[weirdek]

I'm running them through Xonar DG. I didn't mean the Pro 80 has literally no bass, I just didn't feel it had the oomph that the song asked for. Obviously the MH463 doesn't sound like a bass heavy headphone, it just delivers when it needs to. I should also mention that when the track does ask for some thumping and/or deep bass it doesn't even slightly overpower the rest of the spectrum which, for the price, is amazing. Like I said, it feels and sounds better in every way than the Pro 80 imo. Their open design also makes them feel more neutral and detailed, while still retaining a really fun sound signature. 

 

[IamPride]

I totally agree with you with the somic part. And even at $150 it's no brainer let alone $24 .

 

[aaDee]

How about these ones?? Somic Vienna V2, they look exactly same as MH463 but with an additional option of foam pads. But costs around $100.
 
http://www.aliexpress.com/item/Somic-V2-Professional-monitor-music-headset-hifi-subwoofer-Enhanced-Super-Bass-Noise-Isolating-DJ-headphone/32389823267.html

 

[Hal Rockwell]

  How about these ones?? Somic Vienna V2, they look exactly same as MH463 but with an additional option of foam pads. But costs around $100.
 
http://www.aliexpress.com/item/Somic-V2-Professional-monitor-music-headset-hifi-subwoofer-Enhanced-Super-Bass-Noise-Isolating-DJ-headphone/32389823267.html

 
There is also a V1 model which seems like the MM163 with MM185 pads.
If this is the only difference than there are a number of similar pads that can be purchased separately.
 
As for V2, it is stated that the driver is smaller than MH463 - 42mm vs 50mm, but the impedance is higher - 54Ohm vs 45 ohm. The sensitivity rose 2dB, 95dB instead of the 93dB in the MH463.

 

[JesvsR]

What do you think about the Somic MH539?? I've found it very cheap on Aliexpress... Does it have a good sound??

 

[Hal Rockwell]

  Ok, so i have had these MH463s with the turtle beach pads for a long minute and have some issues with their sonic presentation. The highs are pretty dull and lifeless as well as too much mid bass/ sub bass hump which interferes a lot with the rest of the sonic spectrum. The soundstage is large but not as large as i had hoped it to be from fully open back can but the instrument separation is really good on these. Not completely satisfied with what i was hearing i attempted to mod them in order to resolve all the problems stated above. **WARNING** Pics galore after the break!!!
 
HEADBAND MOD:
1st thing i noticed when i got the MH463s was that these cheapo Somics are pretty well built, great weight to them and they felt solid in your hands/ on your head. As a matter of fact toooo solid, this thing is pretty heavy when worn which causes some discomfort after some time.
.... i noticed that the headband area near the wing-plates could be pulled loose, so i took a peep under it and saw hard piece of rubber/ plastic inside. I unscrewed one side of the wing-plate on the headband and proceeded to pull the piece of rubber out. As you can see in the pic above the rubber piece seemingly is used to retain the shape/ hold the headband together. But to my surprise the headband did not fall apart after the rubber piece was pulled out. The weight of this rubber piece is pretty significant so no wonder these somics felt so heavy on the head. It feels a lot lighter after it is removed.
 
..... how the rubber headband piece looks like. My advice is to use some electrical tape/ masking tape, etc to wrap/ tape around the intersecting joints of the headband after you pull the rubber piece out. Would be best to just do one side 1st then go onto the other since you can lift/ pull the headband pleather back in order to get a better view/ handle. I didn't so it and the slider part of headband fell apart which took me some time to put back together. Not really that intricate but there is a small metal ball and a piece of tiny metal which contributes to the smooth sliding/ adjustment of the headband. If you lose those it's GG for you. LoL 
 
........I still have not taped the internal joints of the headband but its still in one piece and is holding together, it was a post thought after the headband mod and i am too lazy to go back and do it now esp. after i stuffed the thing. After the rubber piece is out, the upper area of the headband is really deflated, so i proceeded to stuff it with some cotton. It would make it easier if you take off both wing-plates so you can stuff it more evenly from both sides. This is why it is pretty important to tape the intersecting joints of the headband under the pleather just in case you need to man handle it while stuffing and the headband will not fall apart on you.  I stuffed it real good and lost count of how any cotton balls i used.
 
