Maybe this is OT, but why not add a real tweeter and get over with all these issues ?
xrk971 already added a separate bass, so adding a tweeter should be no big deal.
Round 3 poll is a three way system ?
Then we'll be listening maybe 300 Hz -3 kHz midrange. There the differences between these small drivers should be small. Unless a driver is really bad.
It should be interesting.
.
xrk971 already added a separate bass, so adding a tweeter should be no big deal.
Round 3 poll is a three way system ?
Then we'll be listening maybe 300 Hz -3 kHz midrange. There the differences between these small drivers should be small. Unless a driver is really bad.
It should be interesting.
.
I hope anyone listening to this test realises he's listening to a mono track recorded in a real room. No doubt any of these speakers would be more impressive in your own room.
There are reasons why this part of the forum exists, and no tweeter is used (exept for a brief test in X's other thread on his 10F FAST). Not everyone is sold on the multi way concept.
There are reasons why this part of the forum exists, and no tweeter is used (exept for a brief test in X's other thread on his 10F FAST). Not everyone is sold on the multi way concept.
Yes, X. I could have voted A or E or B (tho the B part was on Logitech desktop speaker) with different subjective reasons:
1) a driver that I can make a great speaker with (ScanSpeak),
2) or the most natural and detailed (Peerless),
3) or the least fatiguing,
4) or the best in performance to ratio... (Vifa TC9)
Here is the TXT content, again (remember it was written after I heard B from round 1):
Rank#1 (A) Alpair 7 (or Peerless or Tang Band)
Most dynamics, vocal and especially instruments are tonally correct, detailed and clear. Typical of metal cone but the distortion is different. Easy on ears like paper but detailed like metal. Initially I guessed this was the mistery driver using cone material I haven't heard before (such as bamboo). Later I found out that the Tang Band is indeed using bamboo! But after a long listening I found this is as fatiguing as aluminum, less fatiguing than the other Aluminum tho. So which one is the better aluminum, Alpair7 or Peerless??? Probably Alpair because it is more expensive and has popularity 😀
Rank#2 (D) Vifa TG9
The detail/balance/tonality performance is close to (A) but the midrange/vocal is close to (E). I think I was listening to a fibreglass cone. Only Vifa and ScanSpeak are using fibreglass in this test.
Rank#3a (C) Tang Band (or Alpair7 or Peerless)
The sound is orange-to-apple to (B). The balance is very similar to (A) which has pronounced HF (on-axis of course! Should be flatter off-axis). (B) OTOH has less detail but surprisingly non-fatiguing and very easy on ears. After prolong listening of B and C, it is obvious that C is slightly fatiguing and B is not, so I change my mind: I prefer (B) to (C).
Rank#3b (B) Vifa TC9 (the mistery driver)
I asked myself "is (A) a straight wire?" The answer is no because the distortion is too high for a straight wire. "So what is the mistery driver? Isn't it logical if the winner from first round is included in the test?" So I downloaded the winner from first round (Vifa TC9) and listened. Intuitively I listened to (B) and I didn't need to think twice, they were exactly similar in tone (tho the clip from first round sounded better).
Rank#3c (E) ScanSpeak 10F
The midrange is "outstanding". A sound that I haven't heard before, probably better than what I have. I'm sure I was listening to fibreglass. Very clean, but the midrange is too much. High frequency is insufficient and tonality is typical of plastic/fibreglass so I gave low ranking. But with proper crossover and tweeter, this is my rank#1.
Rank#4 (H) Fostex FF105
This is one of the hardest to guess because it is quite "detailed" for a paper (so probably expensive paper such as Fostex). That background noise is so pronounced such to make it like metal and artificially detailed. I always want to run when I hear a Fostex...
Rank#5a (F) AHE
This is definetely cheap paper cone. Muffled, no detail and even worse, I thought I could hear the cone flexing. Objectively this is the worst but I didn't put at the bottom because it is honest with its price and appearance.
Rank#5b (G) Peerless (or Alpair7)
Typical sound of advanced cone material with flawed midrange. Very flawed midrange. I can see only 2 drivers with Aluminum cone. The Alpair and the Peerless. The cone is so responsive, and I accidently saw a very small Mms around 2.4g for Peerless... Initially I have no doubt that this is Peerless but when I look at graph from round1 I thought "what kind of company would design a driver with such a f***ed up FR??". May be this one also has similar FR issue like the one in round 1, plus IMD of metal cone? 60% for Peerless, 40% for Alpair7.
If my above prediction is correct, and by considering the price, then the true winner is B (Vifa TC9)! This is the driver that I definitely want to have! Hmmm... may be I should rank it #1? 🙂)
Last edited:
I guess I thought ranking #1 would have meant voting for it. But it seems your writing is just your thought process and mental scratch pad.
I won't be doing a 3-way Blind Comparo thread - that is too many cans of worms. The bass unit was to remove dependency on cabinet that uses driver to generate self-bass. The XO over to a tweeter is tricky and changes with each driver and not all will integrate well. It's not something people will do anyway in full range forum. Troels Graveson does use the 10F/8414 as a mid on his Discovery 3-ways - a beautiful speaker that probably sounds fantastic. If I were to build a passive 3-way monitor - it would be that speaker.
It's nice to finally see the results and to see that my choices tied in with what I was hoping they would tie in to.
A seemed to be doing quite well for a while and until comments were made. This is perhaps not surprising as it's purely pistonic with a slight rise around the 2-3kHz area that could give an impression of greater detail. The bowel shaped 3-13kHz area could also help to keep sibilance to a minimum although the breakup and peaking issues in the top octave too could give cause for concern.
