KB4100347 Intel microcode updates for Windows 10 v1803 - January 8 Win Update

Bree said:

If you look in C:\Windows\System32 you will find two files, mcupdate_AuthenticAMD.dll and mcupdate_GenuineIntel.dll. These are installed by default on any clean install of Windows 10. Each contains all known microcode updates at the time they were published.

At boot only the appropriate one for your processor (Intel or AMD) will be checked to see if it has newer microcode for your specific CPU than is currently loaded, only if it does will the microcode be uploaded to the processor.

On my Intel system KB4100347 has updated my 'GenuineIntel' one to the latest version while my 'AuthenticAMD' is still dated as it was when Windows was installed. I suspect if @Jigsaw99 looks at the dates of theirs it will be the other way round.

Thanks for the reply man!I looked the files you said and the AMD.dll has the date 12/4 which is when I installed Windows10 and the Intel.dll has the date 11/7.I don't have the knownledge all the other guys around here have but since everything is working well,I guess I just have to leave it alone and not try to break something that is working good!

ddelo said:

According to the first post from Microsoft/Intel, for my Broadwell U CPU (306D4), I must have microcode version 0x2A. That's what I had since Feb. 2018, until hp provided a BIOS update in June and it took it up to version 0x2B.
Today I was offered the KB4100347 update, although at a higher version and I used the wushowhide.diagcab to hide it.
Although successfully hidden, it's still offered by WU, as I didn't install it when it was offered and I must force a recheck to hopefully get rid of it, since it seems to be redundant.

Just because it's not needed doesn't mean you need to go to the trouble of hiding it.

If your BIOS level is higher, it's disregarded, it won't affect anything. If it is updated and you don't get a BIOS update, then you'll have hidden it.

I can tell you I have the same CPU and Acer have not provided an update for me, KB4100347 even in it's revised form contains 0x2A.

I am now hopeful that the scope of KB4346084 is extended to older CPUs. This I think is how Microsoft will achieve the patch granularity that I hoped for.

I have an i5 6600k. Wondering if there are any performance hit WRT gaming.

ern88 said:

I have an i5 6600k. Wondering if there are any performance hit WRT gaming.

Pure CPU scores are affected in benchmarks (currently based around game engines with settings maxed out) this affects your GPU framerate so I have elected to block uCode updates on my gaming desktop, use the last release uCode from 2017-11 and turn off the protections via registry.

I'm scoring higher benchmarks than before Spectre patches were enabled, in-game performance has improved by a few FPS on average. My desktop is Haswell-based.

On my laptop that I use for work, all the updates are enabled (as far as it goes) and patched all bar CVE 2018-3639 (SSB).

winactive said:

Pure CPU scores are affected in benchmarks (currently based around game engines with settings maxed out) this affects your GPU framerate so I have elected to block uCode updates on my gaming desktop, use the last release uCode from 2017-11 and turn off the protections via registry.

I'm scoring higher benchmarks than before Spectre patches were enabled, in-game performance has improved by a few FPS on average. My desktop is Haswell-based.

On my laptop that I use for work, all the updates are enabled (as far as it goes) and patched all bar CVE 2018-3639 (SSB).

Have you checked this link for v 1.33 update for your model: https://us.answers.acer.com/app/answ...ulnerabilities

brummyfan said:

Have you checked this link for v 1.33 update for your model: https://us.answers.acer.com/app/answ...ulnerabilities

Thanks yes, it contains 0x2A not 0x2B. I have a support ticket open with Acer (who weren't aware 1.33 was even released as it's not on the product page). Embarassing!