Hehe I was relieved listening to that statement. Because that was how I did my listenings. Anybody (like Planet10) will complain that I will not hear the imaging quality. Well, may be, but I can still hear imaging quality from a single multi-way speaker.
My personal opinion on test tracks is use:
1. Something that is widely popular. Using something popular is great because chances are, most people have heard it somewhere else and have something to compare it to. I recommend something from Daft Punk- Random Access Memories. While not complex, i love the beginning of "lose yourself to dance". "One of the biggest albums of 2013 was Daft Punk's Random Access Memories, with many reviews commenting on the album's great sound.[22][23] Mixing engineer Mick Guzauski deliberately chose to use less compression on the project, commenting "We never tried to make it loud and I think it sounds better for it."[24] The album's opening track, Give Life Back To Music, may as well be understood as hinting at this issue in the great part of modern music productions. In January 2014 the album won five Grammy Awards, including Best Engineered Album (Non-Classical)" - Loudness War, Wiki
2. Something highly complex
3. One of your favorite songs
4. Pink noise
1. Something that is widely popular. Using something popular is great because chances are, most people have heard it somewhere else and have something to compare it to. I recommend something from Daft Punk- Random Access Memories. While not complex, i love the beginning of "lose yourself to dance". "One of the biggest albums of 2013 was Daft Punk's Random Access Memories, with many reviews commenting on the album's great sound.[22][23] Mixing engineer Mick Guzauski deliberately chose to use less compression on the project, commenting "We never tried to make it loud and I think it sounds better for it."[24] The album's opening track, Give Life Back To Music, may as well be understood as hinting at this issue in the great part of modern music productions. In January 2014 the album won five Grammy Awards, including Best Engineered Album (Non-Classical)" - Loudness War, Wiki
2. Something highly complex
3. One of your favorite songs
4. Pink noise
Last edited:
Ah! This is exactly why I don't like B. I couldn't describe it tho, I just wrote that it has unrealistic sweetening effect (2nd HD). And I still don't understand how that "pseudo stereo" is possible with B and not with others, or what is the mechanism involved.
But I think it is the driver ability to expose details that makes it capable of producing room acoustics cues, hence sound more "stereo". And of course this should be a good thing. In the clip#6 (Mariah Carey or whoever) driver B clearly show the room cues, tho in Eric Clapton live performance, driver D was more realistic to me (10F is even better).
Problem with female vocal is that you can have many sounds and you don't know which one is more accurate, and even an inaccuracy (such as 2nd HD) will make it even sound better. Different with male vocal where inaccuracy will make it worse and "intolerable".
There is a rare "quality" that is hard to achieve with speaker. The speaker should trigger those who likes to sing to sing along. I'm not yet sure what design error prevents this from happening. It could be the pitch is unsuitable, it could be the voice/music is "enclosed" in a box and thus unnatural and "miniature". But this is most probably a function of crossover design in listeners' home and in XRK's (so all drivers will behave the same I believe).
If possible, I would suggest to go back to the LR2 crossover. Not because it is better, but because they are very different, and I'm sure the sound of the LR2 is what all of us have been familiar with. It is even better if the crossover/speaker matches those used by the recording engineers!!
Note that the LR2 has a midrange "lift". Many speakers will sound fatiguing and this, fair or not, will make it easier to choose the best one.
LOL, I will download it then and remove most drivers solely by the clip
But care must be taken because 10F could be accidentally removed Thanks for clarification. I'm happy then, knowing that my speaker really have that resolving capability (showing how the singer struggling with inhale/exhale).
With my speaker, MP3s don't sound as good as CD. And surprisingly (but expected), the recorded MP3s/WAVs from many listening tests in this site, didn't sound better than those MP3s...
What bothers me is that... while with my Sennheiser, many files will quickly be disqualified for beeing fatiguing, my speakers could handle all these clips well...
Plain old mono . . . with some excellent demo tracks . . .
Gustav Holst, Ralph Vaughan Williams, Peter Mennin, Vincent Persichetti, Herbert Owen Reed, Frederick Fennell, Eastman Wind Ensemble - Holst: Suites 1 & 2 / Vaughan Willams: Folk Song Suite and Toccata Marziale / Reed: La Fiesta Mexicana / others - A
Gustav Holst, Ralph Vaughan Williams, Peter Mennin, Vincent Persichetti, Herbert Owen Reed, Frederick Fennell, Eastman Wind Ensemble - Holst: Suites 1 & 2 / Vaughan Willams: Folk Song Suite and Toccata Marziale / Reed: La Fiesta Mexicana / others - A
Are we sure that it would be better to listen with mono recordings of the mono clips?
Isn't the point to simulate the Floyd Toole set up? Think about it, Wouldn't a stereo recording of a mono source be more like what we hear in real life and be the better way to simulate the Toole set up? Then the best substitute to physically be at X:s place listening to the mono speaker should then be a stereo recording, at least if you like most of us listening with headphones.
On the other hand, If you listening through your speakers, then it would make more sense to do a mono recording.
Isn't the point to simulate the Floyd Toole set up? Think about it, Wouldn't a stereo recording of a mono source be more like what we hear in real life and be the better way to simulate the Toole set up? Then the best substitute to physically be at X:s place listening to the mono speaker should then be a stereo recording, at least if you like most of us listening with headphones.
On the other hand, If you listening through your speakers, then it would make more sense to do a mono recording.
