CHANGSTAR: Audiophile Headphone Reviews and Early 90s Style BBS
Lobby => Headphone Measurements => Topic started by: Marvey on June 23, 2013, 07:03:08 PM
-
Inspired by LFF's dummy binaural head and the pinna/ear coupler thingie Tyll sent me a few years ago, I decided to try out a alternative measurement method. The measurements you see here should not be compared to the other measurements on this site. (BTW, the "sponges" you see on HF, posted of course by an un-authoritative source who has never spoken to me about my rig, is out-of-date misinformation. The "sponges" are only reflective of how I measure the CSDs for higher frequencies. For all other measurements, they are not used, or at least used exclusively.)
More info later, like in a week. I won't be posting much for a while (busy at RL work.) Still tweaking things like surface coatings to mimic human skin flesh or pork belly which seems to come close to it. For obvious reasons, I can't make a coupler or line a dummy head with pork belly.
The harmonic distortion measurements are based on sweeps, which is different from the other distortion measurements taken so far. SPL was normalized at 83db SPL @ 1kHz for each headphone.
P.S. Added T50RP stock.
-
This is interesting, shows the difference in upper mids between the HD600 and the Paradox. Also the difference in 15-20kHz.
Any chance you might be able to send me one of these for the K550 I sent down? Only if it doesn't take you much more time...thanks for another look at these phones with a different method!
-
Yes I will. Not this coming week though.
-
Kool! Results don't look that different from your previous results IMO. Interesting to see the differences between the Paradox and Mad Dogs. Correlates well with impressions about the tuning of the Mad Dogs being a bit stronger in the bass (not a bad phone though), and the Paradox more balanced. Very Nice!!!
:money:
EDIT: For the distortion numbers, are you using 90 dB Marv?
-
Excellent ! :)p1
-
Purrin.. would you mind if I linked to this in my "Paradox tour" thread over at HeadFi? Reason I ask is if perhaps, it's a WIP-type thing..
-
That's fine. This info is supposed to be public even if it's still WIP. The only disclaimer is that I don't do science, nor do I purport this to be science, nor do I feel that my measurements are more accurate or authoritative than others. This is just a hobby / fun thing for me until I decide to sell a Changstar (TM) headphone correction gizmo.
-
EDIT: For the distortion numbers, are you using 90 dB Marv?
In this case, I took the the FR measurements and adjusted volume so the 1kHz @ ~87db. +/- 1 db.
The FR and HD are derived at the same time using a swept sine with a sequence length of 256k and sampling rate of 48000. I run the test two or three times and get an average. (I should really run them 5-10 times in the middle of the night.) I have no idea how this method works mathematically since I haven't studied the transforms in depth.
FYI, the Paradox and MadDogs are less efficient, especially in the bass and midrange areas than the stock T50RP. This shouldn't be too hard to fathom as its easier to reduce energy than to add it. The MadDogs also seem to take extra step to reduce mid-treble energy. Overall, the Paradox is the least efficient, but not by much compared to the MadDogs.
P.S. It would be correct to say that the Mad Dogs sound the most laid back of the bunch.
-
Nice work! There are indeed a few differences from the other method but the basic shape remains the same. Looking forward to more insights!
-
Very cool and interesting!
If this is the Mad Dog I loaned to Marv, that should be the 3.0 revision (latest as of this post is 3.2). If not, then nevermind. There are more details on that in the MD measurement thread.
When I had my Paradox (No. 18) and MD 2.0 (with older dog pads) on hand at the same time, the Paradox almost certainly had a more emphasized bass response. I'm thinking it was around 80-300Hz or so. I don't really remember how the treble compared. I didn't have a Paradox to compare with the newest Mad Dog pads.
Just my bit of experience with both of them...
-
These meaaurements look accurate. Perhaps slightly more than past ones.
Definitely Harmonic Distortion is easier to interpret.
However, consider aligning 0 dB with 1000 or 500 Hz level and providing a magnified view of the FR.
This means switching the reference to about sqrt(2) V/Pa
Or 0.707 Pa/V.
-
The measurements make t50rp look like they dont need much modding.
-
Those aren't CSDs which pinpoint the T50rp flaw: resonant ear cups.
-
Are you serious? Go back and look at the stock T50rp CSD plots again.
-
it does look pretty odd around 1k and so, but haven't heard t50rp stock so I can't comment on that.
-
I have, it sounds bit resonant and somewhat "plasticky" in nature. I bet it's the wide but short 10k ringing in action combined with bit too long mids decay.
Paradox is a major improvement on this front - if not any others unfortunately - in the basic version.
-
The D4 on the HE500 here looks kinda :gross:. It seems more normal in the Abyss review. Any idea what would cause this discrepancy?