Sennheiser HD800S Review (Headphone)

[Billy Budapest]

The Sennheiser HD800 and HD800S are kings of comfort. I am a previous owner of HD800 and never felt any discomfort even when using it for many hours. A headphone which is not comfortable should never be recommended.

I think the Oppo PM-1 was darn near the most comfortable headphone ever.

 

[tinytea]

Awesome headphone review Amir,

Definitely out of my price range but I will use the opportunity to try to understand the graphs for headphones.

Thanks for the good work

 

[Degru]

I don't at all like the idea, in loudspeakers but especially in headphones, of recommendations based on listening through EQ.

For loudspeakers I can see some semblance of justification. Loudspeakers don't move, and even if most "audiophile" gear is obsolete dross (no or primitive bass management; no room correction) at least speakers are fixed in place and DSP (built into modern equipment, or as an external component) is readily available.

For headphones there is, however, no reasonable justification. I don't think headphones should get a free pass on the necessary condition of pleasing tonal balance, unless the tools to improve it are included in the package.

The primary reason is the headphone amp market is in a state of complete and total failure. Almost universally, the products offered are sonically inferior to a 20 year-old HeadRoom device, There is more SINAD or whatever (BFD after a fairly low level of attainment) but no actually useful innovation in the form of modern processing (PEQ, crossfeed or more advanced room simulation) or the required UI. MiniDSP HA-1's had promise, but ended up a failure because they pulled iDevice compatibility at the last minute. So it ended up being unusable with the devices people actually use with their headphones. Nobody else seems to have even tried to make a useful headphone amp.

Furthermore, software EQ is simply not an acceptable ersatz. I guess there are a few people who prefer to listen to headphones at home. But the bulk of users will use them on the go, or in offices connected to locked-down computers. There is no opportunity to EQ them in such cases, due to the aforementioned market failure in headphone amps. So bottom line a pair of headphones with poor tonality (and no included corrective tools) are bad headphones, regardless of price, brand, etc.

There are things like IL-DSP, PowerDAC v2, and some bluetooth adapters with built in parametric EQ nowadays. If you use local files for your music, that also opens up a good few media player apps with PEQ built in. As for home use, it is not complicated to set up system-wide PEQ on Windows and plenty of plugins for media players and some also have it built in. And, I wouldn't use a headphone like HD800 as a portable headphone (well, maybe I would, but most wouldn't) so EQ would be an acceptable compromise.

But I do agree with you in the context of using gear portably with streaming music. System-wide PEQ for smartphones is basically nonexistent at the moment.

 

[MZKM]

I do think the impedance vs frequency is valuable. It puts to bed all those "these headphones are difficult to drive" statements. What voltage does the Klippel rig use for the speaker impedance sweeps? If it's only around a few volts, you could use that.

I would find it valuable to calculate the mW needed for desired SPL.

 

[maverickronin]

If it's only around a few volts, you could use that.

That would only work for mid to low sensitivity full sizes which have boring curves anyway.

HD800S from Tyll at Innerfidelity

Multi balanced armature IEMs actually have multiple drivers and crossovers so their curves actually get interesting.

Westone W60

The W60 also has an efficiency of 117dB/mW too so the Klippel will just destroy it...

 

[amirm]

Can you export the Harman target curve?

Sure but let's call it ASR Target Curve. There are reasons why this is not exactly the same as Harman's which I won't go into now.

 

[amirm]

But how many "consumers" out there have no clue and listen to HD580, 600, 650 (and other brands, and "99% of HP) just as they are, without EQ ?

Probably a ton of people but I am here to change that. We need to move people to usage of EQ.

 

[MZKM]

Sure but let's call it ASR Target Curve. There are reasons why this is not exactly the same as Harman's which I won't go into now.

Can you export it at the same resolution as the headphone measurements?

Unless someone more proficient with spreadhseets can help, I need the same Hz values for each in order to normalize the HP curve to the target curve. Or, if there is some equation to generate the target curve, then I can do it for any Hz value.

 

[amirm]

I do think the impedance vs frequency is valuable. It puts to bed all those "these headphones are difficult to drive" statements. What voltage does the Klippel rig use for the speaker impedance sweeps? If it's only around a few volts, you could use that.

Klippel that can that with ease but unfortunately it is in a different room altogether. So I am going to have to build a fixture for the AP.

 

[amirm]

Can you export it at the same resolution as the headphone measurements?

Hmmm. It came that way. It is a heavily filtered curve as otherwise it would be wildly different seeing how it is in-room measurement. I have no other options to increase its resolution. One would need to interpolate in-between samples.

What I suggest is filtering the frequency response to its resolution.

 

[restorer-john]

As long as we use the same ears to listen to things in everyday life changing the hifi to make a sound different to real life would obviously sound completely unnatural and wrong to anybody, whatever the state of their hearing.

Frank, you should use your outside ears for day to day stuff, and keep your audiophile ears near the HiFi gear for critical listening.

