Suggestions/info:
- What are the S-spec numbers of your processor pairs?
- Did you get the current bios version from the seller (1006.002 is the latest)
- What memory speed (PC3200 / PC2700) are you testing with
- Which graphics card is pictured

- Try a power supply with a single 12V rail of at least 25A
- Stick to single channel (1 DIMM) memory for now
- Could be bad caps - have you checked

Thanks for the sugestions!

- The 3000DP-pair is a SL7HG. I just tried with a 3066DP SML6GH which I hoarded from a Dell, but the same result: nuts.
- No idea which BIOS is on, never have seen it boot.
- I've tested different kinds and brands of DDR, in single and double config, but nothing gets me a boot screen.
- The one on the picture is a 9700Pro. I triued PCI-based cards, yet again, no difference.

- The power supply pictured gives 17A per 12V, I'll try another one tonight or tomorrow. Yet it should be sufficient to power a single Xeon, I imagine.
- Caps seem ok, I don't have the tools to check them other than visually.

The NB seems to be heating up when the system is on. As the green light does indicate it has had the capability of checking the bare minimum (CPU present or not), something is at least doing something, which is a start.

I switched the power supply from the Antec 450 to a Corsair 650TX with 53A on the +12V line, again, no difference.

I am afraid I was sold a dead mainboard. Might my last hope be a new BIOS chip?

Update: took it back from storage and had another go, with the bare minimum (1 cpu, 1 stick of ram). Still no response from the dead.

I've removed the BIOS-chip, and the behaviour of the board is the same as with it installed. Could this point to a shot BIOS?

I have a similar board asus NCT-D, unfortunatly can't test it since I am unwilling to take the cpus of a working setup to test it.

They are quite cheap if you have the heatsinks

I have a pair of 3.6 GHz Irwindale CPUs in my ASUS NCT-D with BIOS 1009 09/27/2006 running Windows 7 x64 Ultimate V6.01 Build 7601 Service Pack 1

red-ray wrote on 2021-03-27, 19:21:

They are quite cheap if you have the heatsinks

I have a pair of 3.6 GHz Irwindale CPUs in my ASUS NCT-D with BIOS 1009 09/27/2006 running Windows 7 x64 Ultimate V6.01 Build 7601 Service Pack 1

Too bad I don't have any spare heatsinks 😀

I've ordered a new bios chip on Ebay, I'll report on the findings here. I've spent last night hotswapping bios chips and trying various bios utilities to flash the Asus rom onto on of the chips I had available, but to no avail.

It did work for a few other ASUS Bios roms I had (P5PE-VM, P4RD1) but not on the NCCH-DL. At least I have some experience in hotswapping chips, so all's not in vain.

Although I've now acquired one of these boards since my original suggestions (seems close to yours production-wise based on barcodes...mine is 4BZM170875) , I'm afraid I can't offer much else in the way of things to try.

- I tend to associate the long repeating beep tone with hardware incompatibility, but as you've swapped over all the major components that seems unlikely
- Is the heatsink secured properly with sufficient surface-to-surface loading (CEK springs in place) as these cpus seem to overheat rapidly without it
- do you have a case to properly mount it in with the correct Xeon mb tray standoffs
- have you tried a fresh coin cell battery

PC Hoarder Patrol wrote on 2021-03-28, 06:58:

- Is the heatsink secured properly with sufficient surface-to-surface loading (CEK springs in place) as these cpus seem to overheat rapidly without it
- do you have a case to properly mount it in with the correct Xeon mb tray standoffs
- have you tried a fresh coin cell battery

Much obliged for the suggestions!

The system is currently not in a case, and as such, the copper heatsink sits on the cpu solely thanks to the law of gravity. I'd suppose the system might at least show some sign of booting before the thermals kick in. If I'm able to revive it, the system will indeed find itself in a proper housing.
The battery is a new one I use for testing purposes.

I'll keep my fingers crossed that the bios replacement chip will have this board magically spring to life 😀.

flupke11 wrote on 2020-01-03, 08:10:

The power supply pictured gives 17A

You are pushing your luck, my Dual Intel Xeon (Irwindale) system maxes out at 22.750 amps

Thx, I'll keep that in mind when it's revived and ready to be put to use.

The current test setup should be sufficient with just one CPU.

No luck with the new BIOS chip. The fans spin when I push the power button, but nothing else.
I do not want to give up on this main board, but I am clueless about what else to do to get it fixed.

Shame about the BIOS, but it was certainly woth a try.

Done a bit more reading on this, and despite not seeming to have a history of board level component failure, sadly it's beginning to look that way.

As a matter of interest do both VRMs have the big heatsink fitted as apparently some 1st release boards (serial nos beginning 4B & earlier) only came with one, although mine has both. Also, the original TIM on these VRMs was a strip of (blue) thermal pad which some people tried to replace with varying +ve / -ve results. Overall, maybe worth removing the heatsinks / pads to check for visible MOSFET failure.

Going back to the question of heatsink fitting / loading - you're probably right re gravity for a quick power-on check but I'll give this actual example of why a proper chassis install might be worth it...

...I have a used dual skt 771 board (next gen up from the 604s & using basically the same coolers and mounting system) which will not POST if I run it outside proper case mounting, but never fails to post when mounted - same symptoms as you of fans spinning, blank screen & long beeps. My guess is that over time with these heavy / high load heatsinks fitted the board has flexed to the point that electrical contact in and around the sockets only works with a similar load in place (may be pins or solder at work - not sure).

Last thing I can offer is as clear a pic as I can get of my board that you can compare to yours.

For info, I'll be using the ASUS X-PAD mounts on mine rather than the more common CEK springs.

Thanks for your help!

I've taken photographs of the mainboard with the two heatsinks removed. I'll PM the link.

With the CEK springs and one cpu+cooler properly installed, there's still no post. I'll leave it be and turn my attention to a Dell and an HP workstation from the same era.

PC Hoarder Patrol wrote on 2021-04-01, 05:39:

I have a used dual skt 771 board (next gen up from the 604s & using basically the same coolers and mounting system) which will not POST if I run it outside proper case mounting, but never fails to post when mounted

I suspect poor contact is more likely to be an issue with Socket 771 rather than Socket 604, further my Dual Socket 771 (Intel D5400XS Skulltrail) posted all the time when I had it sitting on my desk last month. I am also pretty sure my Dual Socket 604 (ASUS NCT-D) was also OK many years ago before I put it in a case.

Yeah, was meaning this is an issue I have with this one particular socket 771 board running outside its case rather than in general for this socket or for any other board with similar chassis-mounted cooling.