Feb 25, 2010 at 12:09 PM Post #16 of 29
Quote:

Originally Posted by John Buchanan /img/forum/go_quote.gif
Any time you and your wife and son would like to visit, feel free to contact me. WA is a lovely place to visit.


Thanks, John. Yes, I hear WA has so much to offer...sun...surf...sealife...and some of the best headphone sound the planet has to offer
wink_face.gif
 
Feb 25, 2010 at 12:24 PM Post #17 of 29
Quote:

Originally Posted by webbie64 /img/forum/go_quote.gif
Thanks, John. Yes, I hear WA has so much to offer...sun...surf...sealife...and some of the best headphone sound the planet has to offer
wink_face.gif



....and some Quad 989s
 
Feb 25, 2010 at 12:43 PM Post #18 of 29
Quote:

Originally Posted by John Buchanan /img/forum/go_quote.gif
....and some Quad 989s


Stop it.

Stop it, stop it, stop it, stop it. LOL.

(I'm sitting in the loungeroom at 11.40pm with my Martin Logans right before me but 'choosing' headphones because all others in the house are asleep...

...but I guess I could start looking online for cheap airfares to WA...
tongue.gif
)
 
Feb 27, 2010 at 6:16 AM Post #19 of 29
Does anybody have any suggestions for equalization for the Sigma's rolled of highs and lows?

Although I enjoy them a lot, I frequently adjust my parametric equalizer to approximate what Lloyd (http://www.head-fi.org/forums/f4/hi-...-sigma-187076/) quoted:

"On the artificial ear the response to 1.5kHz was smooth and free from major deviation, with the low frequency limit set at about 30 Hz (this agrees with the subjective appraisal which also showed inaudible distortion at reasonable sound pressures.) However, the 2-5kHz band was clearly depressed by some 7dB or so, before recovering towards 10kHz, the latter part somewhat exposed relative to the adjacent areas. Reasonable correlation was obtained on the Neumann head, though a bass hump was indicated at 60 Hz and the shape was somewhat altered in the 750Hz to 8kHz range......

GENERAL DATA

Frequency response 100Hz-5kHz, rel. 500Hz (deviation from mean curve).......+4dB, -9dB

Frequency response overall within +or-5dB......28Hz to 2kHz"

They sound more detailed when I apply those changes.... still with the wonderful mids they are known for.
 
Mar 5, 2010 at 7:13 AM Post #20 of 29
I'm not sure that measurements of headphones are simple. It appears that there are widely divergent graphs for the same phones, and certainly the sound doesn't seem to reflect what some of the graphs portray. Stax's own graphs show a very flat bottom end for the LNS, whereas I have seen them graphed with a climb up to a hump in other graphs. They don't sound like the latter to me.
 
Apr 7, 2010 at 7:49 AM Post #22 of 29
I periodically alternate the Sigma/404s and the LNS on the Stax Monitor. The Sigma/404s sound even better if pushed harder, much as the original low bias Sigma did (up to a point when the latter started to shout a bit). Mid (not low) bass monsters, with a lot more frontal space. These really are great phones. In a way, I'm glad I let the 007 Mk1 go and left these 2 beauties behind. I'm a very happy little camper.
 
Apr 7, 2010 at 8:32 AM Post #23 of 29
Quote:

Originally Posted by John Buchanan /img/forum/go_quote.gif
I'm a very happy little camper.


Excellent news - after all, it's what we all strive for
wink_face.gif


(Personally, though, I've only ever taken the SR001Mk2 or the SRD-XPro camping LOL. So not enough to drive Sigma 404s
rolleyes.gif
)
 
Apr 7, 2010 at 8:34 AM Post #24 of 29
Quote:

Originally Posted by gilency /img/forum/go_quote.gif
Does anybody have any suggestions for equalization for the Sigma's rolled of highs and lows?

