hahah I don't know about 'famous' - maybe when Crin gets his hands on these he'll give us his take. But yeah getting a sense of what people like is helpful. Not everyone has the same sense of 'neutral', or the same preference for that. For me though, I do give most open-back headphones a bass boost, but the key thing is that I put it below 100hz with a rise into the sub-bass. What I don't like is when headphones have that boost that dominates over the lowest part of the ear gain like on the PX8, as that causes things to sound a bit thick and muddy.
Most of your favorites are open back, which I will tell you the Px8 is the furthest from on this planet lol. It’s almost intentionally closed in. Actually, it probably is. Which would also explain some of your misgivings about it.
Most of your favorites are open back, which I will tell you the Px8 is the furthest from on this planet lol. It’s almost intentionally closed in. Actually, it probably is. Which would also explain some of your misgivings about it.
There's a closed-back ranking as well there. But yeah like... with passive open-back headphones there are fewer acoustic constraints (with the exception of bass) - so yeah preferring open-back headphones' isn't exactly controversial. At the same time, it's not like I'd compare closed-backs or ANC headphones with passive open-back ones. That's why there are separate lists for closed and open, and there should be a separate one for ANC headphones. The goalposts are different and require a different type of evaluation.
So no, my misgivings about the PX8 are uniquely down to harmonic imbalances in the frequency response, confirmed by the graph as well. The same scrutiny would be applied to open-back headphones that have these issues.
There's a closed-back ranking as well there. But yeah like... with passive open-back headphones there are fewer acoustic constraints (with the exception of bass) - so yeah preferring open-back headphones' isn't exactly controversial. At the same time, it's not like I'd compare closed-backs or ANC headphones with passive open-back ones. That's why there are separate lists for closed and open, and there should be a separate one for ANC headphones. The goalposts are different and require a different type of evaluation.
So no, my misgivings about the PX8 are uniquely down to harmonic imbalances in the frequency response, confirmed by the graph as well. The same scrutiny would be applied to open-back headphones that have these issues.
This is how Mark Levinson describes sound design of the 5909:
“Designer Sound
The № 5909 has been engineered to provide you with optimal sound for an over-ear headphone. Its sound profile has been expertly tuned to an acoustic response curve that achieved the highest subjective scoring across listening tests. Experience reference-quality sound, the culmination of decades of research into listening preferences and the science of acoustics.”
So they are obviously referring to the Harman curve, which is no surprise as ML is owned by the Harman Group.
Now your graphs show that this reference curve claim is not obtained on the treble areas.
Is this not a massive failure and false claim by Mark Levinson?
This is how Mark Levinson describes sound design of the 5909:
“Designer Sound
The № 5909 has been engineered to provide you with optimal sound for an over-ear headphone. Its sound profile has been expertly tuned to an acoustic response curve that achieved the highest subjective scoring across listening tests. Experience reference-quality sound, the culmination of decades of research into listening preferences and the science of acoustics.”
So they are obviously referring to the Harman curve, which is no surprise as ML is owned by the Harman Group.
Now your graphs show that this reference curve claim is not obtained on the treble areas.
Is this not a massive failure and false claim by Mark Levinson?
Yes. The labs and engineers and quality controls of Bowers & Wilkins AND Mark Levinson screwed up royally here. Clearly that must be the only logical explanation.
Not exactly. There are two things going on with this I think.
They reference their scoring system which is based on their statistical model that predicts listener preference. This model unfortunately isn't able to accommodate the more fine-grained elements of FR deviation that may lead to preference issues among those who notice them, and I've spoken to Dr. Olive about this on a live stream we did with him so he's aware of that. But on that point, they're not lying when they say it scores well on their statistical model.
The second thing is that it may have been engineered to match more perfectly than what the two models I tested show, but in practice with production units there could be some unit variation that leads to the peak issue. With all of these headphones, when you're dealing with so many variables (and imperfect materials like with pads), you're bound to get some unit variation, and this could be the result. Now I'm not saying that's exactly what's going on here as I've only tested two of them, but it's a possibility.
