly brighter balance if wanted. The others were too bright for my liking. And the downward upper tilt on the RS drivers helped to fill out the thin midrange, all of them need more BSC imo.

The one I ran away from was F. The MA driver, no surprise it's the least linear.

I don't know why frequency response flatness comes as a surprise, it is the no.1 parameter required for good sound quality.

5th, many folks here have a hard time accepting that frequency response is the number one parameter for good sound quality even though it has been proven numerous times. I think it is because they are afraid of losing their following.

Btw, I had a very similar experience. I narrowed it to B, E and G. Finally, settled on G despite the muted high frequencies because it had the least number of resonances and sounded the most even through the midrange. I sent these comments to X last week, so not cheating 🙂

 

[irribeo]

Looking at freq respons: Why did E sound darker than G ? Or G brighter than E ?

 

[alwineden]

B, E and G for me too. The price of the Vifa is pretty extraordinary. I wonder how it would compare to a Scan Sp offering? I would be happy to make a PP contribution for future testing...

TM

 

[turk 182]

hum d appears to have the least deviation across the displayed bandwidth (flattest response) but did not win why?

 

[ErnieM]

Dang can't find my notes. I picked D. B, C, and D were my favorites. E and F were my least favorite. A and G were mediocre for me. In any case, I wasn't surprised by the results (ratings) at all except maybe A.

I called E and F crap. A was shouty and bright. D was clear and neutral with maybe a slightly raised top. B was very neutral but just not quite a transparent as D. That's all I can remember without my notes.

My exact thoughts on this. Maybe the slight on axis rise for D in the top octave flattens power response a bit? It's still pretty flat throughout most of the range.

 

[wesayso]

B, E and G for me too. The price of the Vifa is pretty extraordinary. I wonder how it would compare to a Scan Sp offering? I would be happy to make a PP contribution for future testing...

TM

You may get that wish granted. X is already in proud possession of the Scanspeak 10F as well as the Vifa's more expensive brother the TG9FD10-08.

I'm sure we could think of more though 😀.

hum d appears to have the least deviation across the displayed bandwidth (flattest response) but did not win why?

I had rated D high at first glance, but for me it lost in a direct comparison to B. So G and E were second and third for me after B. So there is more than just an even frequency response I reacted to.

I hope we can up the quality a bit for a possible next round. Never was that fond of mp3 as a medium for audio.