Skylake Bugs Aren't Odd, They're Prime

Status
Not open for further replies.

Darkk

Distinguished
Oct 6, 2003
615
0
18,980
This reminds me the time when a FPU bug was found in the Pentium 60 processor. Intel tried to say very few users will ever experience this bug. Intel ended up swapping the processor out after a public outcry.
 

InvalidError

Titan
Moderator

If the bug is fixable through a microcode update with no performance or power penalty, then it was just a minor microcode bug and won't have a stepping to address it specifically since there is no actual hardware bug to fix.

If you read errata for past CPUs, most still have around a hundred known bugs in their last production stepping, all addressed by microcode updates and software patches. Many of those non-critical bugs even carry over across multiple CPU generations since the next-gen chips are usually too far into the design phase to bother fixing those minor bugs by the time they are discovered.

Here, Kaby Lake must be getting close to first silicon tape-out if that hasn't occurred already. If it reused the same logic responsible for this bug, it is likely too late for a hardware fix and Kaby will carry over the microcode fix.
 

Cryio

Distinguished
Oct 6, 2010
881
0
19,160
Skylake has a lot of issues IMO. Boot/shutdown times on my Surface Pro 4 i5 are really bad. From what I read around, its battery life is also a lot worse compared to the Pro 3.
 

bloodroses75

Distinguished
Aug 10, 2009
186
0
18,710
This reminds me of the old Pentium bug years ago. Hopefully the patch doesn't kill the performance like that one did. I know Intel will get it fixed on a hardware revision; I wonder if they'll offer free replacements like they did before to those that want it.
 

blppt

Distinguished
Jun 6, 2008
568
89
19,060
How about the "feature" in my Haswell 4790K which turns it into a miniature supernova when AVX2 instructions are used in Prime95. ;)

One wonders why bother including these features if the chip cant handle them at stock speeds.
 

Kewlx25

Distinguished
AVX2 at a lower clock is still faster than non-AVX2 at a higher clock. It's called a trade-off. The desktop Haswell increases the voltage when AVX2 gets used, the Xeon Haswells reduce the clock speed and keep the voltage the same.
 

blppt

Distinguished
Jun 6, 2008
568
89
19,060
AVX2 at a lower clock is still faster than non-AVX2 at a higher clock. It's called a trade-off. The desktop Haswell increases the voltage when AVX2 gets used, the Xeon Haswells reduce the clock speed and keep the voltage the same.

Not on the 4790K---mine goes straight for 100C before it thermally throttles. That doesnt seem like a great way of reducing the clock, you know, making it so hot that the chip death protection cuts in.
 

vaughn2k

Distinguished
Aug 6, 2008
769
4
19,065
"Intel told us that average users should not be too concerned about this bug, however, as has only a slim chance of occurring in compute-intensive tasks, and it should not occur unless a user is running extremely complex workloads." - A bug is a bug. An average consumer does not know what is a complex task and what is not. There is no measurement for it, until Intel will have a data to provide that the risk is low. Do you want to ride a car that has a bug in controlling the brakes? No. I'll keep my Skylake for now, as I have seen the bug. And wait for a recall, if any (I know there will be).
 

InvalidError

Titan
Moderator

The bug has been fixed by a BIOS microcode update, like countless other bugs in past generations of Intel chips. There won't be a recall since all you need to do to fix it is update your BIOS when your motherboard manufacturer releases an updated BIOS that includes the fix.

The FDIV recall was because someone at Intel forgot an entry in the division lookup table and there was no possible microcode fix for that. The only possible work-around was having the OS intercept the FDIV op and execute it in software, which produced unacceptable performance loss from the overhead of running an exception handler on every FDIV.
 

alextheblue

Distinguished

If the bug is fixable through a microcode update with no performance or power penalty, then it was just a minor microcode bug and won't have a stepping to address it specifically since there is no actual hardware bug to fix.

That's an "if" that we don't know the answer to yet. Hopefully after the fix is out, TH et al will do some testz0ring. I hope there's no performance hit, I helped my nephew pick out parts for a Skylake rig and gave him advice during assembly over the phone.
 

blppt

Distinguished
Jun 6, 2008
568
89
19,060
"Sorry your cooling sucks."

Granted, I'm not watercooling, but given that the 4790K @ stock nearly doubles its rated TDP under a P95 AVX2 load.....doesnt seem like brilliant engineering to me, especially since the coolers that Intel recommends for it cant handle a 150+W thermal load.

I'm using a cheap Coolermaster HyperT4 in a decent airflow Corsair 750D. No, its not the H110 CLC I have on the 9590, but certainly better than the stock HSF. I can only imagine how quickly it would head for throttling with that intel cooler on it.
 
This would explain the random worker stoppages I've been seeing on my 6700k in Prime. They had better correct it via microcode, because if that's not all it is, I'm going to add Intel to the list google is already on, and kill them both. :)
 
Status
Not open for further replies.