Patrick, I just don't buy this argument. In a blind listening test, the bump in the response will not be preferred. By everyone. It sounds dull compared to most speakers with a tipped up treble. It doesn't sound dull compared to a speaker with a flat response.
I have waveguide speakers too and I have the KEF Q900s and the Q100 drive units. No, they don't sound right with a tipped up treble. By your own admission the Summas don't have this response anomaly. Nor do KEFs own Reference 201/2 speakers. If you ask me, this is a flaw introduced to keep the 'upgrade' path open and maybe to save money on a few extra components in the crossover.
Rather than touting this speaker as the next best thing since sliced bread, I would recommend buying the Q100 and upgrading the crossover. The drive units are promising, the end product is not.
This is not a valid reason. Sure the overall tone is determined by the power response, but the direct sound has much higher weighting in determining the balance. Get that wrong and no matter what you do elsewhere, it is not going to sound right.
Looking at the off-axis measurements, I'd say the KEF crossover is near 1.5 kHz and the Chorus one is near 2.5 kHz. Both show the same tweeter bloom. In fact, the KEF does not seem to take any advantage of the waveguide to match directivity. Any way you look at it, that bump in the power response of the KEF is going to be audible.
The Chorus doesn't do much better, but at least they got the on-axis right.
Sorry, don't buy the argument of the power response. The bump in the power response is going to sound bright. This why everyone uses waveguides in the first place, to avoid the tweeter bloom. There is no need to tip up the response. On-axis must still be flat and off-axis will be replicas on the on-axis response.
I admit I'm no expert. But after you've played with these things for a while, you get a flavor of what to expect after seeing the measurements. You are not going to get anywhere with that on-axis response. Sorry.
That would be a questionable tactic given these speakers will typically compete on the showroom floor with flatter PSBs, etc., and not higher end KEF models. I wonder if it's a production issue? For the record, after reading all the accolades I found the LS50 a real disappointment at a local audio show.
Remlab, I can tell by the measurements and sensitivity these are power sucking fire wood. that was the pont of posting the link wasn't it? Why the attitude towards me and horns? Is it because I have the ability to point out the obvious and you fail to see it? ROFLMAO
It appears a slightly higher crossover point and a bit more padding would be needed to correct this speaker. But if they're anything like the Q100 and other newer KEFs, they should image very well.
FYI, Stereophile's measurements appear to be different: KEF LS50 Anniversary Model loudspeaker Measurements | Stereophile.com
Sound and Visions: http://www.soundandvisionmag.com/fi...mage_620/_images/201212/kef ls50 off-axis.jpg
FYI, Stereophile's measurements appear to be different: KEF LS50 Anniversary Model loudspeaker Measurements | Stereophile.com
Sound and Visions: http://www.soundandvisionmag.com/fi...mage_620/_images/201212/kef ls50 off-axis.jpg
I'd say all three sets of measurements show that the "design axis" is not directly on-axis, but about 30deg off.
The OP misses the point by assuming they should be aimed straight at the listener. They should be "overtoed" or straight out, depending on preference.
I have a question about the discussion in this thread, I hope that you guys don't mind helping me understand a bit better.
Are the anechoic off axis measurements taken by toeing the speakers in "x" degrees?
Most people toe their monitors in a bit, 15-20 degrees seems to be fairly standard. If you listen to monitors toed in, say 20 degrees, does that mean that the 20 degree off-axis graph is the one that is most relevant to the way you use the speakers?
Just trying to understand why the on-axis is the graph that is being focused on if most people listen with the speakers toed in. If I am totally misunderstanding the relationship between the anechoic measurements and degrees of axis and toe in, I apologize.
Thank you for your time,
Aaron.
Are the anechoic off axis measurements taken by toeing the speakers in "x" degrees?
Most people toe their monitors in a bit, 15-20 degrees seems to be fairly standard. If you listen to monitors toed in, say 20 degrees, does that mean that the 20 degree off-axis graph is the one that is most relevant to the way you use the speakers?
Just trying to understand why the on-axis is the graph that is being focused on if most people listen with the speakers toed in. If I am totally misunderstanding the relationship between the anechoic measurements and degrees of axis and toe in, I apologize.
Thank you for your time,
Aaron.
Aaron, your question is a very important and relevant one. Look at the first chart on SoundStage. That chart shows the 30 degrees off-axis measurements. I don't see anyone listening further away than 30 degrees. Compare the LS50s measurements with some others, for example:
SoundStage! Measurements - PSB Synchrony One Loudspeakers (7/2008)
Look at 30 degrees off-axis (chart 1).
