Saturday, February 2, 2013

Sony LiveSound™ MH1 & MH1C


Sony LiveSound™ MH1 has caused quite a hype among headphone enthusiasts due to the high value in spite of its extreme budget-friendliness. And once its designer, Mr. Sead Smailagic of Sony Mobile, explained its technical aspects in detail with help of a headphone blogger, MH1 truly kick-started a new era of budget-fi in headphones, and it came to a point of which MH1 was started to be compared to full-sized headphones in $100~$200 range.

In order to verify what has been explained, I ran a series of electroacoustic analysis on a pair of Sony MH1C. If everything was proven to be true, it would've been rated among the very best I've measured. However, with its overwhelming amount of bass boost up to 24 dB, I simply concluded all those explanations were simple lies reinforced with crowd manipulation. Of course, many MH1 and MH1C owners showed frustration, insisting their pairs sounded better, and my statements were too harsh. I knew some were definitely upset, but I continued to press on, as I've always done when dealing with irresponsible manufacturers, such as Heir Audio.

And then, after two months since my initial analysis, something happened on January 8th, 2013: Mr. Sead Smailagic, the designer of MH1, contacted me personally and showed me a friendly gesture, asking me if I'd want to give another shot on MH1, assuming the previous pair was defective. The cause was totally appropriate, and that would definitely make sense. so I promised Mr. Smailagic that if data from the secondary analysis turned out to correlate his explanations, I would give him a public apology.





Introduction


Previously, these opinions have been presented to Mr. Smailagic's statements.

Statement #1: Increased bass for the missing 6 dB effect compensation
- Still debated, but remains debunked by Rudmose (1982) until more conclusive research data come up
Statement #2: Effective Free-field localization 
-  interpersonal differences in ear geometry skew frontal localization, making free-field target inappropriate for headphones, presented by Thiele (1986)
Statement #3: Equal-loudness compensation for low level playback
- Possible, if the bass can be decreased

While scientific verifiability of #1 is still in limbo, #2 can be assumed as an application of the intermediate free-diffuse field hybrid, which is a common practice of the industry. However, Mr. Sead Smailagic states MH1 is designed for shallow insertion, and the frequency response at such insertion depth shall be acknowledged as MH1's reference target. Meanwhile, #3 can be easily verified once the low frequency exaggeration disappears.


General analysis

Note: for the sake of fairness, the insertion depth has been set right at the reference plane, as it is a part of the testing methodology. Three samples of MH1 (6 units), two samples of MH1C (4 units), and the previous MH1C samples (2 units, disregarded for averaging) have been separately analyzed.


PRO: MH1C turns out to be slightly more bass-oriented than MH1, due to the damping material being higher in acoustic impedance, according to Mr. Smailagic. A well-extended high frequency can be seen as well.

CON: Slow & second-harmonic distorted bass oriented sound, but yields a realistic booming sensation at the same time. Depending on individual preference, this issue may turn out to be beneficial.

ON SECOND THOUGHT #1: As seen above, the overwhelming bass seen from the previous experiment is gone. Although the IEMs are still not bass-shy, the data prove MH1 and MH1C exhibit very nice electroacoustic performance, and my previous samples were 100% defective.

ON SECOND THOUGHT #2: As previously verified by myself, and later confirmed by Mr. Smailagic, Sony MH1 and MH1C are originally developed to be shallow-inserted. As the IEMs are situated approximately 3mm away from the reference plane, high frequency bandwidth extends up to 20 kHz, and the overall frequency response becomes closer to that of a conventional diffuse-field target.

ON SECOND THOUGHT #3: modifications such as vent blocking and stem foam modification will be omitted, since they have already been extensively covered by Mr. Smailagic.


In-depth analysis

Interunit variances

The interunit variance of MH1 and MH1C is extremely low, as the channel tracking does not exceed more than 2 dB at most within 20 ~ 10,000 Hz range. Surprisingly, one pair of MH1 shows no more than 0.5 dB deviation within 20 ~ 10,000 Hz range.


A front vent modification

According to Mr.Smailagic, a front vent was originally planned for MH1, but the concept was never actualized due to time/cost issue related to production. The vent is projected

1. To prevent damage caused by over-pressurization
2. To maintain fitting / leak tolerance consistent
3. To prevent occlusion effect

In order to investigate this front vent scheme, a small hole is drilled, and Knowles dampers are inserted for detailed simulation.
To retain the sub-bass response, utilizing acoustic impedance less than 3300 ohm is not preferred.


The diaphonic membrane experiment

According to Stephen Ambrose of Asius technologies, LLC, a diaphonic membrane installed along the canal-to-iem coupling can greatly reduce overly emphasized bass and the occlusion effect by removing the tymphanic membrane displacement caused by the driver's over-excursion. If this can be realized on MH1, the bass boost will be greatly reduced. A 5mm wide hole is created on MH1's sleeve, and the diaphonic membrane is created by applying a bubble of extremely diluted polymer on to the hole from the inside.

