The question for better performances on Citra seems endless… and now I’m going to ask this, my apologies
But atm I’m looking for a new cpu for my quite old 1155-socket, wavering between the probably most potential (relating to my case) i7-3770K and the minor weaker (generally- and single-core performance-related) i5-3570K one. Please note that running Pokemon Sun/Moon with better performance on Citra has absolute priority, cost factors and a better connection to other/future non-emulator games are secondary.
To come back to the question, the reason why I’m undecided, although the i7-3770K has even better single-core performances than the i5-3570K, is that I’m afraid HT might have a negative effect on Citra’s performance. Furthermore, even if there was the theoretical possibility to deactivate HT, I’m quite certain my stubborn BIOS wouldn’t let me to do so, bc it’s a HP-branded one and the adjustment-capabilities seem very limited. Generally I have a feeling that i7-owners complain more about worse performance than i3/i5-owners, but that’s probably an illusion and not relevant to the question.
My specs are:
CPU: i5-2320 (yet)
GPU: GTX 580
RAM: 8GB DDR3
OS: Win7 Home Premium 64 Bit
MB: Foxconn H-CUPERTINO2 H61uA TX, Socket: LGA 1155/Socket H2 (I know I am not able to overclock with this MB, still I prefer the “K”-series of the cpu)
I know this intention is no massive boost, and I don’t expect this will run Sun/Moon at constantly 60 fps. Nevertheless I hope it’s a step in the right direction.
Any answers or hints are appreciated.
Citra right now is only using 1 core of your CPU, and that’s why single core performance is the one thing you need to be looking at if you want to see how well Citra will run. Hyperthreading is out of question, it doesn’t make any difference.
I will tell that the difference in single core performance between these CPUs are tiny. (2084 for i7-3770K, and 2029 for i5-3570K.)
Hey thank you for your response and I’m glad to see my cpu-improvement will probably have a noticeable effect on Citra’s performance (boosting cpu from i5-2320 which has 1693 points on the benchmark scale to either i5-3570K (2029 points) or i7-3770K (2084 points)).
But, if you say HT doesn’t make any difference in performance, does this also mean no NEGATIVE difference? (I just noticed the title-question could also be misinterpreted into me asking of a positive effect HT could cause to Citra, I’ll change it)
This matter once was asked in this post from reddit:(https://www.reddit.com/r/pcgaming/comments/2hti6m/why_does_disabling_hyperthreading_supposedly_give/)
and Citra was not directly named but merely other single-core using games in general and that’s the reason why I’m asking if Citra is affected in a negative way by HT (which would make sense if 1 core is devided into 2 threads whereas Citra uses mostly/completely 1 thread at all).
That’s why I have a slightly bad feeling about going for the i7-3770K…
No, hyperthreading doesn’t make Citra run better or worse.
Source: myself having a i3-6100 and trying disabling all of the hyperthreaded threads for Citra’s process in Task Manager.
Now that’s an interesting point I also wanted to go for on my first post.
You probably set the affinity for Citra to the usage of CPU0 and CPU2, obviously disabling HT for Citra only since now the “real” 2 cores of your i3 would be the only ones Citra now uses, right?
The concerns i have about this methode is that I don’t trust taskmanager that much it would force windows to (partly) ignore the order it receives from the BIOS which still says “HT enabled”…
So my question would be if you are able to deactivate HT via BIOS
and if you do, would you notice a change of Citra’s performance?
I’m also excited if other ppl might have experience with this.
EDIT: In case of your i3 dual-core cpu, I just thought of the higher risk of a possibly system-instability which might occur if Citra is running on high priority + HT completely disabled via BIOS, since other/important processes might have less space to evade.