...... this is how the headband looks like after the cotton stuffing. These MH463s are a lot more comfy due to the reduced weight of the headband.
 
......have no idea why they used that heavy piece of rubber for the headband but i am glad it's gone, so even though it doesn't look as spiffy as it was originally they still look really professional.
 
to be continued.......................Next up!!! Sonic Mod.

 
A long long time ago, when this thread was just forming, @sfwalcer wrote this headband mod guide. I tried to follow it but wasn't pleased with the results mainly for aesthetic reasons. The mod did reduce the overall weight of the headphone and made it more comfortable due to the reduction in the clamping force, but made the headband look puffy weird. I decided to do it differently. I took the headband apart, got rid of the plastic parts that hold the pleather piece in place, got rid of the rubber piece, the pleather itself and the piece of sponge that is glued to the inner side of the headband's frame. I took a Bayerdynamic  DT770, 880, 990 style headband pad and two small plastic wire organizing stripes. The stripes are used to hold the plastic rails for the extension mechanism in place. There are two grooves on every pair of rails, the inner and the outer, that the screws of the plastic parts that held the pleather in place were. I tightened the strips in those grooves so they will stay firmly in place. The Bayer headband pad on the Somic's headband creates a shape that conforms easier with the form of my head and makes the headphone much more comfortable.
 



 
A word of advice: "If someone is going to try this mod, be careful cause there are some very small parts in the adjustment mechanism that can easily be lost".

 

[Cukedaddy]

End? Thats what we all say...lol.

 

[timiop2011]

I've just had my Somic MH463 professionally measured by Sonarworks. Their reference 3 plugin is amazing, and I fully recommend it! I have 20 days left on the 21 day free trial. After that I'll either buy the plugin, or use this CSV EQ file which I made for Voxengo CurveEQ based on their measurement data, which gets you 95%+ of the way to the same sound IMHO. http://www.filedropper.com/somicmh463flateq
 
This is just a rough initial attempt, I later plan to do seperate EQ CSV files for CurveEQ for both the Left and Right Channels
 
The data shows that the Somic MH463 is a bad headphone when used uncalibrated. Its FR curve is nowhere near flat, and there was up to 2dB differences between the left and right drivers, suggesting poor construction methods and poor QC. I was also told by Sonarworks of the MH463 that the metal mesh grill creates unwanted resonances at certain frequencies. Therefore I have removed the mesh grills, although I can not notice any audible difference at all in doing so.
 
Interestingly Joe Bloggs' suggested EQ for a neutral response was quite different to the measured Frequency Response og my MH463: Joe's EQ was a good effort, but as you can see he missed the main problem of the headphone, which is a huge peak around 3600 - 3800Hz
The other main problem is the relatively high harmonic distortion in the bass and subbass (see image at end of post)

 
note the 2dB difference between the left and right drivers in some places, such as 350Hz. In mastering terms 2dB  is huge!
 
I could be wrong but I think the main difference between the sonarworks plugin and an EQ plugin is the plugin not only attempts to flatten the percieved frequency response to dead flat, but also some of the harmonic distortion which was measured at 83 and 93dB. I read somewhere Sonarworks have spent $1M USD on developing it, so I'm sure there is a lot more going on that simple stereo EQ. However, my EQ file does gets very, very close if not identical. I recently read that a good midrange headphone should generally have THD @100db < 2.5% (and <1% for high end), but we can see that at 93dB harmonic distortion of my MH463. It is interesting how the plugin actually worsened THD below 50Hz at both SPLs, but it was made better above this at both SPLs (notice at 93 dB SPL how the plugin removes a 1.5% increase in THD which directly corresponds to the location of the "problem peak" at 3750Hz
 

 
If anyone uses voxengo curveEQ (or a similar EQ  which accepts CSV files in the same format), I would love to hear your thoughts on this EQ for the Somic MH463. Note that this is for the original MH463, not its V2 sucessor version, which I think has a different frequency response. I am quite pleased with the calibrated headphone, used with the corrective plugin or EQ.