 

[amirm]

Can I get someone to volunteer writing a mini article on all the EQ options available for different OS platforms?

 

[amirm]

Time to stir the pot! How about 'next up' - planar magnetics. Maybe a HFM Arya vs. a cheapo Deva (wired) /HE 5XX. Or Sundara..Pick one single-sided magnets, one double-sided.. Are there differences? Is there a God?

I only have HE400i. Would that do?

 

[Harmonie]

Probably a ton of people but I am here to change that. We need to move people to usage of EQ.

Thank you Amir, you got my point.

Frankly, I do not listen with cans.
I own NC Bose with sole purpose to reduce the noise in the air plane as I travel extensively (well, Covid is changing that right now).
I remember the magic experience with Stax in "last century", and was among very first EU adopters of HP with the Walkman in 1980.
But that's about it.
Very curious to move to HP again, though I can still enjoy listening to speakers in the living room with no near field listening.
It's a different experience for sure.

I'm sure that there are millions like me still out there.
PS- Just saw your reaction about the Youtube thread.
It's worth a deeper thought.

Cheers.

 

[Ron Texas]

Headphone reviews starting with a bang! Thank you @amirm for the new adventure.

 

[Degru]

The thing is, Amir said he's measuring SPL at 425Hz, which is basically the same dB level as 1kHz when referencing the Headphone Harman Curve, and most of the frequency response is higher than this point (in dB), so given all that it seems that Amir's comments about 94dB being below normal listening level are a little crazy, I don't get that currently.

I think Amir is just talking about a 0dBFS test tone at 425Hz in terms of his comments on dB level (it's annotated in the following graph):
View attachment 99492
EDIT: ah, I see your point now RayDunzl, you're saying that would be the peak, which it would, and music is not always at 0dBFS......although a lot of modern music is consistently bouncing off 0dBFS if you analyze a few tracks in a program like Orban Loudness Meter....so I think the idea that 94dB (peak) is below normal listening levels, I find that a little strange & unbelievable still.

Thing is you should spec out your amp for *all* scenarios and not just the music you'll be listening to most of the time. Having extra headroom in general is just plain beneficial, especially if you are going to EQ. "Reference" level for music is widely considered to be 85dB, and highly dynamic tracks or stuff like movies can reach transient peaks of ~20dB+ above that. Or you might have songs and bits of audio that are quiet so you want the extra headroom to be able to turn it up. So you should get an amp that will push your headphones to ~110db peaks well.

 

[MZKM]

Hmmm. It came that way. It is a heavily filtered curve as otherwise it would be wildly different seeing how it is in-room measurement. I have no other options to increase its resolution. One would need to interpolate in-between samples.

What I suggest is filtering the frequency response to its resolution.

That’s what I figured. Google Sheets can easily “plot null values” for graphs, but I have no clue how to interpolate the actual raw data. Meaning, I cannot figure out how to replicate what their do for graphs for the data cells.

I downloaded that Webplotdigitizer tool a while back (really cool), thinking that could be a solution to these types of problems, but (to my knowledge) you cannot dictate it to digitize the same x values for both lines.

So yeah, anyone an expert on this that can help?

EDIT: I found this paper, but it involves ~30 formulas!

 

[Degru]

It is a matter of design philosophy. I read that Sennheiser calibrates their headphones against the "diffused field." They set up 8 speakers in anechoic chamber and measure and match what arrives at the "ears" of the measurement fixture. No real room has such a sound field yet no one questioned this until Harman came around and made the reference a good speaker playing in a room.

That aside, I can't understand why their HD650 and HD800S measure differently. Surely if they are all designed to comply with that diffused field, they would have similar response but they do not. So it is a puzzle. Likely it is the "sound they like" as they designed it.

Sennheiser tuned HD800 to some kind of "loudness diffuse field" target which I suspect is just made up to make the HD800 look neutral on their provided measurements

 

[Degru]

The distortion measurements are quite mediocre for headphones. The $400 HD650 actually measure slightly better. So what exactly is one getting for an extra $1000? It looks like $1k buys a boost at 5k and all the associated "detail" that comes with it. To me, that looks like a showroom gimmick and if these were speakers that would be a nearly universal opinion on this forum. Not sure why headphones aren't judged the same, particularly headphones that cost $1400.

Have you listened to the two in person? Actually listening to things makes it quite obvious how limited current headphone measurements are in capturing the full characteristics of any given pair.

 

[Robbo99999]

Thing is you should spec out your amp for *all* scenarios and not just the music you'll be listening to most of the time. Having extra headroom in general is just plain beneficial, especially if you are going to EQ. "Reference" level for music is widely considered to be 85dB, and highly dynamic tracks or stuff like movies can reach transient peaks of ~20dB+ above that. Or you might have songs and bits of audio that are quiet so you want the extra headroom to be able to turn it up. So you should get an amp that will push your headphones to ~110db peaks well.

(that's not really the point that was being made)