Does Overclocking CPU Cache/Ring Clock Boost Gaming Performance?

This is a question that has been on my mind for quite some time. How much does gaming performance scale with Intel’s ring clock/cache speed. Modern Intel processors can be tuned and tweaked in various ways. You can overclock your P-cores, E-cores, and your ring clock. I couldn’t find any thorough benchmarks online comparing various titles at different ring frequencies. Therefore, I decided to take matters into my own hands, and benchmarked 13 various games using my Intel Core i9-13900K at different ring/cache configurations to see how much performance scales. Will overclocking the ring provide the user with any tangible benefits or will it be pointless to even touch the cache? You’ll find out the answer to that question in this video!

Buy and RTX 4080 Super HERE!

BUY A 5700X3D HERE:

BUY A 12600KF HERE:

Check out my 13900K Hyperthreading 40 Game Benchmark Vid

Watch my 13700K E Core ON vs OFF 40 Game Benchmark video

Watch my RTX 4090 vs 3090 42 Game Benchmark video

Watch my RTX 4090 at 2GHz video

Buy an RX 7800XT Here:

Support the channel by donating

Want to build a Gaming PC? Use my Amazon Affiliate link below!

Follow me on Twitter

#Intel #GamingCPU #Overclocking

[Psychx_]

I OCed the core and ringbus on my i5-4570K (Haswell) back then. In combination with pushing the RAM from DDR3-1600_CL8 to 2133_CL10 I got a massive bump in gaming performance. While AVG also increased, the improvement of the min/percentile figures was the effect with the most noticable impact on gaming perf. After dialing back core clocks for testing, I noticed that most of the improvement actually came form mem&ring.

AFAIK the stock ringbus clock on modern Intel CPUs is less important, since it's pretty much sufficient for ootb settings and then some, as there isn't as much core-clock OC headroom as it used to be (rising CPU speed by a whole 700-1000Mhz isn't much of a thing these days). If the core clocks could be pushed much further, the importance of the ringbus clock would likely increase proportionally once again.

[cakestapler]

I have a 13600KF and stock ring is currently running at 4500MHz. Considering my UV/OC is running stable right now seems like not much point in jeopardizing that for little or no gain. Thanks for this video.

[BIG_HAMZ]

I wonder if you see better gains with ring clock OC on slower ddr4 platforms

[ErockOnTech]

Unrelated but I could use some help. My intel system is randomly starting to lag hard with mouse movement. I think Tech Yes City covered this quite well. I’m currently using a 14700K. It was smooth for a while. But recently it’s been very laggy at times. Have you heard of this? Any ideas? Also, if you ever want to chat outside of YouTube comments I’m open to that.

[lassekristensen]

Thank you so much for this test and benchmark. From the looks of it the ring L3 OC dosent pay off! Besides it has like 20w or even more in heat penalty. Now it could be cool to see same test in 4K. How does it behave in super resolution. I only play 4K with vsync 120 fps.

[jaynayy]

Good video, confirms what I didn't test myself :)

[GetRekkles]

With CPU oc there is very minimal difference compare to ram tuning. That's where a lot of performance is hiding.

[vottkal]

Hello! I have some small and possibly stupid question: my asus rog bios (Z690) shows "target cache frequency" is 5000 MHz (CPU 13900KF) without me changing anything (all cache settings are on auto), and everywhere I look I have the information that the stock cache speed should be 4.5 GHz, what the hell? Should I force cache frequency to 4.5 GHz for better stability?

[Ked]

You should have tested cpu bound games like valorant, cs2 etc

[cppctek]

Can you do a video on the best timings and settings for the teamgroup 7200 kits? I have a 48gb 7200 kit just like yours and I’m struggling to find the best timings and settings. Running a z690 Maximus hero I can get 7200 stable for the most part using xmp tweaked but in cod it doesn’t like 7200 with these settings so I’ll drop it to 7000. I feel like I’m doing something wrong. I see time and time again people always asking about the timings for this kit yet no ones talking about what they use. It’s like I’m wondering are people stable on this kit with xmp or no?

[TheAdrianpp]

How are you testing ring stability?

With my 12700k if my ring is unstable I'll get hard freezes at idle, but I can pass every stress test under the sun no problem. Prime95 small fft, occt, cinebench etc.

[firexz]

I saw your vid and my I7 12700kf in 5ghz on p core and 3.8ghz on e core but I can overclock ring clock to 4.6ghz on vcore 1.235v low llc I don't know if I have golden sample or not but the cpu took around 210w to 215w on r23 is this should be okay

[Slixbrah]

with my old 9900k i could gain some fps, but my 13700k, barely anything, just a waste of time🤣

[fatedk]

I really like videos like this man good content, helps me validate my own results. nice vid!

[impuls]

Your 5000Mhz OC is UNSTABLE, there is no chance you should get lower fps unless run to run deviance. Retry at 4900Mhz. The shelf you see will sometimes will change with even faster ram(Lower latency). Tuned can be like 56ns while us harcore guys gets like 43Ns. So lower latency on ring combined with super low latency on ram will ofc give a total of lower latency. Thats something every engineer agrees upon. If this affects fps enough is still up for debate. But your oc has to be 100% stable not to get skewed results. Otherwise a very interesting vid!

[weetjewatikwil]

no one test the latency its always fps...fps is not everything!!

[omegax]

My Question ? Do you really need to over clock this spec

[gladysmaria]

Duh everyone knows though with a few registry tweaks you can go from 40 fps to 300 fps.😉

[-Rizecek-]

Overclocking in 2024 is pointless. One gets a 7% difference in performance for a 30% difference in consumption.

[scamdemc]

will overclocking infinity fabric(ryzen 7000/8000/9000) improve performance?