HIFIMAN Susvara Headphone Review

[solderdude]

But about Susvara.
I quite liked the sound as is (no EQ) but did not listen at loud levels as that requires quite a bit of voltage.
When that is not available during the demo and compares it directly to some other reference it may not sound great.
I have heard people say the same thing about the HE.

Yes, it doesn't measure $6k well but does the HE1 measure €70k well ?

Susvara:

and HE1:

Notice the lightly increased almost Harman bass levels for HE1 which Susvara lacks and similar 'clarity dip' around 2kHz. Aside from the bass levels how far is the tonality off ?

Below the Harman following DCA Stealth:

When looking at these measurements and compare them with Susvara... how poor can it sound (at normal listening levels)
Yes, Susvara does not have the same amount of low bass and does not have the same ear-gain around 2-3kHz as a standard fixture but we all know that not everyone does have the same ear-gain and some prefer less bass.

 

[Merkurio]

In reality, there is no actual reason to expect there will any audible distortion or static at normal or even loud listening levels based on the measurement you conducted. If anything, your conclusion is the one that does not align with data in my opinion.

Actually, there is:

 

[Sengin]

In reference to this chart:

Does the sometimes large (but importantly, inconsistent) differences between the left and right channels' level mean there is (or could be) phase differences between the left and right channels? I have a feeling the answer is 'no' or else it would have been brought up already.

 

[IAtaman]

Actually, there is:

No, just because the graph crosses the line does not mean it creates audible distortion.

Here are 3 examples for you, from headphones which has higher distortion in the same general area, that have have no static or noise issues and are recommended, some of them without EQ.

 

[OnLyTNT]

This is another reason why I don't think we should call these programs AI - because we misjudge what is it that they do. ChatGPT is a large language model, it does not make judgements, It just repeats the judgement of others that it learned. If enough websites says NwAvGuy is Charles de Gaulle, he would say he is CdG.

That is true, LLM actually has no idea about what it is talking about. Simply and roughly, it combines words according to statistical results...You say A, it says B, and no idea why the reply is B. When you ask why the Reply is B, it gives you C and again no idea why the reply is C

 

[amirm]

That is true, LLM actually has no idea about what it is talking about.

Yet what it wrote nearly matches everything a subjective reviewer would say about this headphone! This is why words alone have little value in evaluating the sound of the headphone.

 

[amirm]

That is a false equivalency. You listened to these headphones, you heard some static at undisclosed listening levels and concluded it is not a good headphone.

Are you serious? How many times have I explained what I do? Once more, I measure. That is the cornerstone of my review. Those measurements show major deficiencies in both tonality and distortion. It also massively clips at 115 dBSPL where hardly any headphone does. I listen and verify that those deficiencies are indeed there. It is this totality of information that tells me it is not a good headphone and certainly misses the mark at what it sells for.

Compare that to other reviewers who spend pages talking about a headphone with measurements either absent or afterthought. You can read them 10 times and still not know exactly what they are saying about the performance of the headphone. And even if you did, you read the next review and that will set you right back to confusion.

My reviews have structure, focus, goals and objectives. This is why folks are up in arms about them. The message comes out very clearly and crisply if a headphone is not performant.

 

[IAtaman]

But about Susvara.
I quite liked the sound as is (no EQ) but did not listen at loud levels as that requires quite a bit of voltage.
When that is not available during the demo and compares it directly to some other reference it may not sound great.
I have heard people say the same thing about the HE.

Yes, it doesn't measure $6k well but does the HE1 measure €70k well ?

To 617's earlier point, does any headphone sounds or measures well enough to be worth more than $1000?

Notice the lightly increased almost Harman bass levels for HE1 which Susvara lacks and similar 'clarity dip' around 2kHz. Aside from the bass levels how far is the tonality off ?

Judging from Oratory1990's graphs, I think Susvara is a bit too much treble tilted, and would probably benefit from a bass shelf while HE1 has much better tonal balance.

HE-1

Susvara

I personally don't like high tilted headphones and will for sure EQ them. But as you pointed out, it might be exactly inline with the taste of its target audience

 

[caught gesture]

No I do not. And you are awfully predictable.

