Sounds like you want a warmer tone than what the 600 can give? Phillips X1 might do the trick, just based on what others have written about it.
How do you find the final balance on the other 2 headphones?

Just got my HD600 in, and some of you know I thought they were a bit hot around 3-4KHz. I came back after several hours and some sleep and found this to be less of a problem. Could be the fact that I stretched them out while I was asleep, could be the QRV-08 amp sounded best after being left on for a while, or it could be that I was just tired when I tried them the first time.

I did still find something about them to be a bit edgy the second time around, though trying them through the Vali put some ease back into them. Measurements didn't really show anything really wrong. A nice, balanced response. No major peaks or humps. Very little ringing. Not quite sure what I was hearing at this point! I'd say that broad 2-4KHz hump could explain what slightly bothered me about them, but it's still so mild that I can't say for sure.

Either way, I had a couple open-cell, foam discs already cut and sitting on my desk from a different project, so I figured it would not hurt to give it a shot and apply them in front of the driver. I put them on the front side of the ear pads and stuffed the edges under the ear pads. No disassembly required! About as quick and easy (and brainless) as it gets. This is the material I used, I believe: http://www.foambymail.com/CF0-125LF/1-8-inch-charcoal-regular-foam.html  If not that, I did get the material from there.

Subjectively, I found this tweak made them a touch warmer and smoother sounding. Definitely more relaxed, though I would not be surprised if more people than not would prefer the slightly edgier quality of stock. Very easy for me to listen to, now. On the other hand, this did ever so slightly reduce their already not-the-best resolving abilities, and they did lose some air. I think it's a good trade off for someone like me, and definitely worth the whole 30 seconds it took me to try it out. Comfort could be improved by installing the disc under the ear pads.

Measurements do indicate a bassier/warmer tilt to the response. 2-4KHz rise is slightly reduced. 6-11KHz area is lowered a bit but also smoother. Upper frequencies roll off more readily than before. Interesting! I'm happy with it, but you may not be. But it does go to show one may be able to slightly tweak the HD600's performance closer to their tastes with similar, easy mods.

I want to note that I made some fairly large changes to my compensation curve. I'm more or less using the HD600 as my reference point for measurements at this point, given the wide availability of measurements from a variety of sources and the fairly decent level of similarities among those. The response below 100Hz has received a sizable boost to compensate for my specific hardware. I'm also applying much less compensation to the area above 10KHz than I was before. Essentially, take my Oppo PM-1 and JVC HA-RX700 measurements, and boost the bass response below 100Hz on those (ex: 20Hz boosted by nearly 5-7dB, rest of curve follows a similar shape). The results above 100Hz should basically match the response above 100Hz on the raw results. That will give you a good idea what changes I made to the compensation curve. I had planned on using the HD600 to set my curve in stone, or closer to it, and it showed I had some changes to make. Should be good now...*crosses fingers*

Full and robust set of measurements to come in the next couple days, likely (stock, though might post a couple more basic, mod/experimental measurements as well). Also more subjective thoughts.

Bottom two pictures are with the foam disc.

Cool, any chance you can confirm exactly what the damping material is? Hot upper mids or low treble I think was my main problem with the 600s.

I currently have HD650 drivers and grills installed on my HD580 frame. I was thinking of getting an extra pair of earpads for the 580/600 drivers so I can switch them quickly. The Senn pads are $60. There are 3rd-party earpads from China on eBay for $30. Has anyone tried these? Are they any good?

Riker, I'm almost 100% sure it's the material I linked to in my previous post. In other past situations, I've also used spare ear-pad material, for clip-on HPs like the KSC75 and such, with decent results. I find the type of foam used for these ear pads to have less of an overall effect on the sound that what I experimented with on the HD600, granted this is what I was looking for with the HD600. Sometimes ear pads for "regular" head phones use inner fabric/mesh lining material that works well for front damping as well (i.e. 840 pads, Monoprice 8323 pads both have interesting pad lining material). Just throwing out some ideas, but check out that link from earlier for this particular material.

It is basically the case now that I'm only compensating for the bass roll-off my hardware exhibits. The rest of the hardware measured within +/-0.06dB of flat (excluding mic), which I am technically compensating for as well. I thought calibrating the comp. curve using HW measurements gave exaggerated bass in the past, but since I wanted to use the HD600 as a reference point to better match other measurements on the interwebz, that showed me otherwise. I think the mic I'm using has a known elevation in the upper treble response, but given I couldn't find a consistent source of measurements to compensate the mic, I'm just not going to worry about it. Worst case scenario, take off 1-2dB above the 9KHz line (roughly peaking at ~2.5dB at 15Khz, IIRC). There are also cases where I thought past measurements exaggerated the 2-5KHz area, but since that was not something I could consistently repeat, I gave up on compensating there as well. Now that I have the HD600 measurements looking as they should, my future measurements should stay consistent with this.

Oh, and those measurements were done at 90dB@1KHz like I've done with my others. The overall levels are low because I turned down the mic-in volume setting to avoid some HD artifacts I was getting in other measurements (CSV export weirdness is still a different issue).