What's Your Word on Skylake "Prime95" Bug?

[Abir]

Guys, people over at Mersenne Forum found out about the issue while finding prime numbers using Fourier Transformation that Skylake processors (Specifically, Core i7-6700K) freezes after running Prime95 ver. 28.7 for couple of minutes or hours while using AVX2.0 for calculation. What's your word on that? How long would all the motherboard vendors take to release the new "Intel Fix" as a BIOS update? Have you got yours??

 

Cheers!

[BrightCandle]

Intel hasn't released the change yet as far as I know. When they do we will see the roll out of the microcode updates.

[Abir]

Man, I just bought the system yesterday! But I'm happy, though!

[PrimeSonic]

IMHO, it's really nothing to worry about for most people. Like, only in some very specific, and usually scientific, applications will this bug ever come into play.

Just lay off Prime95 in the meantime.

 

I don't use Prime95 and this bug didn't come up in any of the CPU stress tests and benchmarks I ran:

Cinebench, CPU-Z, Intel XTU, 3DMark FireStrike (Physics).

 

None of them caused instability once I found the stable overclock I wanted. (I'm sticking to 4.7Ghz until I get better cooling.)

 

So honestly, for almost everyone, this bug isn't something to worry about.

And the fact that a simple BIOS update will fix it means it wasn't that big of a deal in the first place.

[MageTank]

IMHO, it's really nothing to worry about for most people. Like, only in some very specific, and usually scientific, applications will this bug ever come into play.

Just lay off Prime95 in the meantime.

 

I don't use Prime95 and this bug didn't come up in any of the CPU stress tests and benchmarks I ran:

Cinebench, CPU-Z, Intel XTU, 3DMark FireStrike (Physics).

 

None of them caused instability once I found the stable overclock I wanted. (I'm sticking to 4.7Ghz until I get better cooling.)

 

So honestly, for almost everyone, this bug isn't something to worry about.

And the fact that a simple BIOS update will fix it means it wasn't that big of a deal in the first place.

Odd that XTU did not crash you. It uses AVX2 Linpack (At least, it used to at one point).

 

That being said, you are right for the most part. Some folding cores use AVX2, such as GROMACS, and i know PCSX2 (emulator) uses AVX2 as well. Outside of that, I can't really think of much else that uses it that a normal consumer might care about. 

 

We should see the fix relatively soon I think. Before the end of the month hopefully. Intel has said they already have the fix ready, just waiting on board partners to apply it.

[PrimeSonic]

We should see the fix relatively soon I think. Before the end of the month hopefully. Intel has said they already have the fix ready, just waiting on board partners to apply it.

And let's not forget that this fix would require a BIOS update.

Something that should never be taken lightly.

 

So if the bug isn't actually impacting you. Then it might not be worth the risk.

 

Now, since my motherboard has dual-BIOS, I might give it a shot anyways. I can recover if something goes wrong.

But for the average consumer, it's likely not going to be a big deal.

[Nena Trinity]

If I where to buy it I would see if the fix works first...

[Bit8xPr0]

It's a trick to get people into putting a BIOS update on their motherboards that stop non K CPU's from being overclocked

[Funtron5000]

It's a trick to get people into putting a BIOS update on their motherboards that stop non K CPU's from being overclocked

[emoji50] I knew it!

[AresKrieger]

It's a trick to get people into putting a BIOS update on their motherboards that stop non K CPU's from being overclocked

 

I is readys with my hats

 

But they might just do that since they already have to patch the bios anyway, so I doubt this was all a plan but the end result is the same potentially