As to why the 10F sounds superior to the TG, my guess would be control of what goes on under the dustcap. I've used the TG, but that resonance caused by the pole vent/dustcap is an issue. If Scan control this better in the 10F then that could easily account for it. Both drivers do sound nice at lower volume levels, the resonance issue with the TG only creeps in as the volume is turned up and more significantly at lower frequencies. My modded TG + tweeter combo sounds superb, the TG on its own didn't quite have enough extension for my liking and beamed like crazy.
I've always wanted to use the 10F as a midrange driver now I've got even more reason to want to try it out! And yes I agree Troels three way discovery should sound excellent. It's one of his better executed designs.
A seemed to be doing quite well for a while and until comments were made. This is perhaps not surprising as it's purely pistonic with a slight rise around the 2-3kHz area that could give an impression of greater detail. The bowel shaped 3-13kHz area could also help to keep sibilance to a minimum although the breakup and peaking issues in the top octave too could give cause for concern.
As to why the 10F sounds superior to the TG, my guess would be control of what goes on under the dustcap. I've used the TG, but that resonance caused by the pole vent/dustcap is an issue. If Scan control this better in the 10F then that could easily account for it. Both drivers do sound nice at lower volume levels, the resonance issue with the TG only creeps in as the volume is turned up and more significantly at lower frequencies. My modded TG + tweeter combo sounds superb, the TG on its own didn't quite have enough extension for my liking and beamed like crazy.
I've always wanted to use the 10F as a midrange driver now I've got even more reason to want to try it out! And yes I agree Troels three way discovery should sound excellent. It's one of his better executed designs.
I am kicking myself for not including the SB65WBAC25 as the surprise driver. It may have given some of these the run for their money. It's only a 2.5in driver though so that is why I did not include it as the title was 3in to 5in.
I really enjoy your observations, 5th Element. You and many others here obviously have a far more experienced ear than I could ever hope to have, and you also do an admirable job articulating your insights. Thanks.
Divorce trial tomorrow! Couldn't find the time to listen and compare but such an interesting thread to read! The Scan Speak is probably the most expensive driver which some may say justifies why it should win - and did.
The poor man's Scan Speak is the Vifa. Trickle down technology done right.
The other drivers vary in price. When you build speakers and spend money on drivers it becomes pretty apparent that $$$ doesn't always equal better... but in the case of the Scan Speak, it does.
Many thanks for the great diversion these past few days!
The poor man's Scan Speak is the Vifa. Trickle down technology done right.
The other drivers vary in price. When you build speakers and spend money on drivers it becomes pretty apparent that $$$ doesn't always equal better... but in the case of the Scan Speak, it does.
Many thanks for the great diversion these past few days!
I think I need to mention that you are correct in that:
(1) The "detail" is not real. Of course. There are many kinds of "false" details.
(2) "Honesty" is not exaggerating, so when there should really be no detail, there shouldn't be. In other words, a driver can have less details because it is honest (even to recording with missing details).
Many of the perceived sound can be seen from the step response (or just call it IR). But first I would like to relate "detail" with driver behavior:
In order to be able to show the detail, a cone should be responsive to the input voltage. First it should be light weight. Second, if it is not light enough (e.g. due to bigger diameter or Sd) the motor should be strong enough. If the cone is light and the motor is strong, it must be rigid as to avoid deflection.
A driver with strong motor and light/rigid cone (usually expensive) will have good looking IR.
Now which part of the IR is mainly responsible for "honesty" and "detail"?
"Honesty": (1) Fast settling time and no ringing (2) Less overshoot
"Detail": (1) Fast rise time.
Imho.
Agreed. No RESA is good.
http://www.diyaudio.com/forums/full...ced-selective-amplification-resa-aka-ddr.html
Not sure but I think "dynamics response" terminology is rarely used in loudspeaker design, but it basically is saying the same thing. Imagine the difference between Step Response and Impulse Response. With "dynamics" we talk about change in voltage amplitude. But basically the same thing, imo.
Morning
😕
As in ringing / cone break up, due to the aluminum cone material maybe?
😕
As in ringing / cone break up, due to the aluminum cone material maybe?
Yes, but unfortunately, sometimes we have to get into subjective "preference/taste" territory.
Imagine this: signal moving from stage A to stage F. At stage A (assume a recording stage) the signal is distorted, or something is missing. At stage E the distortion is maximum. Now at stage F you have the option to be honest/pure, or you have an algorithm to reconstruct the final signal so it will be closer to it is before entering A. I cannot see a reason to force others with "my religion is the correct one, yours is wrong" here.
In loudspeaker design, there is this low level reverberation that is often exist in good recording. This information may be lost in the chain, or it might not exist from the recording. But there is a possible way to create or recreate this "vibrant" sound using loudspeaker panel vibration. So it is up to you to pick your religion here, using perfect box like Magico, or using resonative panel like Audionote, Harbeth etc.
Imo/ime, if you want to be pure/honest and still see a good result, you have to be pure/honest all the way from A to the end of the stage. And you know what it takes...
Last edited:
Possible... but we should not jump to conclusions and criticize the cone material without understanding other factors that might be contributing. 🙂
Nothing unfortunate here... the whole thread is about subjective preferences. And at the end, a home hifi system is about enjoying the music (again subjective).
Last edited:
Exactly, material is not so important as long it's well done, no?
In my opinion these tests shows that subjective opinions have a value. After all, most people preferred flat response.
I was surprised of my choice as I've had speakers with elevation around 3kHz which I didn't like at all. But perhaps it would be tiring in the long run. But this bump should be easy to get rid of with filtering or DSP. The smaller Perless have a bit more even response as I remember…...