Abyss Diana V2 Review (headphone)

[amirm]

The image below shows an SR-507 with a seal (red) and improper seal (yellow).

As you can see, quite a lot of similar characteristics

Whoever made that measurement doesn't own or use a Stax headphone as I have for 30 years. Stax headphones hang on each side of your ears like two speakers. They have almost no clamp pressure which yes, reduces perceived bass. But that is their charm, making them sound less boomy and as some people incorrectly call "fast."

You can surely clamp the hell out of them to the headphone fixture and get more bass but not compared to what you hear when you wear them normally. No way, no how they have flat bass as you are showing above. The peaky one with droop is far closer to what you perceive.

Now, if your goal is make things look good -- as in what a manufacturer would do -- sure, you would take a bungie cord and wrap around it. And put on such duct tape for good measure. That will fix up the bass for you. But will miss why we measure.

The Diana V2 has no resemblance to Stax. It has good clamp pressure and incredibly soft pads. It fit on my fixture better than most headphones I have tested. There were no visible gaps anywhere. Can you get more bass out of them in measurement? Yes, I mentioned that in the review I did. But subjective experimentation with EQ proved the measurements to be wrong. Changing the space and orientation of the headphone to increase measurement bass distorted the rest of the response.

Importantly, this headphone did not have proper bass in my listening tests. It simply didn't. To then go and get measurements that show otherwise is ample proof that the measurement is wrong -- just as the Stax ones you show above.

Bottom line, your evidence in the form of Stax measurements work against you, showing how you can produce wrong bass response.

As I keep saying, I iterate with EQ and listening tests to tie back to the measurements. Unless you do that, your measurements are not trustworthy.

 

[amirm]

BTW, some people use the Gras 43AG for measurements:

This them allows them to put clamp pressure on the headphone resulting in more bass. This is of course not happens in real life as you are completely removing the natural clamp pressure that the headphone itself provides.

 

[amirm]

And so the fact that the EQ 'worked' unfortunately does nothing to show that this set of measurements is representative of the product's actual performance. Something can be consistently bad or wrong in the same manner and have EQ work perfectly.

What on earth are you talking about? I made four sets of measurements, each of which were optimized on their own. The two that had their EQ result in better sound were the ones with more drooping bass. This had nothing to do with "consistency" but rather proof point of getting better sound out of a headphone.

You have some random theory of air leak. Where is your proof of air leak? How do you know bass doesn't increase due to other factors such as changing the volume of the air chamber, distance to microphone, etc.? There is no complete seal anyway on your head.

All I am reading from you is FUD with no substance. I measured this headphone. I listened it. I measured it again and again. I listened to it again and again. I played with everything that the frequency response said was right and wrong with the headphone. You don't have the benefit of any of this yet keep posting that the measurements must be wrong.

 

[amirm]

Abyss is supposedly releasing some measurements tomorrow which I imagine will tell quite a different story. Hopefully we get the chance to see Amir correct this and re-do the testing with a seal.

So between an independent test from me on a purchased headphone, and one from the manufacturer with who knows what sample and test conditions, you are already "hopeful" that their results will be more true? How will you know how much seal their setting has? You have a gauge for this? Or ready to buy into anything that shows more bass?

How will you prove to us their measurements are accurate? You have any ideas on that?

 

[amirm]

As has been stated several times, the measurement seal issue is caused by the pad shape not being flat, and instead being contoured to the shape of a head.

This is a fantasy hypothesis. As I noted, these are ultra soft pads and deform very nicely to the flat plate. We are not talking about some cheap, rock solid padding here. These fit like a glove to the flat plate which at any rate, is much more optimal than any "shape of a head." You have a lay intuition otherwise but it is wrong.

Does this look like a fitting problem to you?

No. the flat glass nicely compressed her lips for a tight fit.

There is high variability in Diana as you move, rotate and change the physical distance to the measurement mic. This is likely due to small acoustic volume as larger headphones are much easier in this regard. What this translates to must not be a chase for another measurement, but accept the variability and then see if we can still figure out how this headphone sounds tonally.

