A Subjective Blind Comparison of 2in to 4in drivers - Round 3

[jeshi]

I just have a feeling like this might not end anytime soon. There will always be a new batch of interesting drivers appearing.

So I think the question is if we would prefer to do a shootout now of the "best of round1/2/3" and then include that winner in subsequent rounds, or if we want to postpone the final shootout (which could end up happening at round8). Remember round3 was supposed to be the shootout but new drivers appeared, and then round4 was going to be the shootout but new drivers appeared......

I would be ok with round4 being just TC9FD/TG9FD/10F/B80 (maybe include the previous second best drivers) and then hold off for round5 to be the winner of round4 + new drivers. and then subsequent rounds could be winner of previous round + new drivers.

 

[xrk971]

Thanks for the input folks. I just finished collecting data for all of the above mentioned drivers so I can slice and dice it anyway now. I think it may be tough finding a winner from the first 3 rounds. Remember how hard it was to settle on a winner in round 3? Now we are going to take the best of the best and try to discern a difference. Although my technique gets better this time with collapsing the stereo source to mono first, and using the 10F/RS225 Dagger Ref Monitor as the basis of the enclosure for all drivers tested this time. Let me think about it...

 

[BYRTT]

.....Let me think about it...

Of course X and shure your decision gets respect from here , its just that you asked in 416 that made us pop up with our pointing directions.

jeshi is right in that new drivers might not end anytime soon and then finale would not replicate all the highest ranked drivers and would continually need to be postponed.

Thinking loud a alternative input is don't do a finale round just normal rounds with new drivers, instead either make a new thread or update post one for every round where we can find updated sound clip with first and second preferred driver where they recorded under same conditions and with same tracks as you have updated your setup and system to use. This will give us the opportunity to listen example TC9FD and 3FE22r16 from round 1 up against never rounds under same reference conditions if you know what i mean, a form of standard for the reference setup. Also you have the great AMT tweeter that is probably kind of king in HF area, it would not be bad having a sound clip from that added too so we can compare what compromise we live with every day being full range lovers.

 

[afa]

Do the Tom Danley regen test and it will become quite obvious the most accurate driver.

 

[xrk971]

Of course X and shure your decision gets respect from here , its just that you asked in 416 that made us pop up with our pointing directions.

jeshi is right in that new drivers might not end anytime soon and then finale would not replicate all the highest ranked drivers and would continually need to be postponed.

Thinking loud a alternative input is don't do a finale round just normal rounds with new drivers, instead either make a new thread or update post one for every round where we can find updated sound clip with first and second preferred driver where they recorded under same conditions and with same tracks as you have updated your setup and system to use. This will give us the opportunity to listen example TC9FD and 3FE22r16 from round 1 up against never rounds under same reference conditions if you know what i mean, a form of standard for the reference setup. Also you have the great AMT tweeter that is probably kind of king in HF area, it would not be bad having a sound clip from that added too so we can compare what compromise we live with every day being full range lovers.

Byrtt,
All great ideas here. However, I feel like there is a monkey on my back to get a winner from Rounds 1-3. If I get that done then maybe we can move on to new drivers and we can always compare back as long as the setup doesn't change too much. I think I have settled on a setup that is close to optimal. I am thinking of doing Round 4 as a finale with the addition of a secret driver of my choice (from Rounds 1-3). This is a driver that I think sounds quite good, and even measures very well and deserves another shot (or should I say, another listen). Certainly if we look at how many shots the A7.3 got, there are other drivers that may deserve another listen.

What I can do is just make this Finale round very easy to do - 3 drivers plus TG9FD and one secret driver for 5 total. With 3 sound clips just 15 tracks total to listen to.

 

[xrk971]

Do the Tom Danley regen test and it will become quite obvious the most accurate driver.

I think this test would be great but you need to do it in an anechoic room otherwise the room effects will dominate and all you will hear are the bad untreated acoustics of my room. I am pretty sure some of these drivers could withstand more than the average recursive copying - normally things fall apart at 3 and if you make it to 5 you are doing very well. The point hear is that any non-flat frequency response gets added or subtracted each time. So a 3dB peak is 6dB on 2nd pass, 9dB on third pass, etc. 5 passes will be 15dB peak, which is quite intolerable.

 

[afa]

But aren't you doing measurements at 500mm

 

[afa]

Anechoic room? seriously?

 

[xrk971]

But aren't you doing measurements at 500mm

Yes, and fine for gated measurements but sound recordings are real time and allow full reverberation of room to be recorded.

