TR Forums

Just flashed the BIOS on my ASUS Z170-A a couple of days ago since it contained the microcode patch against Spectre. I see in the update notes that it also had some ME update included. I have flashed the BIOS several times before in this system and as in the previous tries, I did it from within the BIOS by loading the image from a USB flash drive. This time, after it had loaded the entire file from the flash drive, it just froze. Normally I would see the message that it started flashing but this time it was frozen, i.e. the mouse cursor was frozen and num lock could no longer be toggled on the kb. I left it in that state for about 15 minutes until I saw no other option that to turn off the system. Even the reset-button was unresponsive..

Now the system is completely dead. Drives and fans powers on when I press the power button but there is no POST and no beeps. The MB power LED is always on though whenever the PSU has power. The MB has a feature called ASUS Crashfree BIOS and I tried to see if I could make this work in case it was a bad BIOS flash. I prepare a USB stick with a Z170A.CAP file but the system never tries to read from it (the stick has an activity LED). I should also be able to hear some beeps and see an image on the onboard VGA output but there is nothing happening.

Now I worry that it is not just a corrupted BIOS on my MB but that something went wrong while it tried to update the microcode or Management Engine FW on my i7-6700k CPU.

I have run out of things to try for now, so I have ordered a new Z170-A as well as a Celeron G3900 CPU which was the cheapest CPU I could find which will run on this MB (even with ancient BIOS versions). I can always get a refund for them within 14 days if I find that I don't need them after all, but I don't see any other way of figuring out whether I should RMA the MB or the CPU (if I still can - haven't checked). Any input to this or thoughts into whether it is even possible that a CPU could be bricked by flashing a new MB BIOS?

The ME and microcode are part of the motherboard firmware too. Nothing gets flashed into the CPU.

Ok, when you write MB firmware is that something which can be swapped between MBs for bug-fixing or is it just bad luck and a new MB in case that gets corrupted? Now I haven't checked on mine yet but at least on some boards the BIOS is in a socket so can be swapped easily. I have a feeling though that these security 'features' are probably locked down to the specific board somehow?

FWIW a BIOS/UEFI update did not kill your CPU period. The chances of that are so tiny it's hard to put into words. Try clearing CMOS and flashing stuff again following the directions precisely as Asus has laid it out.

BIOS chips do indeed tend to be socketed on ASUS boards, so you can hot-swap them after booting from a good one and then flashing over the corrupt one. The BIOS embedded EZ Flash may not work if it's read from the BIOS chip immediately before the flash (after you've swapped chips), in which case you'd have to use the DOS version, in this case AFUDOS for American Megatrends BIOS/UEFI.

The software is not locked to a specific board but to any board of the same model, so if the BIOS image is unmodified, there should be no problems booting, flashing or with SecureBoot. However if you had encrypted the drives with Bitlocker, the paired key will be cleared/lost.

What bfg-9000 said, you should just call ASUS and have them send you a new BIOS chip (already flashed to the latest version) for your board, and it'd probably be free if it is under warranty. You can swap out the chip and send back the board you ordered.

The Z170-A does not have Crashfree BIOS, only ASUS' more expensive boards seem to offer it, kinda stupid.

The CrashFree BIOS function is not as user friendly as it should be.

Use the renaming tool to make sure you got the file name correct:
http://dlcdnet.asus.com/pub/ASUS/misc/u ... namerl.zip

Make sure to try multiple different USB ports. There typically is only one port that actually works. Make sure to use FAT32 and certainly try USB 2.0 flash drives.

The better solution would be if your board has the Asus Flashback function. For some reason Asus gave a header on your board (page 1-33 in the manual), but it doesn't look like they ever created/sold the card it says you need to buy.

You might have to just replace the BIOS chip to recover the board.

Ryu Connor wrote:

The CrashFree BIOS function is not as user friendly as it should be.

