I was thinking if i upgrade i might as well go for a MATX board with at least 3nvme slots so i can add nvme storage later on and more than 5 USB type A port in rear IO.

I can't find the gigabyte Z790m aorus elite ax on lazada / shopee so can't compare with the other 2.

i like the 13th gen because the raptor lake cores has way more L2 cache for both the P & E cores compared to alder lake, usually faster performance & cheaper for the 13600k/kf
but i guess i'll wait till newer stock price that follows the new rm4.4 per 1 usd, else just cont use my old pc saja haha

This post has been edited by terradrive: Jan 8 2023, 12:12 PM

125w is the default for base power, PL1.

PL2 is 253w, turbo power, so I set it to 200w for now.

This post has been edited by kmarc: Jan 8 2023, 12:26 PM

Looks like no difference in performance on assassin creed origins whether it is performance mode, standard mode or power saving mode. On HWinfo, max cpu package power was only 160w for performance mode

However, temperature for performance and standard was about the same around 70-85'c with fans noisily ramping up.

But on power saving mode, it was a cool 59-62'c. I like it!


Performance mode


Standard mode


Power saving mode

This post has been edited by kmarc: Jan 8 2023, 12:36 PM

I'm guessing you'll only see some difference in Warzone 2.0

This post has been edited by edmund_yung: Jan 8 2023, 01:15 PM

Issue now is VRM is on fire. Try to add fan to it.

off ecore. or do a ac/dc alignment with your llc + temp/power limit. dat will shave alot.

Missclicked

This post has been edited by Vannus: Jan 8 2023, 03:04 PM

Yeah, I guess. Probably those games that need more CPU power.

Noted. Will do!

Interesting. Will research more on switching off ecores whether it affects anything or not.

The "ac/dc alignment" I think is out of my league. Anyway, did mention earlier that I just want lazy-man methods. Nowadays lazy to tweak the settings too much.

Likely u wont see a difference coz gaming dont consistently stress the CPU to the max. For power test, use application benchmarks; cinebench, blender, furmark has a CPU stress tester

But as long as gaming doesnt hit the max TDP, it should do fine.

Hi Guys, did you all manage to find any article or information which shows the performance vs power consumption scaling of the raptor lake and AMD 7000 series with the power measurement is done physically either from the power outlet at the wall or from the motherboard?

This post has been edited by zzzz52: Jan 11 2023, 11:24 AM

Linus tech tips

Gamer meld

Gamers nexus

Try search on youtube

This post has been edited by adamtayy: Jan 11 2023, 03:00 PM

I tried youtube, example like gamers nexus, they do state that they measure the power from the EPS12V, but they do not perform a power scaling testing.
Linus channel do not do power scaling as well.
Der8bauer do have the power scaling test, but not sure if it is direct measurement or just reading from software like HWInfo.
The closest I found(with my limited time) on google is the PCWorld 13900K testing.

I'm just trying to look at the efficiency comparison between intel and AMD. Example for 2 similar system, one runs the 7950x vs the 13900K, both system power limits adjusted to the very close power draw measured at the wall, which system performs better in certain task at multiple power levels.

It triggered my interest after reading the Anandtech power scaling article:
https://www.anandtech.com/show/17641/lighte...-13900k-7950x/4
It mentions that the settings vs the power draw have some discrepancy.


This post has been edited by zzzz52: Jan 11 2023, 04:05 PM

Not sure what ur trying to ask for, but both Intel 13Gen and AMD 7000 are so much different that apples to apples power consumption just cannot be made. From the base of their MArch construction (monolithic vs chiplet) to silicon process (10nm vs 5nm) to their thermal characteristics (13 Gen has hard PL1 & PL2 power budget, while AMD 7000 has dynamic power limits as long as it hits 95C). Intel CPUs will run faster but not necessarily hotter as long as it can be kept down when hitting the PL2 power limit, unlike AMD 7000 series which will automatically ramp up as high as it can reaching temp limits.

This is a good chart, especially for those who have casing size constraint, to decide which CPU is better for their case/cooler combo.

Not sure if it applies now, but when 12th gen was launched I did some test using Windows 10 and found it weird that my i5-12600K performs not as good as the lower end models.

I then disabled the E cores and the performance went up, result is then as expected.

Didn't FEEL this in i9-12900K probably because it already has a bunch of P cores, or for some reason the E cores didn't impact it that much. No idea.

Yeah, some websites showed that disabling ecores do improve gaming performance but only slightly although they didn't look at thermals when ecores were disabled. I am more interested in thermals. Furthermore, disabling ecores means the pcores have to do those background work.

Anyway, haven't tried disabling ecores yet as I am happy with thermals under the power saving settings for now. If I do try next time, maybe I will leave 2 ecores active to run background tasks......

In my case the difference was huge, not sure if it was Windows 10 issue or game issue.

I haven't face this with 13600K though, it was only 12600K time which was about 10 months ago.

I wished I had gone for 13600k. But oh well....... I guess limiting the power on 13700k for now is still ok as my RTX2080 makes my rig GPU-bound.

12th gen have Intel's Thread Director that is optimized for Windows 11 Task Scheduler. It's not optimized for Windows 10, that causes it to assign task to incorrect cores. After some time later I think MS and Intel have fix it for Windows 10.