I would be more concerned IF the next gen CPU has this fixed. All's I know is that if Intel wants to fix the very next gen, they will need to scrap a lot of silicon that has already been finished. Sent from [ProtonMail](https://protonmail.com), Swiss-based encrypted email. -------- Original Message -------- On Jan 3, 2018, 15:35, Nathan O'Brennan wrote: > I'm more curious to know which versions of Intel's upcoming chips have been fixed already. I would like to upgrade my current workstation in the next year and will stick with Intel despite any performance impact over AMD. > > On 2018-01-03 00:43, Aaron Jones wrote: > >> >> I read the performance hit for Intel chips will be %35 or so after the fix. >> >> On Jan 2, 2018, at 7:49 PM, Eric Oyen wrote: >> >>> so, does this mean that the UEFI might get patched first? OR, does the OS ecology have to do so first? Lastly, how much of a performance hit will this represent? >>> >>> -eric >>> from the central offices of the Technomage Guild, the "oh look! yet another bug!" Dept. >>> >>> On Jan 2, 2018, at 3:39 PM, Matthew Crews wrote: >>> >>>> https://www.theregister.co.uk/2018/01/02/intel_cpu_design_flaw/ >>>> >>>> In a nutshell, it is a major security flaw in Intel hardware dating back a decade that is requiring a complete kernel rewrite for every major OS (Linux, Windows, Mac, etc) in order to patch out. It cannot be patched out with a CPU microcode update. Major enough that code comments are redacted in the patches until an embargo period is expired. Also the reported fix will have a huge performance impact. >>>> >>>> Also crucial to note is that AMD chips are not affected by this. >>>> >>>> How the heck does something like this go unnoticed for so long? >>>> >>>> Sent from [ProtonMail](https://protonmail.com/), Swiss-based encrypted email. >> >> --------------------------------------------------- >> PLUG-discuss mailing list - PLUG-discuss@lists.phxlinux.org >> To subscribe, unsubscribe, or to change your mail settings: >> http://lists.phxlinux.org/mailman/listinfo/plug-discuss