Most important is the listening window response (chart 3). That is the balance you are likely to encounter most of the time. And it clearly shows the glaring error in the response of the LS50. Again, a few additional crossover components should fix it.
All this talk of listening off-axis is rubbish. The only way to get the balance right is by having a flat on-axis response and then making sure the off-axis response is smooth and replicates the on-axis response with a little narrowing in the HF. Bumping up the on-axis response can never sound right.
SoundStage! Measurements - PSB Synchrony One Loudspeakers (7/2008)
Look at 30 degrees off-axis (chart 1).
Most important is the listening window response (chart 3). That is the balance you are likely to encounter most of the time. And it clearly shows the glaring error in the response of the LS50. Again, a few additional crossover components should fix it.
All this talk of listening off-axis is rubbish. The only way to get the balance right is by having a flat on-axis response and then making sure the off-axis response is smooth and replicates the on-axis response with a little narrowing in the HF. Bumping up the on-axis response can never sound right.
I wouldn't be in a great rush to pronounce the designers at KEF as a pack of fools or charlatans, having met and talked to some of those guys over the years. I have friends who are or have been professionals in the industry. I get no sense of the cynicism people seem all to eager to ascribe to them. They know what they are doing. Unlike DIYers like ourselves though, they are building for commercial sales, so their choices may not be the same as ours- and of course our choices are much more varied than most of those who are consumers of audio products rather than builders. I have heard the KEFs, as I mentioned on another thread, either on here or Audio Circle, at the Hifi Wigwam show earlier this year. Used with quite a low-powered modern valve amplifier, I thought they sounded excellent. The owner (the show is based on people bringing their own systems to demonstrate) was very happy with them. I have no intention of buying them to replace what I have and no doubt they will not be to everyone's taste- what is?- but for the price and size they seem outstanding. Only my take of course, but even though this was at a show, it was in the context of a well sorted privately-owned system.
Academically flat small speakers tend not to get reviewed positively. KEF engineers have to make salable products
Toaster, this is a good point and we know they can make a good loudspeaker (201/2). The point is this: why the aberration when an almost identical speaker has it done right?
I'd only be guessing ra7, but I suppose further equalisation might not have seemed sufficiantly worthwhile to the designer/ design team if the speaker sounded good as-is. Put another way, I expect the point at which deviation from a flat fr would have been considered a problem in need of a solution in the context of this design was not reached! The raw fr of the drive units themselves will be the result of a number of design decisions and compromises, the physical properties of the materials used to execute the design, along with the exact shape of the resulting diaphrams, the tweeter waveguide, etc. The design team may also have a preference for minimum componant-count crossover networks- and as already noted, it is a commercial product with a target budget. The 201/2's raw driver response may be intrinsically flatter, eq might have been easier to implement if required, and/ or the (different?) design team might have considered that a flatter fr was a more important design goal for that speaker system. I expect if, say, Jack Oclee-Brown or Mark Dodd were asked politely, they might tell us! Don't get me wrong; discussion about the minutae of speaker design is a perfectly reasonable and worthwhile thing on the forum- of course it is- but we're not only being unnecessarily rude but also making ourselves look silly if we think that someone like Oclee-Brown (MEng, PhD Loudspeaker Acoustics) and his colleagues at KEF can't read a frequency response chart! Reading through the Stereophile review, the measurements are generally very good, and that is still a good fr in the context of the commercial speaker market, even if it is not perfect. For what little it's worth, my own fairly brief listen to the KEFs pretty much agrees with John Atkinson's. In different rooms with different partnering equipment they might sound a lot worse, but in my limited experience with them they can sound great- subject to the listeners tastes and expectation of course!
Last edited:
No, I am not affiliated with any hi-fi company. Just a diyer who doesn't want to settle for compromised performance that can be fixed readily. You didn't answer whether you were affiliated with KEF. Are you?
Designing for a budget I can understand. What I don't understand is the press going gaga over something that has such an obvious flaw. Yes, it's got coincident drivers, which result in good imaging. But the drivers are small, they will not play very loud, and with that frequency response, they are going to sound bright. Don't praise it like it's the next best thing since sliced bread.
Also, a 'flatter' response should be the design goal for all speakers, not just the Reference models.
I want to repeat that the drivers themselves are pretty good, and if you can get your hands on a pair of Q100 and improve the crossover, you will have something better than the LS50.
- Status
- Not open for further replies.