Although the eardrum excursion is definitely eased, the overall bass response still remains to be quite emphasized. Since the acoustic effectiveness of such diaphonic technique is heavily dependent on the amount of acoustic impedance the driver(ventilation, compliance, etc) is designed with, the efficacy on MH1 is proven to be minimal.


Conclusion


Sony LiveSound™ MH1 and MH1C have been extensively analyzed, and my opinions on MH1 are further verified. Upon analysis, the IEMs turn out to be excellently diffuse-field oriented, instead of free-field oriented when inserted shallow. Moreover, as current samples exhibit much less low frequency response compared to previous test samples, it is assumed that MH1 and MH1c can be considered to be tuned in accordance with the equal-loudness contour, which may yield flat perceived response at lower volume level. Thus, I am absolutely convinced his explanations are not commercially oriented excuses to sell a product at all.

Thus, in summary, I indeed owe Mr. Sead Smailagic of Sony Mobile an apology. I am deeply sorry for drawing a premature conclusion, solely based on defective test samples, and using harsh expressions. I am only a human being, and I too learn from mistakes just like anybody else. :) I appreciate Mr. Smailagic for letting me to have an opportunity to correct my mistake.


16 comments:

  1. http://www.youtube.com/watch?v=6B9vNqMh0VQ

    I just watched this video clip on diaphonic membranes and it seems like its something that you'd hear but rather not measure?

    Does it really sound louder modded even though they might measure at similar SPL?

    ReplyDelete
    Replies
    1. The inventor electroacoustically measured the effect, which is called 'trapped volume insertion gain'(TVIG), with Skullcandy Titan: http://asiustechnologies.com/docs/ADEL%20Retrofit.pdf

      I simulated the experiment with my Skullcandy Titan long time ago, and it worked.

      Delete
  2. Mine have approx 15 to 18 db's of bass boost. Seems they have terrible QC. I'm definitely disappointed for the discounted price I paid. Full price would be a crime.

    ReplyDelete
  3. Great article Rin, thanks for using that "read more" link like I suggested as well, I thought you didn't see that comment :p Again, great article, these headphones do seem very interesting in terms of measurements. They do remind me a lot of many of Monster's headphone measurements as well (similar curve; to different degrees though).

    ReplyDelete
  4. Review on mh1a? The one which is short and comes with mw1??

    ReplyDelete
    Replies
    1. Whether the model name is 1A or 1C, if the filter is black, the sound should be of MH1. And if the filter is white, then it is of MH1C.

      Delete
  5. Would you prefer the mh1 over the pfe112 or tdk ba200?

    ReplyDelete
    Replies
    1. I got the mh1, and I think I prefer it. Especially with a bass cut eq.

      Delete
  6. Wtf is this article? Just show the damn unaltered frequency response curve. Its a $30 earphone for god sakes. I THINK the gray curve in the second graph is the real FR but I'm guessing as your colors aren't labeled at all. Then later you say that pair was defective? I'll just take my chances and go with the Interunit variances graph.

    ReplyDelete
    Replies
    1. Lmao was thinking the exact same, and no it should actually be the second graph as it matches the frequency curve for this other analysis: https://www.innerfidelity.com/images/SonyMH1C.pdf

      Plus after testing many IEMs while having access to their FRQ graphs, the MH1C definitely sounds recessed in the lower treble regions (4-7Khz) while peaking in the very upper treble regions.

      Delete
    2. I prefer a dip at 4-7khz than a peak here. Peak in the upper frequency is not that anyoing as one below like many even high-end in-ears offer.

      Delete
    3. This comment has been removed by the author.

      Delete
    4. I also agree with Oluv, it is well known that our ears are most sensitive from around 2-5khz. I'm also a staunch believer of the Harman curve though i beleive a small peak at 9-10khz for headphones or a small peak at around 12khz for IEM adds sparkle and excitement in the treble. I also read an article from innerfidelity about effects of the 5k dip (essentially a trough from 3.5-8khz) which pretty much all headphones have. A commenter tested the effect of filling in this dip by using the hd600 headphone. Using EQ, the commenter added a 5db peak at 5khz and found it "add an uncomfortable edge and glare to everything". You can read the rest of the article written by Tyll from Inner-fidelity here: https://www.innerfidelity.com/content/compensation-curve-innerfidelity-measurements-dialog-part-1

      Delete
  7. This comment has been removed by the author.

    ReplyDelete
  8. This comment has been removed by the author.

    ReplyDelete
  9. I really like the Sony MH1 mid response, and now I'm looking for full over ear headphones. Can you suggest which headphones have mid response like MH1?

    ReplyDelete