Although I enjoy them a lot, I frequently adjust my parametric equalizer to approximate what Lloyd (http://www.head-fi.org/forums/f4/hi-...-sigma-187076/) quoted:

"On the artificial ear the response to 1.5kHz was smooth and free from major deviation, with the low frequency limit set at about 30 Hz (this agrees with the subjective appraisal which also showed inaudible distortion at reasonable sound pressures.) However, the 2-5kHz band was clearly depressed by some 7dB or so, before recovering towards 10kHz, the latter part somewhat exposed relative to the adjacent areas. Reasonable correlation was obtained on the Neumann head, though a bass hump was indicated at 60 Hz and the shape was somewhat altered in the 750Hz to 8kHz range......

GENERAL DATA

Frequency response 100Hz-5kHz, rel. 500Hz (deviation from mean curve).......+4dB, -9dB

Frequency response overall within +or-5dB......28Hz to 2kHz"

They sound more detailed when I apply those changes.... still with the wonderful mids they are known for.



Quote:

Originally Posted by mobbaddict /img/forum/go_quote.gif
Someone posted this article on the DT48 thread, it explains a lot of things about measurements
AST : Vol. 25 (2004) , No. 4 pp.276-285
See the frequency response graph of the Lambda Pro at page 4. Frequency response bass is artificially rolled off with a 100 hz bump, treble is artificially rolled off as well.



Don't run too far with this! Here is an article that explains things:

Stereophile: Between the Ears: the art and science of measuring headphones

But the basic premise is that, since phones generally fire directly in the ear (and bypass certain features of the head/ear) rather than from in front like a live concert or speakers would, a correction factor has to be applied to compensate for this. There is not complete agreement on this, but a feature of both the free-field and diffuse-field models is that there should be a dip between 2 and 5 khz, just as the measurements sited by Gilency show. This dip is intentional and should NOT be EQ'ed out for flattest perceived response. In fact, the models says the dip should actually be deeper...

Kevin
 
Jan 25, 2011 at 10:26 AM Post #25 of 29
Hi All,
  I have a pair of Sigma (normal bias) en route from Japan courtesy of Head-fie'r Marcus, AKA Bat King. I'm really excited to get these and was looking around Head-fi for Sigma EQ information. Being in the music business {performance/composition/design & production} I have no issue applying a *bit* of cut or boost (EQ) to get a smoother frequency response *if it doesn't slaughter the phase* (which is system+frequency specific). I have mastering quality equalization available to me so I'm curious how well these phones respond to a bit of bass boost? How easy is it to cause over-excursion resulting in arcing?
 
From looking at the SR-Sigma response graph on Wikiphonia and comparing it to the standard Fletcher Munson Equal Loudness Contour one might conclude these phones are pretty decent reproduction tools with 2 major flaws.The bass is in strong decline by 70hz and there is a -15dB hole between 5-7khz (which should probably be around -10dB).
 
Minor response problems aside, how can we fix these larger issues? Well, a linear phase EQ in the 5-7Khz range boosting ~3dB on a soft curve would probably benefit most peoples listening experiences without destroying the music. That was easy, but how to bring the bass more inline without altering Midbass phase data making real recordings lose their authenticity?  
 
This is purely conjecture at this point as I haven't received the phones, but one experiment I will try is to set a minimum phase high pass filter at 40hz and then bump the Q to 1.5 so it peaks from 40hz-60hz compensating for the Sigma's natural roll off. The clever part of this config is that the high pass filter boosts our bass response above 40hz and quickly cuts it below which has a beneficial effect on the phones performance. By strongly cutting below 40hz we reduce the diaphragms excursion requirements, reduce the danger of arcing and probably increase the dynamic range of the system (although I'm not familiar with ESL physics so I can't be certain of this last component).  
 
The primary downside to such an EQ profile is phase response alterations at the low end. Considering these headphones are already "phase nebulous" with the amount of reflected information they rely on from the enclosures, I'm considering this method as it may have only positive impact on the sound. I'll report back after I've received the phones and tried some EQ experiments. In the meantime has anyone found lite EQ'ing of the Sigmas to be useful?
 
Thanks to all the posters writing about the Sigmas, esp John B. I've read many of your posts and it informed my decision to seek out a pair.
 