So to answer your question, no it's not at all a false claim by Mark Levinson nor a massive failure by any stretch. I think they achieved a very good passive result, and certainly close to their intended tuning. But at the same time I think the platform itself needs a bit of... polish, and I think that's how we should think about the 10khz peak as well.
Well I actually have a bad set of px8s. I first noticed it with the battery, very unpredictable charging and lifespan, and the percentage reporting is simply broken. But the worse problem is that they seem to have an amplification problem, sometimes dropping off about 20%, sort of like a low level switch on a hifi amplifier. Last night I had the opposite situation, where I was listening and suddenly they came alive with a much stronger sound. After listening that way much of the night I put them on this morning and actually heard them drop back after about 10 seconds. Sort of like Cinderella's night at the ball, the good times were over.
I don't mean to open up a can of worms, but I'm wondering if this is a widespread thing. I know it won't be of everyone's tastes but the below link happens to be what I was listening to before and after at high volume. For me I would say enjoyably loud was about 60% when my amplification was "on", 75-80% when "off". If someone is willing to do an experiment and report back their high volume percentage.
It's hard to say if quality suffers with less headroom but I think it does. I'll be contacting B&W tomorrow about it.
Yuuuuuuup. Sounds exactly like mine. A song would start lower and then "pop" almost like some amplification kicked in or something sometimes. That on top of the battery issue is no bueno.
It's kind of funny because I wasn't sure if I was going crazy or just having "bad days" and couldn't enjoy the sound as much. I was messing with the seal/fit and wasn't sure if it was my glasses or what. Turns out the headphones may have been playing tricks on me!
I am local to audio advice, so I am going to try and go into the store when they open next week. I'm right at the tail end of my return period, so hopefully they've got some stock and can swap it.
Not exactly. There are two things going on with this I think.
They reference their scoring system which is based on their statistical model that predicts listener preference. This model unfortunately isn't able to accommodate the more fine-grained elements of FR deviation that may lead to preference issues among those who notice them, and I've spoken to Dr. Olive about this on a live stream we did with him so he's aware of that. But on that point, they're not lying when they say it scores well on their statistical model.
The second thing is that it may have been engineered to match more perfectly than what the two models I tested show, but in practice with production units there could be some unit variation that leads to the peak issue. With all of these headphones, when you're dealing with so many variables (and imperfect materials like with pads), you're bound to get some unit variation, and this could be the result. Now I'm not saying that's exactly what's going on here as I've only tested two of them, but it's a possibility.
So to answer your question, no it's not at all a false claim by Mark Levinson nor a massive failure by any stretch. I think they achieved a very good passive result, and certainly close to their intended tuning. But at the same time I think the platform itself needs a bit of... polish, and I think that's how we should think about the 10khz peak as well.
Since you understand this better than I ever will as I am a lawyer, not an engineer or really able to understand a ton of the finer technical points… What role do you think a NC has in both altering frequency response and just altering sound in general I feel like with ANC on measuring is more difficult and it can make sound perception different for different people
Since you understand this better than I ever will as I am a lawyer, not an engineer or really able to understand a ton of the finer technical points… What role do you think a NC has in both altering frequency response and just altering sound in general I feel like with ANC on measuring is more difficult and it can make sound perception different for different people
It depends on the implementation, but generally speaking, with ANC headphones there's some DSP in there to fix potential problem areas that arise as a result of compromises made on the acoustic system to integrate the noise canceling function (whether we're talking the electronics built into it, the microphone requirements or other mechanisms). In some cases it's not just that, but also the whatever other features there are in the headphone, like head-detection and so on. And yeah there are some that are 'smart' headphones that will behave differently depending on the coupling as well, to accommodate when the seal is broken for example. This is why on ANC headphones you have to do manual sweeps by ear to see if what you've found in the measurement correlates reasonably - obviously there will be differences since your individual HRTF isn't the same as the measurement rig's, but for coarse-grained information it tends to be very noticeable, and a useful way to confirm the results.
