I’ve only seen it behind the 5600x in maybe 2 titles over many many reviews. Now that I think about it though I saw not much better fps than my 3900x on my trial character. Tried again this weekend and it went WAY up so probably some optimization going on still…
I’ve actually talked to FamerNexus about this and they said the reason that’s not done in their comparison chart is the common consumer is not going to do it.
This is it right here…
Fixed it for you and yes thats it. Just like on Ryzen not everyone is going to use Dram calc and manually plugin timings. I know I did but most people wont.
Thanks for the long chat on you youtube channel btw
Yep, if it’s not 1 button I don’t do it.
1 Like
Most titles the 5900X will be faster. The rare cases when you need every ounce of CCD affinity performance you might see the 5900x slower (margin of error slower). The use case for that is going to be very rare though.
In Shadowlands I’m seeing every one of my cores equally loaded in Bastion. So even in WoW the 2% perf diff between a 5600x and a 5950x might only exist in griffin runs in old zone (ie perf diff is irrelevant).
I think you’re right, shadowlands is improving CPU usage.
Gonna have to test this once the servers are back up.
Before on my 6/12 chip it was maybe around 30% max.
That should probably be most people’s philosophy in life LOL
RIP my GPU need a new one!
Here’s my usage just doing some quests out in Bastion:
https://i.imgur.com/Q7aF5Le.jpg
1440p Maximum settings, FXAA High, unlimited framerate, only 130fps and pegged at 99% 
CPU Usage seems pretty spikey still, and favors two cores for the heavy lifting
I have noticed this on my 5800x also however on my 10900k it still be utilizing two cores and alternating between cores. However on AMD useage is being spread out evenly across all cores.
See above.
I wonder why this is. Shouldn’t this be a windows task scheduler thing?
Honestly may just be the technology in the bios on AMD. I literally got rid of my 10900k for my 5800x and I am not sorry I did it. The fact that I can do a multi-core cinebench run and pull 6108 on a 8 core when my 10 core i9 10900k OC to 5.2 pulled 6723? Proof of how much better technology AMD is. When video rendering gaming etc etc I literally can’t tell the difference between the two processors except the cores on the 5800x seem to be utilized a lot better and just about every scenario. The fact that this 8-core is performing on par with a 10 core overclocked chip is insane
I’ve been wanting to swap to AMD for years because of everything they are doing technology-wise however Intel always had a substantial lead in gaming so I never did. The 5000 series changed all that so I figured it was time to pull the trigger.
Knowing all this is making me really excited considering as soon as I can get my hands on one I’m swapping to the 5900x and selling my 5800x
I wonder if it is actually using more resources on the AMD chip or is it just a sensor polling issue?
We know Ryzen chips tend to have issues with boosting and idle processes, leading to high temps due to the boost behavior on near-idle processes. This is a well documented thing with people complaining about high idle temps, voltages, etc.
Is it possible that the way the OS schedules things on Ryzen, the sensors report higher utilization even if it’s not really taking place the way we think? Or is the boost behavior somehow leading to higher utilization than necesseary?
For example my 2 high cores on my Intel system bounce around, and the others bounce around a lot between cores as well. Most seem to be in the 15-25% with the two hard hitters being in the 40s.
It wouldn’t be using more resources man instead of only using two cores it’s taking that same amount information and spreading it evenly across all the cores. So instead of having high usage on 2 cores you have really low usage on all the cores. In other words seems like AMD has a much more efficient usage of cores
2 Likes
IMO that’s probably software.
I would want to see how this is different on Linux
2 Likes
It’s probably software engineering done through the BIOS because it did not happen on Intel
Also, there’s no actual idle threads on the Intel system. They’re still low usage, and all being utilized. As seen here they are all 15-25% still, which is low usage, and only high usage on a few.
Well coming from someone who builds gaming rigs for a living man I’m going to tell you that engineering wise AMD is light-years ahead of Intel and Intel better really go back to the drawing board and come back with a worthy contender. Which I don’t doubt they will do they’ve just been dragging their feet for so long because they’ve been the only game in town.
For years AMD had been making false promises that they could never deliver on before ryzen series. I don’t think they were prepared for Dr. LIsa Su. So basically she kind of just cut them with their pants down
1 Like
This is not really a debate. We’re on Skylake++++++++++++ at this point. Which is a decade old design more or less, if you take into account development period.
2 Likes
From an overall perspective, there’s a few components here:
-
Intel has been sitting on their butts for a decade, just slowly improving upon their existing technology.
-
It’s taken AMD a very very long time to catch up in terms of absolute performance, and only with the 4th generation of Ryzen CPUs have they finally been able to beat Intel’s ancient technology (at gaming, anyway).
-
AMD’s return is a win for the consumer, but a testament on how little innovation there has been in the desktop CPU market for the last decade.
2 Likes
Could not have been better stated. In my opinion what you just said is going to make 2022 the most interesting year for processors that we have seen in a long time.
That is when we are going to get complete revamps from both sides. If someone is holding off on an upgrade 2022 is going to be the best time to go all in
1 Like