The tournament format has many stages - how do we handle the timing and to make sure sufficient time is given for votes at each stage so that results aren't skewed due to not enough votes at any given stage? Then the time to track all of this?
So would I post clips for each round here, set up the votes for the rounds also here, but use the challonge website to track progress?
I think the idea of self managed to double elimination may make more sense.
So would I post clips for each round here, set up the votes for the rounds also here, but use the challonge website to track progress?
I think the idea of self managed to double elimination may make more sense.
Last edited:
I know only one song is a lot less work, but from the first round, I heard that some of the drivers that I liked with one song, didn't sound so great with an other.
And let's not get overboard with adding rules and formats to this simple and fun experiment. Not worth it. This is not a very precise experiment where all factors are taken care of. It's just a few recordings in a particular situation.
For me, I'd say: keep it the same and not think too much about "what if..."
And let's not get overboard with adding rules and formats to this simple and fun experiment. Not worth it. This is not a very precise experiment where all factors are taken care of. It's just a few recordings in a particular situation.
For me, I'd say: keep it the same and not think too much about "what if..."
Last edited:
I like keeping it the same but stepping up sound quality with higher res sound flies. I like that from a time invested standpoint. It's a lot of work to do this.
xrk971,
I think the simplest would be to just post a link to a double elimination layout configured for the number of drivers and their respective identifiers and let people work through it on their own if they wish. It's just a logical way to do one-on-one comparisons with a large selection of drivers.
I'm just trying to offer another option for people to work through all the drivers themselves, I don't want to de-rail anything. It's more of just offering an approach for anyone overwhelmed with how to compare a larger selection of drivers. However an individual wants to determine their favorite drivers in the bunch is up to them.
Any more than 5 or so drivers and that's how I'm personally going to approach this next round of drivers.
Again, thanks for the effort in providing the clips for comparison... it's a real eye-opener. However you decide to format this next round I'm really looking forward to it!
I think the simplest would be to just post a link to a double elimination layout configured for the number of drivers and their respective identifiers and let people work through it on their own if they wish. It's just a logical way to do one-on-one comparisons with a large selection of drivers.
I'm just trying to offer another option for people to work through all the drivers themselves, I don't want to de-rail anything. It's more of just offering an approach for anyone overwhelmed with how to compare a larger selection of drivers. However an individual wants to determine their favorite drivers in the bunch is up to them.
Any more than 5 or so drivers and that's how I'm personally going to approach this next round of drivers.
Again, thanks for the effort in providing the clips for comparison... it's a real eye-opener. However you decide to format this next round I'm really looking forward to it!
Last edited:
X, can you remind us again what the goal of the next round of subjective comparison is? I thought this was a one time deal. Not sure what you are going to achieve with this second round. Like others have mentioned, it is not like the test is very well controlled, so, it may not be wise to draw conclusions from it other than some broad conclusions like, neutral response is generally audible and preferred.
Yes, put up the sound files and have a bit of fun. I would be more interested in your measurements of the new drivers.
Yes, put up the sound files and have a bit of fun. I would be more interested in your measurements of the new drivers.
Ra7,
Thanks for asking what the objective is. If I want to do a round 2 in this same thread, to be fair to previously tested/auditioned drivers, the same setup and songs, and format for voting should be used. I will just increase the bit rate to 320kbps. Round 2 will not contain any drivers from round 1. Then we will have a finals round between top 2 of each round, and for that one, I will use new songs, still same format of a female vocal, rock, jazz, but add one classical. Oh, and to make it fun (and it's my prerogative as the OP and test conductor - to put one wild card (previoulsy eliminated) back into the finals. 🙂
Again, purpose is to pick what we think sounds best to our own ears and to have fun.
I will start posting measurements when I can rather than wait until the votes are in. It is a blind test and that shouldn't sway votes - to see how they measure before hearing them blindly.
Still waiting for Visaton B80, Tang Band W4-1320, Tymphany Kevlar "TC9" (this driver has to have the worst model number in all history for any driver, completely impossible to remember off the top of my head), and the AEH7.
Thus list of round 2 contestants: Fostex FF105WK, Vifa TG9FD10-8, Peerless 830986, Visaton B80, Tymphany Kevlar TC9, and AEH7. 6 total - not too bad. Did I miss any? Maybe an old Radio Shack 4in full range (I forget the model?).
Thanks for asking what the objective is. If I want to do a round 2 in this same thread, to be fair to previously tested/auditioned drivers, the same setup and songs, and format for voting should be used. I will just increase the bit rate to 320kbps. Round 2 will not contain any drivers from round 1. Then we will have a finals round between top 2 of each round, and for that one, I will use new songs, still same format of a female vocal, rock, jazz, but add one classical. Oh, and to make it fun (and it's my prerogative as the OP and test conductor - to put one wild card (previoulsy eliminated) back into the finals. 🙂
Again, purpose is to pick what we think sounds best to our own ears and to have fun.
I will start posting measurements when I can rather than wait until the votes are in. It is a blind test and that shouldn't sway votes - to see how they measure before hearing them blindly.