Maybe it should be called "CrashLessOften" instead. Or perhaps "CrashFree" means "we don't charge extra for the crashes"...

never ever turn off computer when flashing bios.
It doesn't matter if it looked dead, (frozen) because if you turn off the computer, it is 200 percent dead, whereas before it is maybe 90 percent dead.
If it is in warranty, why ask for a bios chip? Just send the whole thing back.
No, it didn't kill your CPU, it killed your motherboard.
edit: and maybe your graphic card.

Are you guys sure it has the Crashfree BIOS thing? I'm pretty sure it doesn't. The board ships with an empty "USB flashback pin header", but I've only seen the CrashFree BIOS advertised on ROG boards.

If the board does have it then he can use the driver CD the board shipped with instead of a USB drive. But again I don't think that board has it.

Kougar wrote:

Are you guys sure it has the Crashfree BIOS thing? I'm pretty sure it doesn't. The board ships with an empty "USB flashback pin header", but I've only seen the CrashFree BIOS advertised on ROG boards.

If the board does have it then he can use the driver CD the board shipped with instead of a USB drive. But again I don't think that board has it.

The manual says it has CrashFree (via optical or USB device) *and* FlashBack (which requires a special card that doesn't seem to be available anywhere).

Yes, the manual does say that it has Crashfree BIOS (not the same as Flashback which it does not have). Crashfree does require a working CPU while Flashback can work without a CPU installed.
I tried various USB ports both the front panel USB 2.0 ones, the one in the USB 3.0 header as well as a couple of the ones in the back, but I didn't try all the ports though. The manual did not mention that it was necessary to use a specific USB port.

I will try the renaming software as well as all ports just to make sure. I don't think I have the original CD any longer, but I guess if I just put any disc in, I should hear when it spins up the disc to read the content to check whether there is a BIOS image on it. If the drive never activates, it's probably safe to assume that the Crashfree function is not running.

Did you ever contact ASUS about a BIOS chip? Your board does have a socketed chip that can be swapped out in a few seconds. Even if you keep the new board, it would be worth doing just to fix the old board yourself to sell it.

So, I got the new board and the first thing I did was to transplant the BIOS chip from the new board to the old board. Interestingly, it doesn't bring it back to life so the problem is bigger than just a faulty BIOS chip. I tried the BIOS renaming tool and it showed that the file name I had chosen was already correct. I also found in a forum that the Crashfree function requires you to use the USB port that is just next to the Ethernet port (it also has a slightly different blue color). There are no beeps to be heard as there should be so I guess the next thing for me to do is to move my i7-6700k from the old board to the new to verify that the CPU is actually still working..

Is the new board the exact same model and version as the old one? BIOSes are specific to motherboard versions.

-assuming there are no "beeps" indicating a non-functioning CPU, then it "sounds" as if it's a power issue with the motherboard (..like a voltage spike taking-it-out) and/or an issue with the power supply.

CScottG wrote:

-assuming there are no "beeps" indicating a non-functioning CPU, then it "sounds" as if it's a power issue with the motherboard (..like a voltage spike taking-it-out) and/or an issue with the power supply.

Would be a pretty crazy coincidence for that to happen right in the middle of a BIOS flash.

1. Never shut off pc while flashing bios. It tells you not to do it.
2. Stop it with all these "just change the chip." Now he's got two useless mobo instead of one.
3. when flashing bios after failure, do not use the front panel usbs. Do not flash bios with front panel usb in the first place.
4. try every single one of the back panel usb. start with the one with the different color.
5. FAT32?
6. flash with the back panel usb, usb2.0 if possible.

just brew it! wrote:

CScottG wrote:

-assuming there are no "beeps" indicating a non-functioning CPU, then it "sounds" as if it's a power issue with the motherboard (..like a voltage spike taking-it-out) and/or an issue with the power supply.

Would be a pretty crazy coincidence for that to happen right in the middle of a BIOS flash.

The worst stuff always happens in the middle of a BIOS flash.

just brew it! wrote:

CScottG wrote:

-assuming there are no "beeps" indicating a non-functioning CPU, then it "sounds" as if it's a power issue with the motherboard (..like a voltage spike taking-it-out) and/or an issue with the power supply.