As I keep saying, any listening test of deep bass would demonstrate the bass roll off. This is very common among headphones and is shared with every headphone I have tested so far. And at any rate, it is up to taste to optimize. So to focus on it means not understanding the nature of headphone measurements and what I am doing here.

 

[richard12511]

I have zero affiliation with Abyss but I do have contact with their team and those measurements are coming, supposedly, with measurements on an IEC-60318-4 with a proper seal.

I'll trust 3rd party measurements over retaliatory measurements from a manufacturer who responded on youtube in such an immature way.

 

[GoldenOne]

The attached graph shoes measurements taken by DMS of the Diana V2.
As you can see, the difference when sealed (RED) vs unsealed (Yellow) is quite apparent. (Note: The red is not a 'forced' seal, it is just with the headphone resting naturally on the fixture)

DMS Said: "if you force the seal you can force it to be totally flat in the bass more than this, but ofc that doesnt heppen in real life".

So its a question of two extremes, yes, a forced seal can give excessive bass but is unrealistic. And you shouldn't measure it that way. Meanwhile a broken seal from a flatplate will go the other way, and will result in almost no bass, as well as significant resonance around 80hz. And so they also shouldn't be tested in that manner.

Below is the THD measurement for the diana V2

And then below is the THD measurement with the seal intentionally broken.

This shows a clear difference. I am not at all saying anything in regards to who's measurements are more trustworthy than someone else's. The way I see it the more measurements are available the better. No measurement is going to show an "absolute truth" and the more information is available the better.

But this is not a question of manufacturer integrity with their measurements, or performance of a measurement rig, this is about a clear issue with the specific method of testing these headphones, and a failure to mention this in the review.
It would be unfair not to point this out to those looking at these headphones or competitors to it. If I were spending this much, I would want to know how the headphone will perform, on my head, as intended.

If you believe these measurements are not trustworthy, then they can be easily disproven either by measuring again with a correct seal, or by measuring some other planar headphone with the seal broken and showing that it doesn't have the same effect.

 

[richard12511]

Spoken like a true snake oil vendor.
Totally unprofessional.

Yeah, a response like that is definitely not gonna win them any customers. That’s the kinda response I would expect from a young kid. To each their own, I suppose.

 

[amirm]

Meanwhile, look at another similar model, the Lambda Nova Signature, which DID have a seal during measurements. This time it extends almost flat down to 20hz, and THD is below 0.3% for practically the whole frequency range too.

Let me tell you want is wrong with your data there and that of Tyll's:

Notice what I have highlighted? "Noise?" His "distortion" measurements includes noise as it typically does. For electronics we follow the same scheme as it is "the standard." For headphone testing though, this is deadly. Ambient noise is high especially in bass frequencies and will easily bleed into the fixture and corrupt the measurements. You must exclude noise or you are reporting data that is of little to any use.

Here is another example of what is wrong there with noise. See that spike at 200 Hz? Here is another random example from Tyll:

See the same spike? Here is another:

There is that spike again. Clearly this interference/noise and has nothing to do with the headphone. It is at 120 Hz which is double the mains frequency. You can also see other consistent spikes at say 2 kHz.

Naturally then if you don't have a good fit on the headphone, more ambient noise can protrude and mess up the results more. But even without that, Tyll's distortion measurements are not very useful.

My distortion measurements on the other hand use a completely different technique for extraction of distortion and excludes large amount of noise. This is also aided by making measurements at elevated levels starting at 94 dBSPL and going up. As a result, we can clearly see differences between headphones even though there is fair bit of noise around me. Here is for example the HE400i:

And now Diana V2:

See how there is no mains noise and how the profile of the two are different? There is no measurement system imprint on them as there is on Tyll's.

