TR Forums

Hey all, having some problems here and looking for some advice. I have hell getting my new x299 system to work. I have tried 2 motherboards, 2 processors, 3 power supplies, swapped DIMMs, and i get nothing.

On one board, the fans don't spin up.

On the other, the system power cycles, clicking on and off continuously.

I even tried with no GPU to see if it would behave differently, beep or something, but nope exactly the same. I tried inside my case and running the bare board to eliminate the possibility of a short. All parts (with the exception of CPU and board) are tested working in other systems. I am at my wits end here, and considering whether the whole platform is a lemon and going for TR (which feels irrational to me given plenty of people have set theirs up successfully). Anyone have any thoughts?

It might help if you posted the specific models of motherboards, processors, power supplies, memory, etc. that you are using.

On top of that, if your motherboard is capable of emitting an error code that information could be helpful too in conjunction with the documentation.

Also, you might not have one available but do you have a photograph of the motherboard with everything installed on it? That can sometimes be helpful in doing a sanity check although sometimes you have situations where a plug/DRAM module/etc. looks like it is plugged in but isn't actually seated quite right.

motherboards: gigabyte auorus & asus rampage VI extreme

Powers upplies: evga 1000 and thermaltake 850

RAM: corsair 4266 4 dimms (tried swapping all of them) and even tried some 3600 ram from another system just for good measure

CPUs - i9 7900 & 7800

Trust me everything was perfectly seated, I tried every permutation I could think of, the boards never got far enough to emit an error code. On the asus the system fans didn't even spin up.

How are you attempting to power up the boards? If you're using the same case switches for all scenarios, you might have a bad switch that's sticking.

ludi wrote:

How are you attempting to power up the boards? If you're using the same case switches for all scenarios, you might have a bad switch that's sticking.

Good point. Do these boards have the onboard power switch that you could use to bypass the standard case switch?

Whatever is happening here appears to be extremely low-level in nature if you can't even get the motherboard powered to the point where it will throw an error code.

One more idea: Take out the CPU and see if the machine can even get the motherboard to power on enough to throw an error code telling you that the CPU is missing.

Using the on switch on the boards, stripped them down to barebones at this point just trying to get a sign of life.

Haven’t tried with no CPUs yet, what’s flummoxing me is the fact that no combination of what I choose gets anything. Is it possible a bad board could have damaged both CPUs?

leor wrote:

Using the on switch on the boards, stripped them down to barebones at this point just trying to get a sign of life.

Haven’t tried with no CPUs yet, what’s flummoxing me is the fact that no combination of what I choose gets anything. Is it possible a bad board could have damaged both CPUs?

I'm not going to say it's impossible but it's relatively unlikely that even a bad motherboard will fry a CPU.
I'll plead guilty to frying a motherboard once upon a time but the CPU that was in it emerged unscathed from the debacle.

At this point if things seem really REALLY dead then I'd chalk it up to the motherboards or some interaction between the PSU and the motherboards. Regarding the PSUs, you say they work well in other systems but do those other systems also use two (instead of just one) EPS connectors for the CPU socket? From what I could gather of your listed motherboards they both use two EPS connectors and that's a little less common, so make sure both of the EPS plugs are actually working just as a sanity check.

This kind of sounds like a case where the board doesn't recognize the CPU and needs a BIOS update before it can even POST or give error codes...but that's not very likely given the boards and CPUs in question. They ought to be designed for each other.

Check for damaged pins in the socket.

Make sure you're not excessively tightening the bolts of the HSF or waterblock. Cooling solutions that are too tightly mounted can result in all sorts of weird problems from dead memory channels to failure to POST.

These were both common failure points on X99.

Chuckula's last suggestion warrants checking. Is it possible that the second 8-pin CPU (EPS) connector is plugged into a PCIe PSU port instead of a CPU port (on the PSU side, not the mobo side; if the PSU is modular)?

Ryu Connor wrote:

Check for damaged pins in the socket.

Make sure you're not excessively tightening the bolts of the HSF or waterblock. Cooling solutions that are too tightly mounted can result in all sorts of weird problems from dead memory channels to failure to POST.

These were both common failure points on X99.

checked for bent pins, didn't see any, I didn't tighten the HSF too much because of the constant hardware swapping I was doing, no sense in applying max torque if I'm gonna take the ting apart 5 mins later.

Also the 8 pin connectors work fine on my other board. I just ordered another PSU and a new UPS, if that doesn't work I have to give up, I mean what more can I do?

leor wrote:

Ryu Connor wrote:

I just ordered another PSU and a new UPS, if that doesn't work I have to give up, I mean what more can I do?