Hopefully I haven't veered too off-topic. I'm new here and noticed EQ details being discussed above.
Anthony Bisset
 
Jan 25, 2011 at 5:24 PM Post #26 of 29
Anthony, take care with trusting frequency response graphs. I've seen many Stax graphs for Lambda headphones included in their manuals and they are always as flat as can be in the lower mids down to deeper bass, whereas the same phones measured by hifi magazines always shows a 100 - 150 Hz peak followed by a roll off in the deeper bass region (Stereophile did a mass comparison of phones many years ago - all phones had that same basic response in the bottom end). The frequency response cited in Wikiphonia is one of the latter. I would suggest that you would be better listening rather than depending on those graphs. I hear the normal bias Sigma as rolled off at both frequency extremes (I found the treble roll off particularly sad), and they become a much better phone with a SR-404 driver implant. Not a single drawback to that venture that I've ever heard. The phones are about the same efficiency as an SR-007, if you are familiar with that model, so be prepared to turn 'em up. The original Sigmas also came with silistors to limit input into these phones, as they were sold with an SRD-7. The SRD-7 could be driven by any amplifier and some people drove them beyond mortal limits, so Stax decided to limit how hard they could be pushed to avoid driver arc welding. I hope you like the Sigmas - they certainly aren't for everyone, but they scale up very well with improved amplification. An acquaintance over here in Perth has ordered a BHSE and I really want to hear them through that. They sound great via the SRD-7 backed by a largish amplifier. In some ways, they remind me of the old Magneplanar Tympani 1D speakers I had - they like big amplifiers and wake up when pushed a bit. The Tympanis had a unique quality to them - it was quite easy to converse when listening to them. Total lack of the ubiquitous midrange peak found everywhere else. And so goes the Sigma. Vocals and piano reproduction to die for, in my opinion. 
 
Jan 27, 2011 at 12:51 AM Post #27 of 29
Hey John,
 
Thanks for the data.
 
As far as I can tell, many of these test variations are due to one group measuring the phones and then computing an adjusted frequency response graph while the other group simply measures the phone and shows the data unadjusted. One graph is compensated for an average listeners ears while the other is simply what the headphone did to a microphone. Of course, real world situations are more complicated, but even companies that pride themselves in rigorous scientific process frequently fail to communicate clearly when the marketing dept is trusted to publish the facts.
 
Yesterday I read your summary of diffuse field equalization and that further complicated my understanding of measurement frameworks.
 
I would be pleased if industry leaders who value engineering ie., (Stax, Sennheiser, Audeze, Beyer, Etc.) embraced a standard showing 1/6th or 1/12th octave graphs with explicitly stated adjustments and detailed test conditions.
 
Here's a good example of smoothing and how it masks reality:
 
http://communitypro.com/files/literature/tech%20notes/SMOOTH_TECH.pdf
 
cheers,
 
Jan 27, 2011 at 6:28 AM Post #28 of 29
Also, I'd like to add that the condition of the damping material inside the housing plays a significant role in the sound. If I look inside the pair I currently have here which supposedly had a service not so long ago, the shape and quantity of the damping is certainly a bit different and I hear a bit more treble than I used to compared to the past sets (I had two others and they also sounded different compared to each other side-by-side) and a bit less bass as well, both are a welcome change. Still, it will never be a balanced phone (much more of a "fun" phone), that's not the reason why we all like the Sigmas.
 
So I think that the biggest possible tuning would probably be to find an optimal damping material for the phones or just reach an optimal tuning with the stock one.
 
Jan 27, 2011 at 7:38 AM Post #29 of 29
Quote:
From looking at the SR-Sigma response graph on Wikiphonia and comparing it to the standard Fletcher Munson Equal Loudness Contour one might conclude these phones are pretty decent reproduction tools with 2 major flaws.The bass is in strong decline by 70hz and there is a -15dB hole between 5-7khz (which should probably be around -10dB).

 
If you take a look at Tyll's measurements, many hi-end headphones show a similar dip beyond 1 khz (the most similar being the LCD2 and the HP2). According to Lunatique's measurements the O2 seems to have it as well. I believe HRTF naturally compensates for that.
You can find the measurements here: https://sites.google.com/a/headphonehobbylab.org/filehost/home/files/CanJam2010HeadphoneMeasurements.pdf?attredirects=0&d=1
 

Users who are viewing this thread

Back
Top