 

[Barleywater]

I am recording data for the final Round 4 - which was originally meant to be the final round to select an overall winner from the top two of previous rounds.

Is this what we want or would you all rather have round 4 be another round with mostly all new drivers (I have 4 new ones) and can add some reference ones like the entire TC9/TG9/10F family or TC9FD/10F/B80 - top from each round ?

I have the three sound clips selected already and have converted the source to mono but will record clips in stereo. You can play in mono of you wish but stereo is closer to what it would sound like if you actually used your own ears if auditioning the drivers in my speaker lab.

The three clips will be:
Rock - Barracuda - Heart
Jazz - piano/bass/drums - Chick Corea/Further Explorations album
Vocals - Rachelle Ferrell - Satisfied

New drivers that are available for round 4/5 include:

Fountek FR88EX 3in aluminum cone Nd motor
Tectonic Elements TEBM46C20N-4B 3in BMR Paper cone Nd motor
Vifa TC7FD00-04 2.5in paper cone ferrite motor
Tymphany NE65W 2.5in aluminum cone Nd motor

How do these clips sound as mono mix played through one speaker?

Can you EQ for H4 microphones?

EQ below 400Hz needs improvement for direct comparison of reference clip to speaker clips. Here is mono mix comparison for clip-2 of reference v B:



In listening I look for speaker which sounds most directly like reference clip. For this round I ended up using steep 400Hz high pass to make reference clips and recordings sound more alike.

Also, recording at .5m, but referenced to level of 95dB at 1.5-2m would help bring more driver distortion into play.

I found mixing H4 stereo to mono suppressed some of the room factor.

I would be most interested in TC9FD/10F/B80; it is great comparison of similar cone geaometry and ferrite v neo motors.

 

[BYRTT]

In the air

round 4 seems in the air http://www.diyaudio.com/forums/full...ind-comparison-2in-4in-drivers-round-4-a.html

 

[xrk971]

You mean as "up in the air" - who knows where it will go?

Honestly, these are all very very nice sounding drivers and I think most of us (Mr Goldenears Jay, excluded) will have a tough time sorting it out.

It will be fun though.

 

[jurjenh]

I've been wondering about the Hivi B3S and would be happy organising one for testing... Zaph seems to think quite highly of it, and it seems relatively available (at least I can get it here in NZ)

One of the questions on my mind is the performance at volume - pretty sure once you start getting louder, the selection starts thinning out, and comparisons would possibly shift to different drivers.

Also (just thinking here) could it be possible to do a "reference" recording of each driver using the best setup as per the testing to date - then do all the drivers and post them in a separate thread perhaps. Then when new drivers come along, they could be added and compared without having to redo all the tests.



Sent from my iPhone using Tapatalk

 

[xrk971]

I've been wondering about the Hivi B3S and would be happy organising one for testing... Zaph seems to think quite highly of it, and it seems relatively available (at least I can get it here in NZ)

One of the questions on my mind is the performance at volume - pretty sure once you start getting louder, the selection starts thinning out, and comparisons would possibly shift to different drivers.

Also (just thinking here) could it be possible to do a "reference" recording of each driver using the best setup as per the testing to date - then do all the drivers and post them in a separate thread perhaps. Then when new drivers come along, they could be added and compared without having to redo all the tests.



Sent from my iPhone using Tapatalk

Good idea - a reference library of how they sound in same setup. Lock down the setup.

 

[xrk971]

New speaker with B80 is now underway in case you are interested.

http://www.diyaudio.com/forums/multi-way/297171-rs28f-rs180p-b80-hole-filler-3-way.html

B&O Hole filler concept:

 

[Godzilla]

Looks like a nice, refined project. I've gone off the beaten path of full rangers for a bit and into pro drivers - just for curiosity. My collection of smooth sounding full range drivers (on baffles) are neatly stored away in the closet for another day. Big bruiser pro drivers with enormous waveguides sit atop my H-frame at the moment!

Your new project looks great so far!

 

[xrk971]

Hi Godzilla,
Nothing wrong with big bruiser pro compression drivers on a waveguide. I have done that as OB before and it sounds great. Here is PRV 5MR450NDY mids, PRV comp driver and WG, Eminence Beta 12cx woofers (no tweeter). With miniDSP I was able to do some cool things with the whole transient response a la Harsch XO. Going down memory lane...

http://www.diyaudio.com/forums/multi-way/277691-s-harsch-xo.html