At that point, RMA both motherboards, if they're still within warranty. Odds of both being bad are one in a million but the odds of the Intel CPUs being toasted are even lower.

Reaching here, but have you got the system in the case or on a test bench? If in the case, it's possible there could be a grounding issue or short with the case itself.

Ran into a similar problem on my last build. New processor/board/RAM, system would not POST, no display, and the diagnostic LEDs suggested either CPU or RAM. And it was way past the return period. Reseated everything, tried RAM in every slot, moved things around, etc. Nothing works. Then the epiphany. I had to reset the BIOS. That took care of the issue. Since you didn't mention doing that, I'm throwing this out there.

toastie wrote:

Ran into a similar problem on my last build. New processor/board/RAM, system would not POST, no display, and the diagnostic LEDs suggested either CPU or RAM. And it was way past the return period. Reseated everything, tried RAM in every slot, moved things around, etc. Nothing works. Then the epiphany. I had to reset the BIOS. That took care of the issue. Since you didn't mention doing that, I'm throwing this out there.

I'll second this because my brand-new Z270 board wouldn't POST with my 7700K, also new.

I cleared the BIOS (after a bit of panic, I admit) and away it went.

I wouldn't have thought doing it straight from the factory would be necessary, but lesson learned!

Same thing happened here. Prime X299 Deluxe. Did not post by any means. Bios was reset and tadaaa. It's gone..

Can you get the CPU RMA ? You are not new to system building hence i don't think you have missed anything, it seems its just the CPU that might be bad.

SkyWarrior wrote:

Same thing happened here. Prime X299 Deluxe. Did not post by any means. Bios was reset and tadaaa. It's gone..

That's a fascinating situation that I've never encountered in a new motherboard before. However given all the posts about it here it looks like maybe that's a new wrinkle to worry about with these boards.

All these cannot be a coincidence but a weak post production QC before packaging. Probably due to a rushed release. Even the bios was very immature when I first booted. (Refused to install ubuntu in UEFI mode on NVMe drive. )

SkyWarrior wrote:

All these cannot be a coincidence but a weak post production QC before packaging. Probably due to a rushed release. Even the bios was very immature when I first booted. (Refused to install ubuntu in UEFI mode on NVMe drive. )

FYI, mine was not an x299 board, but a Gigabyte Ryzen board. My post was just to point out that sometimes it pays to do a CMOS reset straight out of the box

sigh... I had hit all the bios reset switches, trying one more board, then it's off to Ryzen for me!

Or maybe I say screw it and go coffee lake.

ok, tried the gigabyte pro AC gaming, and I got error code AF

I tried swapping DIMMs, clearing the bios, but I can't get any display to show up. I would try to flash the bios, but it doesn't POST at all.

anyone have any thoughts before I give up for good?

POST code AF appears to be an, "Exit Boot Services event".

To which nobody on the Internet seems able to explain. It's also terribly vague to interpret. By 'boot services' does it mean bootstrapping the board? By 'exit event' does it mean an error stopped initialization? If yes, that means hardware failure.

A bit of Googling around seems to imply there is no "fixing" an AF code.

I was able to get the machine to boot with a different CPU. What's weird is the first CPU posted a dead 00, and the second the AF, what's the likelihood of getting 2 bad CPUs??? Everything else in the system was consistent I checked this multiple times.

Chances are quite slim to get two bad CPUs, provided they were purchased from a reputable source.

Glad you found the answer!

Is this third CPU the same model as one of the other two? If not, update the bios to the latest version now that you have a combination booting and try the other two chips again.

It reeks of "motherboard needs newer bios with updated microcode in order to work" but as Ludi suggested, the combinations you listed ought to be recent enough to work.

Then again, I've heard nothing but horror stories from users and reviewers alike about the X299 platform. The whole thing just sounds like a total mess, as if Intel were caught off guard (which is laughable because Ryzen has been a slow, long-awaited launch and reputable leaks of Zen's silicon performance first hit the web 15 months ago!)

I've just started rolling out x299 systems and so far it has been smooth sailing, I'm thinking the issues were worked out shortly after launch.

leor wrote:

I was able to get the machine to boot with a different CPU. What's weird is the first CPU posted a dead 00, and the second the AF, what's the likelihood of getting 2 bad CPUs??? Everything else in the system was consistent I checked this multiple times.

Knew something was up the CPU (2 Cpus ), glad things worked out for you.