F5F Stay Refreshed Power Users Overclocking Realbench's massive score dropped following the 1803 event.

Realbench's massive score dropped following the 1803 event.

Realbench's massive score dropped following the 1803 event.

K
kgavaga
Member
112
03-07-2016, 12:33 PM
#1
4770k on Asus Maximus VI Hero with 1603 bios, EVGA 1070 SC, 16Gb TridentX 2400. Around 12 runs before the Win10 1803 update, averaging about 106k. After the update, roughly 10 runs average 97k. Post-1803, I used DDU to remove graphics drivers, updated to the newest version, still maintaining an average of 97k. The only idea that came to mind was that Haswell was more affected by the Meltdown/Spectre patches (my bios are a few years old, but the latest on ASUS site for my board doesn’t mention Spectre fixes). Any suggestions?
K
kgavaga
03-07-2016, 12:33 PM #1

4770k on Asus Maximus VI Hero with 1603 bios, EVGA 1070 SC, 16Gb TridentX 2400. Around 12 runs before the Win10 1803 update, averaging about 106k. After the update, roughly 10 runs average 97k. Post-1803, I used DDU to remove graphics drivers, updated to the newest version, still maintaining an average of 97k. The only idea that came to mind was that Haswell was more affected by the Meltdown/Spectre patches (my bios are a few years old, but the latest on ASUS site for my board doesn’t mention Spectre fixes). Any suggestions?

B
Blureux
Posting Freak
797
03-12-2016, 10:14 PM
#2
Are there any other examples besides benchmarks? What about actual performance in games?
B
Blureux
03-12-2016, 10:14 PM #2

Are there any other examples besides benchmarks? What about actual performance in games?

G
gcosta2
Member
96
03-26-2016, 07:07 PM
#3
That are just standards, what about actual gameplay results? I value the point, because benchmarks don't always match real performance, but here are a few things:
- It's not just about the final number, but how much better or worse it is compared to before and after.
- Second, it might seem a little trivial, but I want to boost my Realbench ranking for personal satisfaction.
I've seen a general slowdown in CPU-heavy tests, with less impact on GPU-heavy ones.
Just curious if others on similar older systems have experienced any specific CPU problems since 1803.
G
gcosta2
03-26-2016, 07:07 PM #3

That are just standards, what about actual gameplay results? I value the point, because benchmarks don't always match real performance, but here are a few things:
- It's not just about the final number, but how much better or worse it is compared to before and after.
- Second, it might seem a little trivial, but I want to boost my Realbench ranking for personal satisfaction.
I've seen a general slowdown in CPU-heavy tests, with less impact on GPU-heavy ones.
Just curious if others on similar older systems have experienced any specific CPU problems since 1803.