Still waiting for Visaton B80, Tang Band W4-1320, Tymphany Kevlar "TC9" (this driver has to have the worst model number in all history for any driver, completely impossible to remember off the top of my head), and the AEH7.
Thus list of round 2 contestants: Fostex FF105WK, Vifa TG9FD10-8, Peerless 830986, Visaton B80, Tymphany Kevlar TC9, and AEH7. 6 total - not too bad. Did I miss any? Maybe an old Radio Shack 4in full range (I forget the model?).
Last edited:
Hi X, I think it would be best to start a new thread for a new round of drivers.
Otherwise it will be hard for people to find the new downloads and you won't be able to do a new poll (unless you were planning not to have one).
You could use the same title with "Take II" on the end or something similar 🙂
Tony.
Otherwise it will be hard for people to find the new downloads and you won't be able to do a new poll (unless you were planning not to have one).
You could use the same title with "Take II" on the end or something similar 🙂
Tony.
I was going to suggest the same.
I realize this is only supposed to be a bit of fun, but man, that's a lot of work you're doing there. I hope I'm not overstepping myself, but one of the things I've learned over the years is that when bombarded with requests to do this, and that, and that, and so on, sometimes you just have to say No (or, as a typical, polite Canadian might say, "No thanks"). 🙂
I hear they go for $4 ea in India. That has to be the deal of the century. 🙂
Maybe Linuxguru can chime in where to get them.
Thanks for realizing how much work this is. I did not realize that it would take this long, but the out pouring of support for this project by other members has made it all worthwhile. The ultimate support is to send me drivers to test - and these are not just someone's unwanted el cheapo stuff - the 10F, B80, FF105WK, all the Daytons, the TG9FD, 3FE22, etc. are all drivers anyone would be happy to use in their own speaker. I used to be a one driver TC9FD kind of guy, but not so sure anymore. 🙂
I am running into some work priority issues at present so between that and waiting for final drivers to arrive, it may be a while for me to get around to posting the next thread for round 2.
Last edited:
India has a "fake/clone" Peerless, maybe they switched name to Tymphany too LOL but interested in a weblink.
I’m very curious of the results. I am finishing my latest multichannel audio system with three Vifa/Peerless TG9FD10-8 drivers, and I have to admit that possibly for the first time I’ve heard so well sounded full-range drivers with no correction whatsoever.
The completed speakers have famous IKEA Blanda Matt 3.5 litre sealed enclosures.
The on-site measurement containing all early reflections revealed that only one correction filter is necessary, giving this remarkable result:
The completed speakers have famous IKEA Blanda Matt 3.5 litre sealed enclosures.
An externally hosted image should be here but it was not working when we last tested it.
The on-site measurement containing all early reflections revealed that only one correction filter is necessary, giving this remarkable result:
An externally hosted image should be here but it was not working when we last tested it.
This is certainly consistent with my measurements of the TG9FD10-8 here:
http://www.diyaudio.com/forums/full-range/268626-vifa-tc9fd18-08-best-bang-buck-23.html#post4248013
It certainly is one of the smoothest and lowest distortion 3.5in drivers out there at any price. Perdormance approaching the much more expensive Scanspeak 10F/8424.
Rather than being fake or clone, it is just a separate company called Peerless. They aren't trying to fool anyone. My experience of them has been that they make their own products and none of them are trying to emulate any products that Peerless/Tymphany actually produce.
The company in question is probably Peerless India. They make cheap drivers, but the workmanship can be a bit down and quality control isn't that tight. They perform well, though.
Oops - I missed this discussion. I'll get a pair of the black Kevlar Tymphanys to xrk971 for testing ASAP.
The lot that I obtained was a surplus TV-manufacturing lot with both 6-ohm and 12-ohm drivers (the 12-ohm unit is mainly useful as a 2-driver parallel load, and maybe in line arrays; while the 6-ohm driver is more versatile for everything from Tabaqs, line arrays, mini-Karlsonators, etc.).
The main show-stopper in making them generally available to members here is that shipping costs for small packages out of India to anywhere else are among the highest in the world, thus largely negating the discounted price of the drivers (which I resell for between $8 to $10 each, depending on quantity).
One curious observation: the ScanSpeak 10F/8424 is very similar to this one, but with a fully circular flange (vs. squared-off flange), 8-ohm voice coil (vs. 6-ohm voice coil) and Fs of 90Hz (vs. Fs of 123 Hz). All other dimensions seem to be identical, including the cone effective area. Mounting locations on the flange are rotated by 45 degrees.
Given that they get $95 for the 10F/8424 vs. maybe $10 for the OEM Tymphany, it's perhaps no real surprise that the latter was discontinued a while ago.
Edit: These are from Tymphany (China). I inspected the cartons, and the workmanship on the drivers is very good, better than Peerless (India). There are good driver manufacturers in India, like DMPSL which makes drivers under the Boston and Kraalx brands, but these Tymphanys are in a different class...
Last edited:
- Status
- Not open for further replies.