Would be a pretty crazy coincidence for that to happen right in the middle of a BIOS flash.

I suspect it was from flipping the power supply on/off.. (..and it might not have happened from the turn-off during the BIOS flash).

-but yeah, Murphy's Law.

Darthutos wrote:

1. Never shut off pc while flashing bios. It tells you not to do it.

He waited 15 minutes. While I would've would've waited longer in that situation, odds are nearly 100% it was already dead. I've never had a successful BIOS flash take more than a couple of minutes.

Darthutos wrote:

2. Stop it with all these "just change the chip." Now he's got two useless mobo instead of one.

Where did he say the second mobo was dead?

So the status is that my computer is now up and running again with the new MB and all the other old components including the CPU. The new MB also boots fine with the BIOS IC from the old MB so something clearly happened to that board other that a simple duff BIOS flash. The new MB is the exact same type (Part# on the box label is identical) and there is no other revision of this model.

I know that turning off a system while flashing is not good, but I had no other choice. Should I just have left it on and frozen for eternity? It should normally not stay unresponsive for more than maybe 5-10 secs during flashing, so I doubt that waiting another hour would have made any difference.

Now I am not sure what happened with the old board, but they do write that it is covered by a 3 year warranty and since I bought it 2½ years ago I am doing the RMA process for it now. I hope they will accept it although ASUS is not known to be particularly good in this regard. The backup/test CPU that I bought will be returned for a full refund since I never opened the package.

I did take a gamble and updated the BIOS on the new MB since the one that it was supplied with was quite ancient but this time I did not use the latest version which has the Spectre microcode update in it (which we now know can cause a multitude of problems and Intel advices people to stay away from for now), but instead I took the version just before that. And for reference, booting the new board with the BIOS chip from the old board booted to the version that I was using before I tried to update it, so it looks like it never started the actual flashing process.

Well that's pretty crazy. So the motherboard just happened to die as you were attempting to update the BIOS. What were the odds against that?

demolition wrote:

I know that turning off a system while flashing is not good, but I had no other choice. Should I just have left it on and frozen for eternity? It should normally not stay unresponsive for more than maybe 5-10 secs during flashing, so I doubt that waiting another hour would have made any difference.

Now I am not sure what happened with the old board, but they do write that it is covered by a 3 year warranty and since I bought it 2½ years ago I am doing the RMA process for it now. I hope they will accept it although ASUS is not known to be particularly good in this regard. The backup/test CPU that I bought will be returned for a full refund since I never opened the package.

No, you did what you had to. Again, it might not have had anything to do with the time you turned it off during the BIOS flash - it could well have been a turn-on afterwards.
Note: in this respect then I'd be careful if using the same power supply.

I've had more problems with mother-boards than any other component (on about a 10-1 basis) - and NOT the cause of careless ESD, nor was it always a complete failure of the mother-board. Power supplies run a distant second for me. (..I once even had a bad memory stick.) My mother's laptop had a new Windows 10 Radeon driver kill her GPU - over-volting to heat death, and I've heard of this with CPU's as well with BIOS updates (..but fortunately that wasn't your problem).


-anyway, I hope the RMA goes well for you. (..I've also had bad luck with them from various manufacturers past a 90 day window.)

I gave up trying to get a new BIOS chip from Asus the time I had a flash go bad. Ended up ordering a replacement BIOS chip from a third-party BIOS recovery site. Probably cost me about as much as it would've cost to ship the board back to Asus for an RMA, and with less hassle.

I don't know about Asus's service in the USA, but don't expect Asus Canada to honor whole mobo RMA. They will make all kind of excuses to charge you full price on the dead board.
In fact, I'm surprised they gave you a "bios chip".
One time my ethernet port died. They took pictures of my board with tiny white dust/scratch on the edge of the board, and tells me there was damage so they won't honor the warranty. Funny thing is they ignored the missing gpu bar. They focused on tiny grains of white (scratches?) on the edge of the mobo.
I bought the mobo for 300 dollars. They want to "fix it" for 275 and shipping.

tl;dr: Don't count on Asus warranty.

