HIFIMAN Susvara Headphone Review

[amirm]

Here is a video I did on importance (or lack thereof) of phase (timing):

 

[Rhamnetin]

You cannot take apart the social representation facts. Lot of ads are working on this subject.

Here is how it works

These headphones (you can change headphones by DAC, amp, speakers...) are massively expansive.

They look expansive.

There is no measurements.

The experts on the tube are very impressed, as usual when it is expansive.

Some people buy the product.

@amirm review the product and measurements are not good.

Owners are proud of what they paid and have build mental construction about the product with the help of YouTubers: you need an expansive amp to use the headphones. Replace the two terms with what you want, IE amp by DAC and headphones by cables.

The mechanic is the following: defining a set of conditions unobtainable. There is a name for that: synergy!

And then messages on the thread are like the previous 13 pages.

Can't forget your 5 figure price tag power regenerator and power cables!

 

[Byght]

Something was wrong in the vocal and treble area. Not sure if it was distortion resonances or both maybe on top of that something from inverted polarity.... Who knows
But to me they sounded hazy etheral, yes open sounding but just strange and sometimes wrong.

Indeed Susvara is famously polarity inverted. To get the right sound stage you have to invert it again somewhere in your chain. I've seen other reviews mention this, and then used a mobile app to check my own and confirm the inversion. No other headphone in my personal experience has this issue.

@amirm, I wonder if swapping the polarity of the test setup would change any of the results?

 

[Coverpage]

Only when they have such a messy, non-minimum phase response as the Susvara. Even then, equalization works very well based on measurements.

Can you EQ one headphone to sound the same as another headphones?


That is the TF (represented as the FR Curve) of headphone 1, vs the TF (represented as the FR Curve) of headphone 2, going through a filter should sound the same for an LTI system.

 

[majingotan]

Indeed Susvara is famously polarity inverted. To get the right sound stage you have to invert it again somewhere in your chain. I've seen other reviews mention this, and then used a mobile app to check my own and confirm the inversion. No other headphone in my personal experience has this issue.

@amirm, I wonder if swapping the polarity of the test setup would change any of the results?

My test tracks when plugging in the Susvara showed this. I have performed the Absolute polarity test by playing this test track while inverting the DAC polarity and found the correct sound presentation with my DAC in the inverted polarity setting (I have my exact setup on my sig

)

 

[Coverpage]

In some cases, those $20 wipe the floor with the Susvara. So measurements absolutely tell the truth here. If they didn't have other drawbacks, they would indeed obsolete all of these headphones and in a hurry.

In the same way there is confirmation bias for those in the audiophile community who purchased the Susvara and expensive headphone gear, there's confirmation bias for those in the ASR community who believe the given measurements aresufficient to make the decision on which headphones to buy (without listening).


A blind test between those 20-dollar iem/headphones against the Susvara to subjects who are unbiased. I really think the answer is obvious.

 

[majingotan]

In the same way there is confirmation bias for those in the audiophile community who purchased the Susvara and expensive headphone gear, there's confirmation bias for those in the ASR community who believe the given measurements aresufficient to make the decision on which headphones to buy (without listening).


A blind test between those 20-dollar iem/headphones against the Susvara to subjects who are unbiased. I really think the answer is obvious.

LOL. There's no obvious answer. You'll get random choices due to poor sample size. Have at least 10,000 sample size then because if not see here:

 

[Coverpage]

Everything starts with tonality. Get that wrong and you don't go to these.

In real life, everything you hear has a messy time domain response. When someone talks next to you, their voice reflects from every surface and corrupts any sense of timing. I call this "phase soup."

Spatial qualities are evaluated subjectively in my reviews but they have strong correlation to lower treble where this headphone is poor at.

Finally, there is nothing "life like" when it comes to headphone listening. Sound comes from your head and maybe a bit around you. This is not remotely life like. Audiophiles make up stories about soundstage, and such but it is all myths as witnessed by them claiming they hear such differences in cables, amps, etc.

If you agree that the system is not a simple LTI one, then we need more than a frequency response curve to evaluate the headphones.

While tonality can be adjusted with an equalizer to achieve the desired FR curve, this often results in increased total harmonic distortion (THD). The critical, yet unanswered, question is: what specific phenomena contribute to this increased THD?

THD represents the deviation in the spectral content of the output compared to the input signal. In the time domain, these manifest as errors in reproduction. Identifying whether these errors stem from frequency domain or time domain phenomena is crucial for understanding how they affect the perceived quality. Not all THDs are equal.

The concept of 'lifelikeness' in headphones is a shot at describing the quality of a headphone; we then have to model and define it.

We could do this through a survey. For instance, in one survey comparing the Sennheiser HD600 with Stax headphones, known for their fast transient response, participants evaluated the 'lifelikeness' of these headphones using a curated set of recordings. It's likely that headphones like Stax, with quick transient responses, might be preferred.

We then need a set of metrics that will allow us to reliably predict the results of the above survey. These metrics will be more relevant to my audio preferences as a headphone buyer and people like me.

However, the goal doesn't necessarily have to be about modeling 'lifelikeness'. What's crucial is not to infer the superiority of one headphone over another based solely on current measurements. It's not sufficient.

 

[Coverpage]

LOL. There's no obvious answer. You'll get random choices due to poor sample size. Have at least 10,000 sample size then because if not see here:

Nope. This specific one. From mesurements from ASR. Susvara vs I think some IEM from moondrop .

