ASRock Kills Intel Skylake Bug With BIOS Update, Fire

Status
Not open for further replies.

Bannereus

Reputable
Oct 19, 2014
22
0
4,520
ASUS has also had their update out since the 18th for some mainstream boards, and the ROG series by the 19th.

It was never going to kill Skylake, but I'm glad the mobo manufacturers and helping Intel remedy this bit of silliness.
 

EdgeT

Distinguished
Jan 8, 2009
280
7
18,815
Click-baiting much, Tom's ? They didn't kill it with fire (like an overheat to resolder something in place), they just released a BIOS update.

Stop click-baiting, or you'll lose people, just like the other countless click-bait sites. Pathetic.
 
Good on ASRock to get this fixed, they have made alot of ground.
However, who at ASRock got paid to design that picture? I want their job...
On an more serious note. A worm is not a bug, why is it there??
 

none12345

Distinguished
Apr 27, 2013
431
2
18,785
"ASRock Kills Intel Skylake Bug With BIOS Update, Fire"

The word fire did not need to be there at all, as fire has nothing to do with it. Having that picture is fine, but the word fire in the link to this article is pure click bait. And ya, i too man fed up with click bait link titles.
 

farlandprince

Reputable
Jan 11, 2016
9
0
4,510
How can a bios update fix CPU bugs ? I dont get it at all.

besides , we need Tomshardware to test the CPU using the new bios . and some benchmarks please.
 

bit_user

Polypheme
Ambassador
Yeah, I found that a little misleading... was wondering whether the BIOS update that fixes the bug actually set someone's PC on fire. But I guess it makes sense if ASRock designed and provided that pic.

I'd have gone with something like this:

Code:
   ASRock on Skylake bug: "Kill it with fire" and a BIOS update
 

farlandprince

Reputable
Jan 11, 2016
9
0
4,510


You did not explain how.

If the fix means "do it again" for every Error , then this is not a fix and will lower the performance.

This is more "not freeze" and not a REAL FIX.

Intel should recall all the CPUs and release new ones .

why on earth Intel can get away with this?
 

Xaltar

Reputable
Dec 9, 2015
10
0
4,510


Changes in the microcode alter the way the CPU functions and can and do resolve issues like this. A google search on the term "CPU microcode" will provide more info than I can provide here. Think of microcode as firmware for the CPU. Suffice to say, the issue is resolved as Skylake CPUs are now, with the new microcode update, able to process the same data that caused instability without errors or crashes. In other words the CPUs will now function correctly. The microcode is not a part of the BIOS though it is usually released along side a BIOS update and is written through the BIOS. This bug was very specific and was only triggered when forcing AVX instructions to be used in prime95 or any other workload that triggered the same instructions sets, in other words very, very few real world tasks. This is not the first time a microcode update has been used to patch a bug in a new CPU line and it will not be the last. Recalling an entire line of CPUs for what is in essence a firmware glitch would be a massive waste of time and resources.

I hope this clarifies things for you.
 

Tim_34

Reputable
Jan 26, 2016
1
0
4,510
Did ASRock pull the new BIOS updates? The latest update I see for my Z170 Extreme4 is v2.40 with a date of 12/2/2015.
 

Xaltar

Reputable
Dec 9, 2015
10
0
4,510


The guys in Tech Support are very reliable with BIOS updates and fixes. I am a big fan of how quickly ASRock respond to bug reports and take down problematic updates. According to the list of updated boards on the ASRock forums v2.70 was the update that included the 0x74 fix so given it is no longer available for download tech support must have found a problem with it and taken it down before too many people experienced problems with it. A new update should be available soon. Luckily the update isn't critical to system usability. If you do for some reason require the update sooner you can contact ASRock tech support directly and they will assist you.
 

bit_user

Polypheme
Ambassador
A simple benchmark should show whether the fix hurts performance. If it doesn't hurt performance and does fix the problem, then what's to complain about?

As for being a "REAL FIX", we just don't know unless Intel at least publishes the details of the problem whether it might be properly fixable via a microcode or BIOS patch. Have they detailed the problem? Nobody has mentioned an errata number, so I assume not. If you that concerned about it, maybe you can find it.
 
Status
Not open for further replies.