Topic on Glossary talk:Central Processing Unit (CPU)
This page shows the changes between two versions of a post by Mirh in the topic "AMD hidden shenanigans" on Glossary talk:Central Processing Unit (CPU).
You can see other versions of this post at its history page.
Line 1: | Line 1: | ||
I'm not talking about the usual just-so-frequent incompatibilities, crashes, or other obvious problems, and not even about the [https://www.overclockers.com/laptop-hacking/ various] [https://semiaccurate.com/2012/01/09/spotlight-brazostweaker/ hacked] [https://www.overclock.net/forum/10-amd-cpus/1730316-undervolting-bristol-ridge-laptops-feat-a10-9600p.html overclocking] [https://github.com/FlyGoat/RyzenAdj tools]. But something a bit more in the shadow: | I'm not talking about the usual just-so-frequent incompatibilities, crashes, or other obvious problems, and not even about the [https://www.overclockers.com/laptop-hacking/ various] [https://semiaccurate.com/2012/01/09/spotlight-brazostweaker/ hacked] [https://www.overclock.net/forum/10-amd-cpus/1730316-undervolting-bristol-ridge-laptops-feat-a10-9600p.html overclocking] [https://github.com/FlyGoat/RyzenAdj tools]. But something a bit more in the shadow: | ||
+ | |||
+ | * Original Phenoms cpu had a very [https://www.anandtech.com/show/2477/2 concerning] bug with TLB. Gaming isn't as sensitive of a task as compiling tax returns though, so bold people could try to [http://www.xtremesystems.org/forums/showthread.php?171105-Howto-disable-the-tlb-fix-with-CrystalCPU disable] the software workaround. | ||
* Llano has [https://community.hwbot.org/topic/185623-fm1-and-the-unexplainable-gpupi-performance/ quite reduced] DIV/IDIV performance <small>(I'm really [https://bugzilla.mozilla.org/show_bug.cgi?id=1281759#c39 not] the the guy recommending to undo errata fixes, but considering all the recent guarding around speculative execution.. maybe there is a chance?)</small> | * Llano has [https://community.hwbot.org/topic/185623-fm1-and-the-unexplainable-gpupi-performance/ quite reduced] DIV/IDIV performance <small>(I'm really [https://bugzilla.mozilla.org/show_bug.cgi?id=1281759#c39 not] the the guy recommending to undo errata fixes, but considering all the recent guarding around speculative execution.. maybe there is a chance?)</small> | ||
* Bulldozer has [https://web.archive.org/web/20131007224201/http://www.xtremesystems.org/forums/showthread.php?286448-The-Book-of-Bulldozer-Revelations-Episode-2-(SuperPI-x87) crippled] x87 performance <small>(there are no public errata regarding the feature, so this might be all free real estate?)</small> | * Bulldozer has [https://web.archive.org/web/20131007224201/http://www.xtremesystems.org/forums/showthread.php?286448-The-Book-of-Bulldozer-Revelations-Episode-2-(SuperPI-x87) crippled] x87 performance <small>(there are no public errata regarding the feature, so this might be all free real estate?)</small> | ||
+ | |||
+ | * Bulldozer (and I could swear Bobcat) CPUs are liable to [[Store:Steam#Steam CEG protected games crash or freeze the system (AMD Bulldozer CPU)|system locking up]] on steam CEG games loading. Given the time window, and given some [https://bugs.winehq.org/show_bug.cgi?id=32515#c49 wine discussion] involving debug registers, I believe this might have been [https://www.amd.com/system/files/TechDocs/48063_15h_Mod_00h-0Fh_Rev_Guide.pdf#page=57 erratum 671]. | ||
* Possibly every architecture before Ryzen [https://www.overclock.net/forum/10-amd-cpus/1499562-undervolting-editing-p-states-piledriver-using-amd-mrstweaker-mini-guide.html could hit] the TDP ceiling before even reaching the maximum frequencies <small><strike>(unclear whether this couldn't just be specific to some motherboards and/or workload though)</strike></small> | * Possibly every architecture before Ryzen [https://www.overclock.net/forum/10-amd-cpus/1499562-undervolting-editing-p-states-piledriver-using-amd-mrstweaker-mini-guide.html could hit] the TDP ceiling before even reaching the maximum frequencies <small><strike>(unclear whether this couldn't just be specific to some motherboards and/or workload though)</strike></small> | ||
Line 21: | Line 25: | ||
* I couldn't find [https://www.overclock.net/forum/78-pc-gaming/1487105-want-higher-fps-what-timer-do-you-use-2.html#post26423128 anybody] on the net [https://www.anandtech.com/show/12678/a-timely-discovery-examining-amd-2nd-gen-ryzen-results/4 seriously] reporting that forcing HPET helped them save for touching BCLK [http://forum.notebookreview.com/threads/must-read-setfsb-o-c-game-problems-fixed.569893/ live] or under [https://web.archive.org/web/20160314074421/https://www.ocaholic.ch/modules/smartsection/item.php?itemid=1071&lang=english W8](+?). Still for some reason [https://forums.guru3d.com/threads/another-look-at-hpet-high-precision-event-timer.368604/page-42 something] in the AMD drivers always does/did that. | * I couldn't find [https://www.overclock.net/forum/78-pc-gaming/1487105-want-higher-fps-what-timer-do-you-use-2.html#post26423128 anybody] on the net [https://www.anandtech.com/show/12678/a-timely-discovery-examining-amd-2nd-gen-ryzen-results/4 seriously] reporting that forcing HPET helped them save for touching BCLK [http://forum.notebookreview.com/threads/must-read-setfsb-o-c-game-problems-fixed.569893/ live] or under [https://web.archive.org/web/20160314074421/https://www.ocaholic.ch/modules/smartsection/item.php?itemid=1071&lang=english W8](+?). Still for some reason [https://forums.guru3d.com/threads/another-look-at-hpet-high-precision-event-timer.368604/page-42 something] in the AMD drivers always does/did that. | ||
− | And I could swear I knew more, but time was hard on my memory. | + | And I could swear I knew more, but time was hard on my memory. I wonder if all of this couldn't be enough to resurrect the old [[Special:Undelete/AMD_(CPU)|AMD (CPU)]] page we had. |
<small>p.s. Ryzen is so [https://www.bitsandchips.it/english-news/xfr-and-its-anchestors stacked] [https://www.gamersnexus.net/guides/3491-explaining-precision-boost-overdrive-benchmarks-auto-oc complex] and [https://www.techpowerup.com/review/1usmus-power-plan-for-amd-ryzen-new-developments/ wide] that you'd need a [https://www.anandtech.com/show/14873/reaching-for-turbo-aligning-perception-with-amds-frequency-metrics- manual] just to understand the stock situation.</small> | <small>p.s. Ryzen is so [https://www.bitsandchips.it/english-news/xfr-and-its-anchestors stacked] [https://www.gamersnexus.net/guides/3491-explaining-precision-boost-overdrive-benchmarks-auto-oc complex] and [https://www.techpowerup.com/review/1usmus-power-plan-for-amd-ryzen-new-developments/ wide] that you'd need a [https://www.anandtech.com/show/14873/reaching-for-turbo-aligning-perception-with-amds-frequency-metrics- manual] just to understand the stock situation.</small> |