[ASDFer]

PLEASE SEE REVISED MEASUREMENTS AT END
Alright, so I finally built up a little rig to test players under load (it doubles as a ER-4p-> ER-4s or whatever else I want). The Karma suffers on the lowend when low impedance headphones are used. Here are the rightmark results for the comparison:

RMAA 5.1 Rio Karma v. iHP-120

Testing procedure: player into jumpered test rig (straight through to headphones). Test end into ER-4P and other end into sound card. The sound card was set at its maximum input level and the volume adjusted on each device to get the same output response. This meant ~26/40 on the iRiver and ~22/30 for the Karma.

I then added a 100 ohm resistor in series with each channel (only into the headphones). The volume levels stayed the same (low output impedance?)

The results look much better for the Karma now:

100 ohm resistor added

So, if you're a bass-head with a low impedance headphone, you might think twice about getting a Karma. Otherwise, it's still a great unit IMO.

A few notes: loopback was tested without load. EQs were off. If I think of anything else I'll add it later. It's late.

 

[penbat]

So would you say that Shure E5Cs probably work well with a Karma ?

E5Cs Sensitivity (at 1kHz): 122dB SPL/mW
E5Cs Impedance (at 1kHz): 110 ohms

They have high impedance but also high sensitivity.

 

[shibby]

what about the senn 280 PRO.

which of the two would drive tham better?

 

[ASDFer]

It looks like the Karma uses a 220uF coupling capacitor, while the iRiver perhaps forgoes the coupling C? (I calculate that it should be around 800uF).

The Karma should work fine with the e5c, but it will probably be quiet for some tastes. With my er-4s+ (127 ohm) setup it has plenty of power for me. The bass definitely sounds better with the inline resistors - it sounds _slightly_ flabby without.

As far as the 280s go, I think either will be fine since the impedance is 64 ohms. I calculated about a 1.4dB suppression for the Karma at 20Hz, something you will not likely notice. For reference, the iRiver may have about .2dB suppression at that frequency.

EDIT: it actually seems the Karma has 330uF coupling cap...

 

[dj_digital]

So do those tests indicate that the ihp has theoretically better sound than the Karma? I'm not too familiar with rmaa so could someone explain these results a little?

Thanks

dj_digital

 

[penbat]

Shure E5Cs give very high volume output. much more than Etymotic Er-4Ps.

 

[ASDFer]

Quote:

Originally posted by penbat Shure E5Cs give very high volume output. much more than Etymotic Er-4Ps.

Well that's great, I'd say they should work well then and you've answered your own question.

 

[ASDFer]

Quote:

Originally posted by dj_digital So do those tests indicate that the ihp has theoretically better sound than the Karma? I'm not too familiar with rmaa so could someone explain these results a little? Thanks dj_digital

There isn't really a huge difference between the players. The frequency response just shows that there is some roll-off (reduced response) below 100Hz for the Karma when driving a low impedance (the output capacitor and and resistance of the headphone act as a high pass filter). The Karma also has slightly higher distortion numbers (so a pure sine wave sound file won't be exactly a sine wave...) The other numbers are roughly comparable - both devices have a lot of stereo crosstalk.

 

[CrawlingEye]

One thing to ask, that I can't clearly see in the screenshot is this:

Is the 1KHz bump accredited to your soundcard or the Karma?

It seems a lot of RMAA screenies seem to have a bump around 1KHz. I'm trying to create a good EQ setting for frequency correction.

So far, I have 40Hz bumped up about 3dB with a Q of about 2, as it doesn't seem like an extremely high Q is neccessary considering the natural roll off.

It sounds a bit cleaner with 1KHz cut about 6dB but I'm not so certain if this is just the fact that sometimes a chunk out of the mids can give this effect falsely.

Great thread and great information though.

 

[ASDFer]

Which 1kHz bump are you refering to (which graph?) In the dynamic range and THD tests, this is the test signal that is used to calculate the reported values. As far as the bump at 2.5 kHz in the low impedance frequency response test goes, it is caused by a bump in the impedance curve for the headphones. It is not an intrinsic property of the player.

Here is the impedance measurement for the headphones (done on a HP 4294a)

 

[CrawlingEye]

I was speaking in terms of dynamic range.

 

[ASDFer]

I just realized that having a frequency dependant load might be a bit confusing (the response curve plots voltage...) so I've redone a measurement with a constant 51 ohm load:

51 ohm test

As you can see, it no longer looks like there is increased high frequency response (it's actually reduced for the ER-4p, since the power goes as V^2/Z)

 

[CrawlingEye]

Ok, thanks. After investigating a bit more (and thinking clearly after some much needed sleep last night) I've changed it to just a simple 1dB increase at 40Hz it seems unnoticable at first (especially on the stock mx300's) but long-term, even on the mx300's I can notice a difference. It makes the rhythm section ever so slightly more pronounced, which in my opinion brings it back on level with the rest of the instruments. I use a Q of 4 for the 40Hz boost.