 

[timiop2011]

I've superimposed the EQ curve onto the Percieved Acoustic Frequency Response. It is mostly symmetrical about the x axis, but there are some interesting deviations - circled in red.

 

[timiop2011]

This is the data for 0Hz to 10,000Hz which I will use to create a seperate CSV EQ file for each driver.+/-dB values are along the bottom, each level with either a line or an interval. Circled areas are places I need to be careful to include as they dont sit on a line or space, and deviate from an extrapolated trend line. Accurately EQing 10-20Khz will be much trickier, not least because of the lack of logarithmic interval lines (I have put on some small estimations of these above 10-20Khz which I might use). There is also much more misalignment of the percieved frequency response above 10k, compared to below (looks a lot like out of phase audio). 

 

[Joe Bloggs]

This is the data for 0Hz to 10,000Hz which I will use to create a seperate CSV EQ file for each driver.+/-dB values are along the bottom, each level with either a line or an interval. Circled areas are places I need to be careful to include as they dont sit on a line or space, and deviate from an extrapolated trend line. Accurately EQing 10-20Khz will be much trickier, not least because of the lack of logarithmic interval lines (I have put on some small estimations of these above 10-20Khz which I might use). There is also much more misalignment of the percieved frequency response above 10k, compared to below (looks a lot like out of phase audio). 

You seem more serious about equalization than most I've seen around here, so try this:
https://www.dropbox.com/s/bp4dwqfv1qtdx38/SineGen.zip?dl=0

Unzip this, run Sinegen.exe, try actually listening to your MH463 with your ears. (the Level sliders is the volume, pull it down to decrease volume. Start with a low level)

Now, all frequencies are not made equal for the human ear, so the goal is not to equalize until all frequencies sound as loud to you, (equal-loudness contours and all that) but the profile of those narrow-band peaks and troughs above 2kHz should not be much affected by this. Yet I'll be quite surprised if the peak and trough frequencies captured by the Sonarworks measurements are not off by at least 5-10% for each peak / trough frequency compared to what you actually hear with the headphones on. For the higher treble frequencies the errors should just about be enough to turn trough frequencies into peak frequencies and vice versa

So, I wouldn't spend too much time with Excel or CSV with that data, if I were you I'd sooner start actually tuning the phones for my ears and not some dummy ear's microphones

I must also ask if you sent the MH463 to Sonarworks with stock pads on, because you would have wasted a good opporunity to profile your earphones (for all the good that does) with some better / more comfortable pads

 

[timiop2011]

You seem more serious about equalization than most I've seen around here, so try this:
https://www.dropbox.com/s/bp4dwqfv1qtdx38/SineGen.zip?dl=0

Unzip this, run Sinegen.exe, try actually listening to your MH463 with your ears. (the Level sliders is the volume, pull it down to decrease volume. Start with a low level)

Now, all frequencies are not made equal for the human ear, so the goal is not to equalize until all frequencies sound as loud to you, (equal-loudness contours and all that) but the profile of those narrow-band peaks and troughs above 2kHz should not be much affected by this. Yet I'll be quite surprised if the peak and trough frequencies captured by the Sonarworks measurements are not off by at least 5-10% for each peak / trough frequency compared to what you actually hear with the headphones on. For the higher treble frequencies the errors should just about be enough to turn trough frequencies into peak frequencies and vice versa

So, I wouldn't spend too much time with Excel or CSV with that data, if I were you

I'd sooner start actually tuning the phones for my ears and not some dummy ear's microphones

I must also ask if you sent the MH463 to Sonarworks with stock pads on, because you would have wasted a good opporunity to profile your earphones (for all the good that does) with some better / more comfortable pads

To answer your questions:
1) Regarding the Pads, I tried a few different pads out. The original red backed Somic pads were uncomfortably soft (as though they had not put enough padding in or something), so i binned them. My Sennheiser HD439 velour pads are very comfortable on them but unfortunately didn't quite fit the size of the headphone. I thought about glueing them on, which would have worked but I need them for the HD439s too so I didn't. In the end I settled on sending them off to sonarworks with faux leather ATH-M50 pads. 
 
