Discussion in 'Headphone Measurements' started by sorrodje, Apr 27, 2016.
Awesome compilation of measurements and info - thanks Sorrodje!
Will fix sr007 mkII FR plot. He60 measurements Will be provided as updates une the dedicated thread.
Will provide later some compared FR.
My comments about measurements will come as well.
SRM-727II modded and SRM-1/mk2 are not that bad...besides, final payment done three months ago, this loch-ness-monster-of-an-amp BHSE should be here before summer.
I've always had a look at all those DIY kg-something, especially those from Birgir who's slowly becoming a sort-of-industrial builder, but I've never seen anything from DIY that can compete with look / finish of the BHSE (IMO). Yes, DIY T2 is above, Carbon is in the same league, other crazynesses (nanotubes, grounded grid, megatron, DHT from Frank Cooter, etc...) are tempting...but what I really want is a BHSE
One of the big weaknesses of Stax-branded amps is in the PSU -- AFAIK virtually all of them (except maybe the SRM-T2) have wimpy unregulated power supplies. Some KG designs have PSUs that weigh more than entire Stax amps!
I feel the same way. The BHSE is a stunner and sets an aesthetic benchmark for all amplifiers to aspire to (including non-electrostatic and power amps IMO).
P.S. @Ali-Pacha it's refreshing to hear someone talk shit about the SR-007 Mk1.
Be prepared for an ass kicking!
Great thread guys! Well done!
A thread I can relate to! Great work Pierre65 and nice collection Ali.
SR-Lambda + transformer box/entry level amp is still the king of price : performance for true high end in my mind. Really nice measurements there. I also liked to see another personal fav - the Lambda Sig - showing super fast decay. A king of resolution.
Stax really aren't that expensive. Plenty of the stuff you guys measured is quite affordable given a person doesn't mind vintage gear. No, they won't show up in youtube unboxings, magazine ads, forum reviews, group buys, or at tradeshow booths, but these antiques still wipe the floor with just about anything offered these days. The 'newer = better' mentality largely doesn't apply to audio. Give some of these an honest look and I think most people will be impressed or shocked.
Pierre65..... you little piece of Serbian Shit @n3rdling
Awesome work guys!
BTW, don't mess with Milos. He is also half Mexicano with large huevos.
What can Stax do what a well amplified HD800 can't?
Less effin' ringing.
Except for the unicorn modded ones with accurate midrange and treble timbre.
Spoiler: I'm not qualified to make such a judgement, but...
Clearer and faster sounding. Bass is probably more extended in terms of frequency response (depending on seal).
(at least, going off the many occassions I've heard them at meets or shows or dealers, ... - I've never heard the SR009 and HD800 amplified to their best so I really hate commenting on this.)
EDIT: I'm very interested to hear what @sorrodje thinks about this.
Sorry, this Thread is dedicated to Stax headphones and measurements. Not to another sterile HD800 vs Stats debate a la Head-fi.
To be honest, I'm a bit tired of " headphone X vs headphone Y , what's the best" . The more I learn, the less I know and I consider my summit fi is composed of a bunch of headphones and amongst them some Stax, the HD800 and some other headphones . Each of them does thing not other can do and all can sound stellar and average depends on what I listen to.
For example , I own a HE60 and a HD800 and each night is different enough to make me enjoy one or another equally depending on my mood and the music I want to listen to. Just bought the 007 mkII and I enjoy it a lot as well even if it's very different from the two other ones.
Large eggs?... Slang for testes or what?
LOL! I don't want to derail this thread by the slightest amount. It's too awesome. But yes. That's slang in the Spanish language. The equivalent in English would be balls.
More in line with this thread, I relate to the measurements here and the discussion makes sense to me. I'm not certain the treble response is too far of from what I hear relative to dynamics. It is possible that what I hear as a little liveliness on the lambdas is related to their elevation in the 4kHz range relative to other frequencies. They are not ear drilling lively in the DT990 sense by any means. The liveliness in such a region together with the low distortion may bring up detail in ways many other headphones cannot.
I don't remember hearing the SR-5, but based on what I see relative to the other Stax peers, I can see how many would consider them a little boring (relatively speaking). I'm now kind of curious about them, because I like what I see and would love to hear them.