HIFIMAN Susvara Headphone Review

I own it and I don't see anything wrong with the measurements. Susvara is literally designed to have almost no dampening whatsoever hence the massive group delay / reflections and that zig zag FR curves. Shocking to see AHB2 running out of steam at the same level A90 at 114 dB SPL, wow I'd be...

audiosciencereview.com

Apologies, I did not see your previous post. In that post you wrote, I am merely objecting to the idea that one can make objective claims about a headphone unless it is demonstrably broken. Are you saying that it is impossible to measure a headphone and then make an objective claim?

 

[amirm]

@amirm, have you thought about reviewing the Smyth Realiser, it would be super interesting, and quite a little project of a review!?

A bit of background. I met the designer at Cambridge AES conference in 2008. And listened to what was one of the most incredible things I had heard. He had a 5 channel system and you would put on and take off the headphone and could not tell the difference whether you were listening to speakers or headphones! The designer by the way was the same person who came up with DTS compression algorithm for movie sound. So we had a lot in common and had a great chat about the industry.

So yes, would be great to test one although it would not fit the style of these reviews. It is like testing room EQ or something like that.

 

[amirm]

Apologies, I did not see your previous post. In that post you wrote, I am merely objecting to the idea that one can make objective claims about a headphone unless it is demonstrably broken. Are you saying that it is impossible to measure a headphone and then make an objective claim?

That is what he is saying which makes no sense. He is confusing variability/accuracy of measurements with objectivity (calling them subjective).

 

[OnLyTNT]

Yet what it wrote nearly matches everything a subjective reviewer would say about this headphone! This is why words alone have little value in evaluating the sound of the headphone.

Ahh yes, a subjective reviewer/listener behaves like LLM in a sense, they all repeat what other so called "experts" say or what is "statistically accepted as true"...

 

[Merkurio]

No, just because the graph crosses the line does not mean it creates audible distortion.

Here are 3 examples for you, from headphones which has higher distortion in the same general area, that have have no static or noise issues and are recommended, some of them without EQ.

No, just because the graph crosses the line does not mean it creates audible distortion.

Here are 3 examples for you, from headphones which has higher distortion in the same general area, that have have no static or noise issues and are recommended, some of them without EQ.

I'm just saying that, contrary to what you were asserting, there is indeed information within Amir's measurement methodology that objectively demonstrates the existence of audible distortion at 94 dB in a region where our ears are quite sensitive to it (5 kHz).

Amir's subjective impressions or final conclusions about other headphone models are beyond my point, as I hope you can understand.

 

[IAtaman]

Those measurements show major deficiencies in both tonality and distortion.

With all due respect, both of those statements are incorrect.

It's tonality is not bad. It is actually good according to Harman reasearch. Its deviation is less than 2dB from target and it has a treble tilt, but overall it performs in the same ballpark as Stealth or HD650, your 2 go to headphones. You might not like the tonality, but it does not make it objectively deficient.

Distortion is also not great but not terrible either and looking at the graphs objectively, there is no reason to think any of that will be audible.
In fact, you yourself did not find them to be audible for other headphones as I have pointed out in this post.

It also massively clips at 115 dBSPL where hardly any headphone does. I listen and verify that those deficiencies are indeed there. It is this totality of information that tells me it is not a good headphone and certainly misses the mark at what it sells for.

Stax SR-009S does, according to you. In that review you said "I did have to back off to 110 dBSPL as anything above that caused severe distortion". A headphone you found out to be a wonderful headphone and highly recommended for low to medium level listening. Why exactly distortion at 115dB SPL is a no go for this headphone, objectively?

Compare that to other reviewers who spend pages talking about a headphone with measurements either absent or afterthought. You can read them 10 times and still not know exactly what they are saying about the performance of the headphone. And even if you did, you read the next review and that will set you right back to confusion.

I am not comparing your reviews to that of other people. I don't care about other reviews. I am pointing out the fact that many people have listened to Susvara, including myself and Solderdude, and have not heard the static you heard.