2) I am not using Excel, I am using notepad to create CSV files based on the Sonarworks data. It is much quicker. And then I am loading the CSV file in Voxengo CurveEQ
 
3) I know that due to psychoacoustic effects not all frequencies are equally loud the human ear, and measuring headphones can also be made difficult due to the proximity effect (as the microphone inevitable has to be very close to the driver). The reason I went with sonarworks is because they use 

Perceived acoustic power frequency response ​

(

A patent pending acoustic measurement system that measures sound the way humans perceive it). They have put A LOT of research money into developing that system. Their software is also able to somehow magically correct small harmonic distortions at some frequencies. Simple EQing with a sweeping sine wave by ear can not achieve this​

 

[timiop2011]

... and on that note I regret not sending one of my other headphones off to Sonarworks, especially my HD439 which Sennheiser claim has <0.1% THD. If that figure is true, that is almost on par with the best headphones in the world for THD%. If I knew that THD was so much more "uncorrectable" than frequency response by the plugin, I would definitely have sent off my Sennheiser HD439 to be measured instead (Although I do still have 20 days left on the free plugin trial to match my Sennheiser HD439's and Samson RH600's to the plugin-calibrated Somic MH463's )
 
Dont get me wrong. I'm very happy with the calibrated sound of the MH463, which are very poor when used uncallibrated. Maybe when we are looking at data for which headphones to buy we should be paying more attention to THD and less to the frequency response curve.
 
P.S. Where Sonarworks calibrated Somic MH463 headphones could be a real bargain is for anyone who mixes on Yamaha NS10 monitors and wants an inexpensive portable solution for mixing away from a studio that translates very accurately. This is because the NS10 has very little bass response below 100Hz so it totally removes the problem area of the MH463. To achieve this all they would need to do is open the Sonarworks Reference 3 Plugin and use the "Japanese White Cone Studio Monitors" Preset in the Speakers section in Simulation Mode.
 
There are also some excellent headphone simulations within the reference 3 Sonarworks plugin based on various headphones; Beats Solo HDs (named "Popular Consumer Headphones" Preset), Sennheiser HD650s (named "German Studio Headphone Simulation"), a Grado GS1000i simulation (named "Popular Hifi Headphone Model Simulation") and an AKG K712 simulation ("Studio Headphone with Austrian Heritage"). The AKG K712 preset sounds particularly great to me (for listening purposes as opposed to mixing purposes, where I would always use calibration (flat FR) mode, not simulation mode)

 

[timiop2011]

 

[Joe Bloggs]

Well, you having actually changed pads before sending them to Sonarworks, the FR profile you obtained is different from what I heard and posted using the stock pads. No surprises there.

The wiggle in the THD graph where the EQed THD is lower than original THD is 3xxxHz, corresponding to the frequency that the their dummy ear heard to be 10dB louder and EQed down accordingly. Looks like they're reporting THD based on setting (say) 1kHz to be 93dB and running a constant volume sine sweep through your headphones, as opposed to adjusting the volume of the sine sweep until all frequencies are actually measured 93dB loud by the dummy ear. For the latter practice (as seen e.g. in Innerfidelity) it's the weak parts of the frequency response (e.g. 7000Hz on your phones) that would tend to get a blip in the THD. And in that case there would be no point measuring THD with and without EQ, since the effect of EQ on THD is zero (unless there are calculation errors in the EQ, in which case THD would increase)

Have you actually tried Sinegen? It doesn't generate sine sweeps (which you'd of course have a hard time following in real time). It's a real time variable frequency sine wave generator. So what I'm asking of you is simple. From your Sonarworks graph I see peaks at roughly 3700Hz, 5850Hz, and (say) 8400 and 10400Hz, troughs at roughly 5050, 6950 and 9300Hz. Now fire up Sinegen and set the frequency slider to these frequencies--and (careful with the volume when you start) are 3700, 5850, 8400 and 10400Hz actually louder than the frequencies above and below as you wiggle the frequency slider? (You can use the up and down arrow keys as well as the mouse on the frequency slider, for slow precise movement of frequency) Are 5050, 6950 and 9300Hz actually quieter than their adjacent frequencies? Or are the actual peak and trough frequencies subly (or completely) off the mark? It doesn't get much more "perceived acoustic power frequency response" than that.