You've tried it with the 590, and it's the same? If true, then sounds like a bad bios flash, or may be settings in bios. Check to make sure all bios settings are correct.
Or are you saying you updated to new drivers for card? If that's the case then go into safe mode, remove it in programs, restart, custom, clean install the last known good driver.
It's possible you have a loose connector somewhere, check all connections, pull 24 pin from mobo, check for overheated connections. Could be psu, but likely is bios issue, what does bios say for voltages?

I flashed the new BIOS successfully and booted while I still had the 590 and 260 installed. It passed POST and started windows same as always with the new BIOS so I'm sure the BIOS flash worked. When I booted, I used driver sweeper to remove nvidia drivers so when I shut down and installed the new cards I could boot up and install the new drivers.
 
Even if the BIOS flash had failed, the Rampage IV has an on-board button for BIOS swapping which I also tried and even with the swapped BIOS the computer did not start. It doesn't reach POST before dying. I can't even get an error code off the board to see if it could be anything else. I really think it's the power supply, but I appreciate any other suggestions.
 
EDIT: Yes, reset/cleared BIOS.

Just for fun try using the 500 with one 670 and all extras disconnected. It won't be enough to game, but should let you know if psu is the problem.

I just tried using the 500W PSU with only the motherboard, CPU, CPU waterblock+fan, and one GTX 670 attached to see if I could reach BIOS.
 
It started for a second then died, then started, then died, in an endless loop. It was doing this once when I was trying to get it to work with the other PSU. I have no idea what's going on now. The only other time I've encountered immediate shut-off was with a poorly-seated CPU cooler, but I didn't even bump the cooler and it seems just as firmly in place as before.
 
Tried the 260 as well. In slot one and in slot three. Same thing happens, endless shut-off loop. I think I heard a pop this time. There could be a short somewhere, but with only a few things plugged-in it must be in the mobo itself. The fan on the 670 spins when it fails to start up.
 
EDIT: The pop was probably from one really loose fan connector, at least I hope that's what it was.

Even if I started the build process completely over, I don't have any thermal paste to reseat the CPU cooler. All I did was switch out some GPUs. I used the same two black 6-pin power connectors that were already in place having been used with the 260. I took out the two 8-pin power connectors that were used with the 590 and exchanged them for two 6-pin connectors for the second 670. Throughout this process I didn't touch much of the board, just the right edge while routing cables behind it and of course the PCI-E slots. I didn't think I did enough to cause this. Yet, once everything was in place, the two 670s, the SLI bridge, power connectors plugged in, I started the computer and it shut off immediately. Only through all this troubleshooting to solve that initial problem did I do anything invasive. With the BIOS working before I wiped the video drivers I don't thing that's to blame and I even used the BIOS swap button on the board to use a different BIOS. Since the mobo refuses to start with a different PSU and GPU, signs now point to the mobo being at fault rather than the PSU.
 
The only thing left I can think of trying is to plug the two fans for the CPU rad directly into the cooler. I have them hooked up with molex because they are LED fans and aren't designed to be throttled as the H100 does. Either that, or plug the pump directly to the PSU with molex connectors. I had to do that with an older build. It wouldn't start if the pump was plugged into the board because the delay in the pump's startup was too great. Maybe the updated BIOS caused the board to start reacting this way. I'm not sure how it managed to start up with the 590 and 260 before, but fiddling with the CPU cooler is the only thing left I can think of doing.

I've been removing the power and reset pins since post one. Tried removing the battery earlier, but I didn't have it out for 2 minutes.
 
I just fiddled with the CPU cooler, changed the fans back to the ones it came with and plugged them into the pump. Still the same result with the Coolmax PSU (immediate shutdown). I had a non-sleeved 8-pin extender for the CPU power connector that I also removed and plugged the CPU power connector directly into the board without the extender in case it was the cause, but that didn't change anything either. Might contact Asus and troubleshoot with them.

Thanks for taking your time to brainstorm with me a while even if I didn't manage to fix it.
 
My cheap PSU has worked well over many years, but I lack the PCI-E connectors to expand beyond two graphics cards. I might get a new one as I've been putting it off. After all the testing it doesn't appear to be the problem though.
 
I've sent a message to Asus. If they can't troubleshoot me out of this mess then they'll just have to replace or repair the board. Pretty sure it's covered by warranty, but I'm going to need thermal paste if or when I have to remove my CPU to mail them the mobo and then reseat it again when the board is returned. What a pain, at least I have a pretty good laptop to use in the meantime.

I had a similar problem once installing a new GPU.  The mounting of the motherboard was loose in one spot so when I pressed down on the GPU to get it in the slot, something flexed and cracked.  A new motherboard fixed it right up.
 
Good luck
ps. get yourself some Thermal paste to keep on hand.