I think we also need to remember the purpose of these tests. It is not to pick the best drivers (objective or subjective). There are too many variables here (personal preference, individual hearing responses, our system used for listening to the recordings, microphone pickup response curve, xkr's room acoustics, crossover effects...). The only thing these tests do is provide us with a little more data in a controlled side-by-side comparison to help us decide if we want to buy some of these drivers to use at home. Maybe we are lucky and have one of these drivers at home and then we can extrapolate from our experience on that driver to how it sounds here (how it sounds different to how we know it) and then guess how another driver might sound in person.
We should not get too caught up in these tests. Just because something "wins" here does not mean that it is the "best" driver. And just because we like something in these tests, does not guarantee we will like it when we get to hear it in person.
I really don't care which one "wins", but of course it is fun to root for a favorite like a sports game. I just want to make some good personal decisions when buying drivers so that I don't end up with a box of drivers I don't like.
These tests have been extremely helpful for me, and right now I have all the data I need to make some decisions. Personally I don't need round4 (a final shootout). After a break, I wouldn't mind another fresh round of drivers.
(edit: I just realized that all rounds are run in FAST, but that the woofer changed in round3. I originally thought that rounds 1&2 were run full range without a woofer. Personally I would prefer to hear the drivers fullrange without the FAST even though it might give a "voting bias" to those drivers with more bass response. I think fullrange, both listening and measurement with natural rolloff, will give us more honest information about how the drivers sound)
We should not get too caught up in these tests. Just because something "wins" here does not mean that it is the "best" driver. And just because we like something in these tests, does not guarantee we will like it when we get to hear it in person.
I really don't care which one "wins", but of course it is fun to root for a favorite like a sports game. I just want to make some good personal decisions when buying drivers so that I don't end up with a box of drivers I don't like.
These tests have been extremely helpful for me, and right now I have all the data I need to make some decisions. Personally I don't need round4 (a final shootout). After a break, I wouldn't mind another fresh round of drivers.
(edit: I just realized that all rounds are run in FAST, but that the woofer changed in round3. I originally thought that rounds 1&2 were run full range without a woofer. Personally I would prefer to hear the drivers fullrange without the FAST even though it might give a "voting bias" to those drivers with more bass response. I think fullrange, both listening and measurement with natural rolloff, will give us more honest information about how the drivers sound)
Last edited:
I agree. I think the most valuable part of this process is the opportunity to hear and compare various drivers side-by-side (and to see if any trends emerge in the expression of individual preferences). I too would rather hear a round of fresh drivers than a Round 4 shootout of drivers that we may already have listened to multiple times.
Putting more science into speaker building.
Floyd Toole says some most interesting things about speaker testing. Everyone should be aware of how biases work. Anyone building and listening to speakers should pay attention to what he has to say.
https://www.youtube.com/watch?v=zrpUDuUtxPM
Floyd makes a pretty good case the the JBLpro M2. Any chance of learning something that might get some of the more inexpensive drivers to perform like the M2?
The suggestions that each round should contain completely different speakers, and not comparing the best with the best, seems a waste of time. Our memory is too fickle to make a comparison over a period longer than a few seconds.
Floyd Toole says some most interesting things about speaker testing. Everyone should be aware of how biases work. Anyone building and listening to speakers should pay attention to what he has to say.
https://www.youtube.com/watch?v=zrpUDuUtxPM
Floyd makes a pretty good case the the JBLpro M2. Any chance of learning something that might get some of the more inexpensive drivers to perform like the M2?
The suggestions that each round should contain completely different speakers, and not comparing the best with the best, seems a waste of time. Our memory is too fickle to make a comparison over a period longer than a few seconds.
But we can always go back to a previous round and listen to those older recordings. We can even A/B compare something between the rounds.
Ideally there is a standard test procedure for all rounds (maybe from this point forward?) so then in theory xrk can compile a "database" of recordings for each driver. Then we don't need to retest the "best-of" with each new round and we can freely blind test any drivers in the database of recordings.
I do not agree that we can compare between rounds... X has managed to improve his setup with each round so far. So comparing to older rounds wouldn't be fair I.M.H.O.
Floyd Toole was just trying to sell the M2
Seriously, there are too many trade-offs in speaker design, such that, just because directivity is important it doesn't mean that every tweeter should use waveguide, or, just because dynamic and power response is important it doesn't mean that everyone must use compression driver.For learning purposes, I think GedLee has some materials regarding to the design concept of M2.
I'm sorry, but I honestly don't understand your point. The fickleness of memory applies to all tests, doesn't it? Including testing the same drivers many times?
By the way, I have nothing against including more popular drivers in more than one listening test. I just think it might be more interesting to get exposure to (and a listening impression of) a wider range of drivers rather than treating it as a multi-tiered round robin tournament.
It all depends on what xrk971 sees as the goal or primary purpose of his listening tests. If it's to determine what drivers sound best in his setup then the multi-tiered round robin is the way to go, but if not...
Perhaps xrk971 can help clarify things. After all, he's the one doing all the hard work setting up these tests.
It all depends on what xrk971 sees as the goal or primary purpose of his listening tests. If it's to determine what drivers sound best in his setup then the multi-tiered round robin is the way to go, but if not...
Perhaps xrk971 can help clarify things. After all, he's the one doing all the hard work setting up these tests.
OK that makes sense. We probably do need round4 to be a retest (of the top picks/best) with the new setup.
But at some point, if X stabilizes his test setup, then we should be able to compare drivers from say round6 and round10.
sorry X I think you have created a monster here.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.