Well can't get any more definitive than that. Wasn't the BIOS update itself that killed the board. Waiting longer wouldn't of mattered an iota since the old BIOS chip was still good.

just brew it! wrote:

Well that's pretty crazy. So the motherboard just happened to die as you were attempting to update the BIOS. What were the odds against that?

Aye, that is amazingly bad luck. That it died before even beginning the flash process as well... even a partial flash would have left the chip in an unbootable state when he placed it in the new board. But since it worked, the flash process had apparently never even begun.

Does ASUS perform failure testing on returns? I am dying of curiosity to know what happened now to that board.

Kougar wrote:

Does ASUS perform failure testing on returns? I am dying of curiosity to know what happened now to that board.

I'm sure they do but they are probably not going to tell me the result - only whether or not they think it is covered by the warranty. They will probably fix the board if they can and then put it into their pool of refurbished boards (which is fair enough).

Since replacing the MB, I've had a new problem that I'm trying to figure out now.. I now get audio crackling/pops. At first, I thought that maybe it was just my X-Fi sound card that had developed a bad connection to the socket or something, but I have exactly the same issue with the on-board Realtek audio. It is not just from playing audio on the PC, but also with audio that I loop straight through the audio card from the line input (I think it goes through an ADC and is then mixed digitally with the played back audio). The noise sounds to me like a buffer underrun issue and it gets much worse if I choose 192kHz as the endpoint standard sampling rate rather than 48kHz where the problem is less noticable but still present.

I do think that it is a driver issue though as it seemed to be working with the onboard audio after I uninstalled the Creative drivers but after reinstalling the X-Fi drivers the problem has returned. Perhaps it is a BIOS setting issue - maybe some power saving thing that interrupts the data?

demolition wrote:

Since replacing the MB, I've had a new problem that I'm trying to figure out now.. I now get audio crackling/pops. At first, I thought that maybe it was just my X-Fi sound card that had developed a bad connection to the socket or something, but I have exactly the same issue with the on-board Realtek audio. It is not just from playing audio on the PC, but also with audio that I loop straight through the audio card from the line input (I think it goes through an ADC and is then mixed digitally with the played back audio). The noise sounds to me like a buffer underrun issue and it gets much worse if I choose 192kHz as the endpoint standard sampling rate rather than 48kHz where the problem is less noticable but still present.

I do think that it is a driver issue though as it seemed to be working with the onboard audio after I uninstalled the Creative drivers but after reinstalling the X-Fi drivers the problem has returned. Perhaps it is a BIOS setting issue - maybe some power saving thing that interrupts the data?

Are all cards in the same slots they were in before? This really shouldn't matter with PCIe (and if your X-Fi is PCI there's only one slot it could possibly be in), but we need to eliminate potential variables here.

Were you using 48kHz before, or was it possibly set to 44.1?

just brew it! wrote:

Are all cards in the same slots they were in before? This really shouldn't matter with PCIe (and if your X-Fi is PCI there's only one slot it could possibly be in), but we need to eliminate potential variables here.
Were you using 48kHz before, or was it possibly set to 44.1?

Yes, all cards are in the same slots that they were in before, however I think that Windows could well have reinstalled all the drivers anyway. I did see it install some things on the first boot on the new board and it also lost the activation so I had to get a new license # since I had not linked the old activation with an MS account.. The X-Fi is PCI so can only be in one slot as you write. The only peripherals that I did not install in the same locations are all the USB ones. Even all the SATA cables are in the same place as they are all numbered. The BIOS settings may obviously be different from before as I don't know for sure how they were all set on the old board.

I believe it was set for 24bit / 96 kHz before, but it is now noisy on all format settings, be it 16/24/32 bit and 44/48/96/192 kHz but generally the higher ones are more noisy than the lower ones. I do remember that I have had a similar problem some time ago but it was resolved by changing the mode of the X-Fi to Audio Creation Mode instead of the default. I like to have bit-matched playback, lower latency and with all the fancy effects turned off so Audio Creation Mode was always the preferred mode for me.