That is... kind of how I feel about them to be honest. 'Unlistenable' is a bit of a tricky term, because like... yeah I've heard worse for sure. But at $799 I would expect better sound quality out of a headphone than what the PX8 delivers. The thing is, we create a lot of expectation around how well something should perform, and a lot of that is anchored to things like price, aesthetics, build quality and so on (there is actually research on this, as far as what influences the headphones people choose). But I don't think we should assume that just because it costs a lot and looks/feels nice that it's necessarily going to sound good - and the PX8 is a great example of that in my view.
I don't think people enjoying the PX8 generally are doing so because of biases, or solely because of biases. Perhaps some are, but that alone just wouldn't explain the overwhelmingly positive response the headphone is getting.
I literally popped the Bathys and PX8 in front of my girlfriend, and just let her have a listen. She preferred the PX8 and said that vocals seemed more 'real' sounding. She didn't know the price, that's for sure (that's the last thing I want her to know).
For me at least all my biases were opposing this headphone. I'm already heavily into Focals, and had the expectation I'd prefer Bathys. But something about the PX8 enamoured me, vocals sound fuller, it has a more energetic, fun presentation. That's just to me, but my girlfriend did agree too - and she's got a damn good ear & has been a musician/voice actress for almost a decade. I say this only to qualify her, she's not just someone running around with airpods looking for a big bass response.
I'm not saying this to suggest PX8 is indeed exceptional, there are clearly other high performing BT headphones, only that some people do prefer it. And I think this preference is down to more complex elements than simple cognitive biases.
I don't think people enjoying the PX8 generally are doing so because of biases, or solely because of biases. Perhaps some are, but that alone just wouldn't explain the overwhelmingly positive response the headphone is getting.
I literally popped the Bathys and PX8 in front of my girlfriend, and just let her have a listen. She preferred the PX8 and said that vocals seemed more 'real' sounding. She didn't know the price, that's for sure (that's the last thing I want her to know).
For me at least all my biases were opposing this headphone. I'm already heavily into Focals, and had the expectation I'd prefer Bathys. But something about the PX8 enamoured me, vocals sound fuller, it has a more energetic, fun presentation. That's just to me, but my girlfriend did agree too - and she's got a damn good ear & has been a musician/voice actress for almost a decade. I say this only to qualify her, she's not just someone running around with airpods looking for a big bass response.
I'm not saying this to suggest PX8 is indeed exceptional, there are clearly other high performing BT headphones, only that some people do prefer it. And I think this preference is down to more complex elements than simple cognitive biases.
Yeah I don't mean to suggest that overwhelmingly those who report they enjoy it are doing so because of biases towards expensive equipment (even if in some cases that's sure to be going on), merely that oftentimes our expectation is that it would be better when that's not necessarily the case. For some folks 'sound quality' is uniquely associated with how much bass a headphone has, and more is almost always better to them - I don't mean to lump you into this category, just as an example. In fact I know a number of folks who think of it like that. This is the subjective nature of the hobby... And just because I think it's not good sounding doesn't mean people who like it should like it any less. You like what you like.
Funnily enough, @Denkmal felt perfectly entitled to make a snide comment in response to my post a few pages back when I called out this thread for lapses into poor conduct.
In a similar fashion that the Harman curve can apply to the interpretation of desirable audio frequencies so might the Bell curve apply to the interpretations put forth by audio “reviewers”.
@Resolve , one page back @Denkmal asked you if the specific PX8 you tested is the only one you've had a a chance to listen to and measure. I'm curious about that as well.
I won't question the harmonic and compression issues that you noticed, even though I haven't noticed anything that sounds nearly as bad as what you describe, but the bass response on your graph is pretty outrageous. The graph shows the entire bass range boosted from anywhere between 6 to 9 db over the Harmon target. I do hear the bass as being boosted, but I certainly don't hear a nearly 10 db boost.
That, combined with the fairly extreme difference between your impressions of the headphone, and that of most others who've listened to it, do lead me to wonder if there could be something wrong with the specific unit you tested.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