So be careful with marrying yourself to someone's measurements without thoroughly understanding what is going on. I have literally spent months refining the distortion measurements that you see here. In this regard, Abyss folks are right. Let's hope they also understand what I just explained.

 

[amirm]

And then below is the THD measurement with the seal intentionally broken.

No one has "intentionally broke any seal." As I keep saying, I have placed the headphone very well on the fixture and optimized it.

 

[Degru]

No way, no how they have flat bass as you are showing above.

Here is the measured (and perceived) bass extension of my Stax 404LE taken with the headphone actually on my head, with no additional pressure applied. This is taken with a capsule from a measurement microphone attached to an earplug. I also had no trouble getting deep subbass out of a Lambda Signature I borrowed that had a looser headband and harder pads. And yes, I do listen to songs that actually contain content that low.

 

[amirm]

But this is not a question of manufacturer integrity with their measurements, or performance of a measurement rig, this is about a clear issue with the specific method of testing these headphones, and a failure to mention this in the review.

The heck you are talking about? This is in the review before any measurements are shown:

The softness of the pads presented serious challenges in measurements. Mere lifting them and letting good would change bass response by as much as 5 dB! I am not talking about moving the pads. Just pulling and letting go. Not only did the bass change, so did the response above 5 kHz.

And in the conclusions:

Equalization was hampered by difficulty in measuring the headphone and the fact that distortion can work against you, creating more harshness and artifacts as you attempt to fill in the holes.

There couldn't be more transparency than I provided. You simply have no case here.

 

[amirm]

Here is the measured (and perceived) bass extension of my Stax 404LE taken with the headphone actually on my head, with no additional pressure applied.

And this is my measurements of Stax SR-007 with B&K's HATS:

Here is multiple placements of the same:

Yours are excluding the the ear canal and which is a completely different animal and protocol. Or maybe you have a wider head than B&K HATS and I.

As I said in the review, I am not here to validate what you are doing, I am here to validate mine.

You guys really have the wrong war you are fighting. There is no way to rationalize measurements across people and measurement systems. Don't ask me to solve that puzzle. It can't be solved.

What can be solved is making useful progress in the face of such barriers. I have figured out how and that is what I am expressing in these reviews. If you think you have figured out how to produce more truthful measurements, then set up shop and prove that to people. I am not in that business with headphones (speakers and electronics, yes. Headphones no.)

 

[richard12511]

The attached graph shoes measurements taken by DMS of the Diana V2.
As you can see, the difference when sealed (RED) vs unsealed (Yellow) is quite apparent. (Note: The red is not a 'forced' seal, it is just with the headphone resting naturally on the fixture)

View attachment 105724

Below is the THD measurement for the diana V2
View attachment 105725

And then below is the THD measurement with the seal intentionally broken.

View attachment 105726

This shows a clear difference. I am not at all saying anything in regards to who's measurements are more trustworthy than someone else's. The way I see it the more measurements are available the better. No measurement is going to show an "absolute truth" and the more information is available the better.

But this is not a question of manufacturer integrity with their measurements, or performance of a measurement rig, this is about a clear issue with the specific method of testing these headphones, and a failure to mention this in the review.
It would be unfair not to point this out to those looking at these headphones or competitors to it. If I were spending this much, I would want to know how the headphone will perform, on my head, as intended.

If you believe these measurements are not trustworthy, then they can be easily disproven either by measuring again with a correct seal, or by measuring some other planar headphone with the seal broken and showing that it doesn't have the same effect.

I think you’ve done a good job of showing that Amir likely didn’t get a good seal. At the very least, you’ve convinced me. Where I think we disagree is with the measurements we want to see.

Correct me if I’m wrong, but you’d like to see best case scenario (perfect seal) measurements for all headphones. I do see the value in that, for sure, as the best case scenario for all levels the playing field.

