You work around the bug. It's done every single processor generation. http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/desktop-6th-gen-core-family-spec-update.pdf Skylake has 53 known errata issues, 54 if you count this one. Haswell had 108: http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/5th-gen-core-family-spec-update.pdf I'm sure they'll end up finding more as Skylake matures. But regardless Intel says it has a fix. Yeah sure, if the Fix ends up crippling Skylake performance by a huge amount then Intel should recall the processor. But there is zero indication of that being the case.

Heh, never knew they screw up so many things. Don't they test their stuff on the usual, like prime number calculation? Reminds me of that bug with AVX, or previus Intel's that crashed too at certain prime numbers and calculations...

stick with '-E' system..never doubt it:)

It's an architectural problem, isn't it? If so, they should have the same bugs, or did I get that wrong?

Along with this update they will block OC for standard processors and all will be happy 😉

You work around the bug. It's done every single processor generation. http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/desktop-6th-gen-core-family-spec-update.pdf Skylake has 53 known errata issues, 54 if you count this one. Haswell had 108: http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/5th-gen-core-family-spec-update.pdf I'm sure they'll end up finding more as Skylake matures. But regardless Intel says it has a fix. Yeah sure, if the Fix ends up crippling Skylake performance by a huge amount then Intel should recall the processor. But there is zero indication of that being the case.

Interesting. I'm really curious how those workarounds work (in human language). Also, do they ever start manufacturing such CPU without certain bugs after some time?

You work around the bug. It's done every single processor generation. http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/desktop-6th-gen-core-family-spec-update.pdf Skylake has 53 known errata issues, 54 if you count this one. Haswell had 108: http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/5th-gen-core-family-spec-update.pdf I'm sure they'll end up finding more as Skylake matures. But regardless Intel says it has a fix. Yeah sure, if the Fix ends up crippling Skylake performance by a huge amount then Intel should recall the processor. But there is zero indication of that being the case.

Thanks, I didn't know that there is such an errata list available. 108 would be for Broadwell as of 5th generation. 155 for 4th generation Haswell 114 for 3th generation So Skylike is the most robust so far...

Interesting. I'm really curious how those workarounds work (in human language). Also, do they ever start manufacturing such CPU without certain bugs after some time?

According to the documentation most of the bugs in errata have NO FIX status which means there are no plans to release a newer CPU stepping which would fix those bugs in hardware.

T So Skylike is the most robust so far...

Or the problems just haven't been discovered yet since it's been out the least amount of time.

This could affect performance, but they may simply implement a BIOS-level exception handler to handle the error the processor will throw which might not noticeable hamper performance. I personally don't know a huge amount about how the BIOS functions, but it is in itself very much like an operating system. You can program it to do different things in different scenarios, so I can see it being possible to implement the ability to handle certain errors. Bugs occur all the time. It's nigh impossible to create something so complex without there being bugs, and to see less than 100 bugs on something that implements over a billion transistors is frankly (in my opinion) astounding, especially as many (read: the majority) of them are correctable via the BIOS. Many of the bugs are not correctable at the physical level due to several common factors: 1. Fixing the bug in hardware would require major reworking that simply isn't work it or possible at a certain point of production/design. 2. The bug is minor in the way it affects the system. 3. The bug affects relatively very few people (not referring to this bug in specific, but some of the other bugs listed).

Sorry guys i have a friend that is claiming that AMD don't have these problems could anyone link me a PDF or site that shows that AMD have microcode updates or defect in the processor

Sorry guys i have a friend that is claiming that AMD don't have these problems could anyone link me a PDF or site that shows that AMD have microcode updates or defect in the processor

http://support.amd.com/TechDocs/51810_16h_00h-0Fh_Rev_Guide.pdf http://support.amd.com/TechDocs/48063_15h_Mod_00h-0Fh_Rev_Guide.pdf Every computational device is going to have Errata. A company like Cadence builds entire systems dedicated to help simulating systems before they are manufactured. http://www.cadence.com/products/sd/palladium_xp_series/pages/default.aspx Validation is like a huge part of designing a chip. But when you have upwards of 1B+ transistors its becomes near impossible to get everything, even for companies with advanced validation and testing systems.