 

[amirm]

Can you EQ one headphone to sound the same as another headphones?

You can get them within statistical tie as far as tonality.

 

[IAtaman]

Are you telling me my criteria for a headphone to play clean for as loud as I want when countless competitors do that with ease, is faulty???

Depends. What loudness level is that?

 

[Roland301]

Looks like a lot of resonances across the band, which correlates to the messy FR, and distortion behavior at high SPL in those spots.

 

[ergre]

If you agree that the system is not a simple LTI one, then we need more than a frequency response curve to evaluate the headphones.

While tonality can be adjusted with an equalizer to achieve the desired FR curve, this often results in increased total harmonic distortion (THD). The critical, yet unanswered, question is: what specific phenomena contribute to this increased THD?

THD represents the deviation in the spectral content of the output compared to the input signal. In the time domain, these manifest as errors in reproduction. Identifying whether these errors stem from frequency domain or time domain phenomena is crucial for understanding how they affect the perceived quality. Not all THDs are equal.

The concept of 'lifelikeness' in headphones is a shot at describing the quality of a headphone; we then have to model and define it.

We could do this through a survey. For instance, in one survey comparing the Sennheiser HD600 with Stax headphones, known for their fast transient response, participants evaluated the 'lifelikeness' of these headphones using a curated set of recordings. It's likely that headphones like Stax, with quick transient responses, might be preferred.

We then need a set of metrics that will allow us to reliably predict the results of the above survey. These metrics will be more relevant to my audio preferences as a headphone buyer and people like me.

However, the goal doesn't necessarily have to be about modeling 'lifelikeness'. What's crucial is not to infer the superiority e.a. of one headphone over another based solely on current measurements. It's not sufficient.

There are also distortion graph in Amir’s review. they are the measurements that indicate how linear the system is. So purely mathematical speaking he has covered both linear (frequency response) and not linear ( distortion ) cases. That Is it.

 

[IAtaman]

Can you EQ one headphone to sound the same as another headphones?

No, you can not.

You can mimic the tonality of an in-ear headphone with another in-ear headphone to the extend which they would score similarly in a preference test. That does not mean you can make two headphones sound the same.

 

[Roland301]

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

The driver itself should be designed such that these resonances don't occur.
By "design of the driver" I mean tensioning of the diaphragm, film thickness, magnet gap, magnet arrangement, trace design/manufacturing technique, etc.
An ideal planar driver shouldn't need external damping outside of that inherently provided by the pads/front volume to perform well.
(By "perform well", I mean in terms of resonance/distortion behavior and a relatively well behaved FR, not strict Harman compliance as that target does require external damping to achieve)

Internally, the Susvara looks more like a refresh of the original HE-6 than a totally fresh design. It definitely shouldn't cost $5k.

 

[solderdude]

We could do this through a survey. For instance, in one survey comparing the Sennheiser HD600 with Stax headphones, known for their fast transient response, participants evaluated the 'lifelikeness' of these headphones using a curated set of recordings. It's likely that headphones like Stax, with quick transient responses, might be preferred.

The reason Stax headphones are found to have 'faster transient response' is the peak nearly all of them exhibit at around 1kHz.
Add the same peak to a HD600 and it sounds 'faster' or lower that peak on the Stax and the 'fastness' of the Staxes is removed.
Arguably the HD600 is 'fast enough' as it can do well over 20kHz. It is a long standing myth about Stax headphones (the speed and transients).
The 'attack' people hear is in the 1-3kHz range. No 'fast membrane' needed for that.
Below some Stax headphones (and HD600/650) to illustrate the point:

Stax Lambda Pro.pdf

Shared with Dropbox

www.dropbox.com

Stax SR-007 mk1.pdf

Shared with Dropbox

www.dropbox.com

Stax SR-L300 LTD.pdf

Shared with Dropbox

www.dropbox.com

Stax SR-009.pdf

Shared with Dropbox

www.dropbox.com

and the HD600:

Sennheiser HD600.pdf

Shared with Dropbox

www.dropbox.com

HD650:

Sennheiser HD650.pdf

Shared with Dropbox

www.dropbox.com

 

[Eldus]

Although I guess Hifimans does not worth their price by any means, I haven't experienced their listening experience. In the other hand while 7hz Zeros almost nail the Harman curve, having them I can say their sound is very mediocre with poor imaging and soundstage. Truthear's Zero Red for example have more or less the same response but have a much higher detail retrieval and a much better imaging and soundstage.

With IEMs that probably has a lot to do with the tips and their seal/position in your ears. It is possible the shape of the bulk of the IEM facilitates a better fit for the Red, at least for you.

 

[McFarley]

One thing that could be deduced from the "static" sound amirm was hearing is that the poor Susvara's diaphragm was being smashed onto the magnets due to the SPL of 114db, bearable to neither human ears nor headphones.

 

[Coverpage]

No, you can not.

You can mimic the tonality of an in-ear headphone with another in-ear headphone to the extend which they would score similarly in a preference test. That does not mean you can make two headphones sound the same.

I agree with you. And anyone who have tried knows it’s not possible.

The point is it’s not an LTI system. Frequency response through a sine sweep cannot completely describe the system.

 

[LuminousMoth]

Of course there are. We don't do it to pass the time.

that response had me laughing out loud