My issue with that is that it hides the fact that some headphones are less likely to realize that scenario in actual use. I’d actually rather see measurements for the most common case. Hard to really get at that with just one human, but in this case it’s all we have, and in this case, the non sealed response is closer(verified by EQ) to what was actually heard. That means Amir was unable to get them to properly seal on his actual head. That’s valuable information, as I’d likely have the same problem, and the non-sealed response is likely to closer match what I’d actually hear. I want to know how the headphone is most likely to perform on my head.

Ideally, I’d like to see best, average, and worst case measurements, though I understand that takes more time. If I have to choose just one though, give me the one that correlates best with what was heard. At least that will give us a more valuable record to use in our database for the purpose of making our own target curve

Finally, I think it’s important to reiterate that Amir did measure a sealed response. He just didn’t show it, as it wasn’t the response he heard on his actual head. Headphone measurements aren’t at a point where we can trust them over our ears, unfortunately.

 

[Degru]

Yours are excluding the the ear canal and which is a completely different animal and protocol.

This will make the rest of the frequency response affected by ear canal and pinnae gain inaccurate, which is why I excluded it. But bass extension should be the same.

Or maybe you have a wider head than B&K HATS and I.

I was going to suggest that perhaps your head is narrow, but I felt that could have been perceived as rude. However, I know many people who own various different Stax and do not have any problems with seal.

 

[KTN46]

Here is multiple placements of the same:

I find this really interesting. I think seeing this information in reviews, or figuring out a way to show the "average of placements" would be super cool.

Obviously there are limitations as to how much we can interpret and use this information, but it would be cool to see your other placement attempts regardless.

 

[mshenay]

No one has "intentionally broke any seal." As I keep saying, I have placed the headphone very well on the fixture and optimized it.

Do you have a picture of that?

Like yourself I really enjoy the Ether CX but if I don't get a proper seal with it on my head... I notice it suffers an obvious loss of low frequency energy. An as much as you talk about CX I imagine you too must have taken a short moment to adjust it's fit to your head. Out of the box the "stock" fit doesn't work for everyone.

Still I think in this instance we all agree how important a proper seal is when listening, I mean I've had friends over before and most of them when hearing stuff in my collection will often adjust the fit when the sound is "off" as every one whose experienced listening to headphones has likely been thru that.

Whoever made that measurement doesn't own or use a Stax headphone as I have for 30 years. Stax headphones hang on each side of your ears like two speakers. They have almost no clamp pressure which yes, reduces perceived bass. But that is their charm, making them sound less boomy and as some people incorrectly call "fast."

You can surely clamp the hell out of them to the headphone fixture and get more bass but not compared to what you hear when you wear them normally. No way, no how they have flat bass as you are showing above. The peaky one with droop is far closer to what you perceive.

I'm a bit lost what Stax model is this? The L700MK II I have here has OBVIOUS clamping pressure and even my original purchased brand new 009 with less than 20 hours does as well... even the ES Lab ES-1A I tested had some clamp, not a lot but it certainly didn't just hang there

I'm concerned to hear your Stax Headphones just kinda of hang on each side of your head... they may need to be serviced or have you bent the headband too much and they no long clamp?

I really think every one here would appreciate seeing how your physically placing headphones for measurements moving forward,

 

[nerdoldnerdith]

 

[amirm]

Still I think in this instance we all agree how important a proper seal is when listening, I mean I've had friends over before and most of them when hearing stuff in my collection will often adjust the fit when the sound is "off" as every one whose experienced listening to headphones has likely been thru that.

I do not actually. As I make the tiniest adjustment on my measurement rig, there are profound differences in measurements. No such thing is perceived or we would go nuts as we moved our head around and tonality changed. I have the Aeon Flow Closed headphones as I type this and I just rotated their cups good 20 degrees forward and back with zero perceived change. Yes, the bass gets reduced if I pull them away from my ear but pushing on them gently and letting go makes no difference.

I think some serious research is needed on why our perception is not tracking measurements like this. It could be partially adaptation where we filter out variations.

 

[amirm]

Do you have a picture of that?

I do not. Didn't think I would need photographic proof as such!