The sigma pro I've heard in the past, and my impressions match the measurements to a large extent. This is not a headphone I like.
Also, based on what I'm seeing, the 007 does indeed look more laid back than the 009, and does show a little more extra hump in the bass region.
This is one of the best threads in Stax I have ever seen. Very informative and concise.
@sorrodje and @Ali-Pacha Awesome thread !!!
I just bought a SRD-7/mk2 adapter and a Lambda Pro from another source (rotten ear pads and foam, so I ordered a set of leather EP-507 ear pads). How do the Lambda Pro and Lambda Standard/Normal Bias compare to each other?
Here are some measurements of Milos SR-009 to complement Sorrodje's measurements. There is also some distortion showing up in the 5-6 kHz region in the left driver. I think this shows in Sorrodje's measurements but in the right driver.
These measurements are not averaged. I did not run multiple measurements to get better distortion measurement resolution and my ADC is not that great single ended.
And here are some measurements on the SR-307 (Lambda). For some reason the SR-307 I measured did not seem as recessed in the 5-10 kHz region. I had a lot of problems with seal on the left driver. I also get higher distortion numbers in the 4-5 kHz region, but these measurements might be limited by the amplifier used which I think was a stock SRM-323 instead of a T1 (as was the case for the SR-009 above).
The deep null between 4-5 kHz may be a measurement artifact.
(EDIT: Please note I screwed up the coloring scheme here, green is right and red is left.)
Subjectively I did not find the SR-307 out of the SRM-323 all that engaging. Pushed sufficiently hard, the SRM-323 exhibited some distortion issues.
This one goes to both @ultrabike and @sorrodje:
What shows up as higher distortion in those measurements is very likely a measurement artifact caused by the dip in the frequency response (which is also a measurement artifact): Because the response is lower at that point, the S/N ratio drops and what we see as a distortion spike is simply the null in FR. Kind of like how room resonances shape the distortion in speaker measurements in the listening room.
This also makes sense because the left channel in your SR009 measurements, which also has a deeper dip at 5.5kHz, shows much higher "distortion" at that frequency. I think the same measurement artifact could be seen with the unicorn LCD2 - interaction with the v2 coupler and not indicative of the headphone.
The distortion of STAX electrostatics is so low that to we would need much more expensive measurement equipment to properly meausure it (higher S/N ratio). It could even be that a big chunk of the D2 distortion is the measurement system, depending on the mic, preamp, etc.
EDIT: This is also why we see D3 and D4 distortion rising at higher frequencies with most of @sorrodje's measurements: The measured frequency response is lower at that point and the S/N ratio drops. Again, this is not indicative of the headphones.
@Serious, your explanation on SNR is indeed reasonable.
However, I do not think that we are experiencing mic or preamp issues at all. It is possible it is a coupler issue. I don't think you need much more expensive measurement equipment to overcome this.
One needs IMO to come up with a coupler that can absorb possible standing wave issues in the region. In fact, one may attempt a free air measurement which may be valid in the treble region and should not have any coupler interaction problems.
Another way to verify that these are indeed coupler problems would be to run a few quick tone sweeps. I understand there are Fletcher-Munson curve issues when doing this. But I think sufficiently deep nulls dominate the curve characteristics. I think I talked about this with Milos in the past.
As far as the theory that Stax electrostatic headphones are incredibly low in distortion and reasonably priced measurement equipment is not up to the task, consider that the mics used for these measurements are (small diaphragm) electrostatic in nature themselves. Note also that a lot of Pro audio mic interfaces are likely balanced, and Pro audio stuff seems pretty good when using balanced I/O. Most stuff IME is background noise which can be averaged out and indeed show reference grade driver limitations.
@Serious. Thks but I think neither @ultrabike nor me are ignoring those facts. I wrote warnings in the first post for good reasons. Please read it and wait until I'll add comments for each headphone.
@ultrabike. Yup I always listen myself to sweeps each time I think measurements show artifacts. Huge dips in all above FR are mostly artifacts.
BTW. I still think the best way to present measurements is to leave them as they are. Flaws included. It should help readers to keep the good distance with them and don't take same as gospel.
Separate names with a comma.