Install Steam
login
|
language
简体中文 (Simplified Chinese)
繁體中文 (Traditional Chinese)
日本語 (Japanese)
한국어 (Korean)
ไทย (Thai)
Български (Bulgarian)
Čeština (Czech)
Dansk (Danish)
Deutsch (German)
Español - España (Spanish - Spain)
Español - Latinoamérica (Spanish - Latin America)
Ελληνικά (Greek)
Français (French)
Italiano (Italian)
Bahasa Indonesia (Indonesian)
Magyar (Hungarian)
Nederlands (Dutch)
Norsk (Norwegian)
Polski (Polish)
Português (Portuguese - Portugal)
Português - Brasil (Portuguese - Brazil)
Română (Romanian)
Русский (Russian)
Suomi (Finnish)
Svenska (Swedish)
Türkçe (Turkish)
Tiếng Việt (Vietnamese)
Українська (Ukrainian)
Report a translation problem
If you don't want to replace the mobo and RAM that limits your CPU options to 11th gen Intel, and that's generally not worth the fuss going from one generation to immediate next.
GPU-wise a 4060 would provide a marginal increase, with a 4070/4070 ti/4070 super being a proper step up. AMD's 7800 XT would be an upgrade too.
You could easily just get a 7800-XT or 4070S and call it a day, better off waiting until AM5 has more options and Intel is bound to change their socket any day now.
and what games you wanting to play....
For a quick validation of your hardware and configuration; download Userbenchmark[www.userbenchmark.com] and run the executable. Share the url of the homepage with the results.
Evaluating the performance of the GPU:
Download the 3DMark Demo from the Steam store and share your score. Post the url your browser opens.
Evaluating the performance of the CPU:
Run Cinebench (either 2024[www.maxon.net], R20 or R23) and share the single- and multicore results.
It's not an issue. The only time CPU X + GPU Y becomes a significant issue is if you're doing something like a 4090 with the lowest end i3. Outside of that the only argument someone can muster is, "better hardware gets better performance", which is a water is wet kind of statement. And if someone is going to be that silly, is their argument anything but the highest end available hardware is pointless? Not a very good argument.
Once you get out of the silly extremes there's a lot of reasonable flexibility, variability and compromise that can be made. Besides if you get something like a 4070 ti, it will be an upgrade over your 3060, so mission accomplished, that's the point of upgrades, improve performance. But if you still want to upgrade your CPU within the next couple of years, you already have a decent GPU to use, so still no real issues.
If you cut corners on a GPU upgrade because you imagine your CPU isn't good enough... well I doubt you'll be able to find some really strong evidence that confirms that feeling.
I meant for you to post the cinebench scores you get after the 10+ minutes.
Just the two numbers for CPU (Single Core) and CPU (Multi Score) and which version of Cinebench you used.
The intend of the various CPU & GPU stresstests; to compare the performance level of your hardware to the expected mean value so we can exclude if a - software, overheating, configuration or whatever - issue persists.
Before discussing possible hardware upgrades, we should at least know your PC's hardware configuration. It is rather careless behaviour to recommend CPU or GPU without the monitor's resolution, refresh rate in mind and if VRR is supported.
From the userbenchmark link you posted in one of your previous threads, your hardware looks roughly like this:
If you run your games at the monitor's native resolution of 3840x2160, your entry- to midlevel GPU should be the limiting factor regarding performance. That contradicts the statement you made about the GPU usage in 'Squad' and that different settings like DLSS do not affect performance in other various games.
Use an on-Screen display to monitor your hardware like MSI Afterburner (+ RTSS). Frequency, usage, temperature and power consumption are metrices to look at.
I highly recommend to use Intel Presentmon[game.intel.com].
It offers a metric called GPUBusy that is (currently) not present in MSI Afterburner. It shows you how much time it took the GPU to render a frame. The information from this metric allows you to analyze f.e. if a stutter is "caused" by the GPU or the CPU(-side).
For example, I'm currently playing the excellent game Nioh and it exhibits the so called "traversal" stutter. These stutters appear when moving from one area to the next at the time when assets are loaded by the gamecode into (video) memory: https://imgur.com/a/PyZPxgK
The light red frametime graph (top) shows a frametime graph, as seen in MSI Afterburner. The hitching in the graph indicates that the successions of the past frames have not been smooth.
The bottom graph depicts the frametime (purple) and GPUBusy (green) graph, measured by Intel Presentmon.
The purple line exhibits (y axis is meaured in milliseconds) stuttering, which is not present in the green graph. Such a behaviour indicates that the stutter (or the limitation in performance) is caused by the CPU and/or the the memory system (and the gamecode).
A more 'powerful' CPU, memory and storage system will not remedy these kind of stutters, but reduce the impact on performance.