 

[Galliardist]

I had a friend who who lived in an apartment (in the US), he preferred speakers but is a headphone user, it wasn't necessarily for space savings, but really because he just doesn't want to disturb his neighbors and have them complain. After awhile, he really got into headphones, after all, enjoyment of something less preferable is still better than no enjoyment.

What I don't understand is, he owned his apartment, why didn't he just sound proof it, not just for listening to audio, but for general privacy and decibel freedom for when you have friends over, etc. Understanding it's easy said than done (particularly for those who aren't as handy) I priced out for my friend how much it would have cost to sound proof a 1,000 sqft apartment if he did it himself (and I offered to help out and how him what/how to do it), it's about $6k-$7k at the time, which is the cost of this horrific performing headphone. He ultimately decided against it because it was just too big of a project to take on while the apartment was occupied.

For those who own and lives in an apartment, seriously considered sound proofing your unit. It's not all that expensive and it will increase your resale value.

You'd have to get something like that past building management. I wouldn't be allowed to, for one.

 

[MayaTlab]

Actually, there is:

The audibility of harmonic distortion depends on frequency, level, order and stimulus, and can't be represented with a straight line. Its impacts on preference for headphones is not well established (per Harman among others). You can disregard the threshold line above, it's mostly arbitrary.
Without knowing whether or not the Susvara behaves like a minimum phase device, you can also disregard the group delay graph - in all likelihood what you're seeing there is just a different way of looking at the same data that produced the frequency response graph and provides no additional information (in other words the latter graph is enough to highlight some issues with the Susvara's performance).

That's not to say that the Susvara performs particularly well - I've always felt like Hifiman, for their open planars, stumbled by accident over numerous iterations on a formula that performs alright-ish - particularly in practice, in situ - instead of doing so deliberately, as a product of sensible engineering (and as soon as they're trying anything else than open planars the results are crap), seems very eager to milk some people's money with difficult to justify prices given the performance vs cheaper headphones in their lineup, and I'd gladly see this sort of product not being released, but it doesn't measure worse than some other headphones that have been recommended with EQ here.

 

[IAtaman]

I'm just saying that, contrary to what you were asserting, there is indeed information within Amir's measurement methodology that objectively demonstrates the existence of audible distortion at 94 dB in a region where our ears are quite sensitive to it (5 kHz).

Amir's subjective impressions or final conclusions about other headphone models are beyond my point, as I hope you can understand.

That line is a line Amir draw. It is not a proven fact of THD auditability. I don't think Amir meant it as an absolute truth either - it is just a guideline. Just because something crosses guideline does not necessarily mean it will be have audible distortion. That is a fact. I have provided the other graphs as examples of that fact.

 

[caught gesture]

That line is a line Amir draw. It is not a proven fact of THD auditability. I don't think Amir meant it as an absolute truth either - it is just a guideline. Just because something crosses guideline does not necessarily mean it will be have audible distortion. That is a fact. I have provided the other graphs as examples of that fact.

So one is able to determine objective facts from measurements of headphones (without the headphones being broken)?

 

[IAtaman]

Apologies, I did not see your previous post. In that post you wrote, I am merely objecting to the idea that one can make objective claims about a headphone unless it is demonstrably broken. Are you saying that it is impossible to measure a headphone and then make an objective claim?

So one is able to determine objective facts from measurements of headphones (without the headphones being broken)?

This also has been addressed a few times, can you search through the thread maybe.

 

[Merkurio]

That line is a line Amir draw. It is not a proven fact of THD auditability. I don't think Amir meant it as an absolute truth either - it is just a guideline. Just because something crosses guideline does not necessarily mean it will be have audible distortion. That is a fact. I have provided the other graphs as examples of that fact.

If we're talking about an aproximate guideline that can meet the arbitrary condition of showing (or not) the existence of audible distortion in that region (~5 kHz) and at that level (94 dB) for that particular headphone unit, what's the point of being surprised by Amir's findings then?

The way you phrase it suggests either you're very familiar with the specific model (even though you say you don't have it), or you're relying on the credibility of a majority that hasn't made clear the methodological conditions under which they don't hear such static/distortion.