We benchmarked 33 CPUs in Battlefield 6 to see what really drives performance. After testing 43 GPUs last week, it's time to find out which processors keep the action smooth when chaos begins.
Battlefield 6 Benchmark: 33 CPUs Tested in Multiplayer
- Thread starter Steve
- Start date
To fix CPU usage copy and paste this into notepad. Save it as user.cfg and place it into your battlefield 6 directory
Thread.ProcessorCount 8
Thread.MaxProcessorCount 8
Thread.MinFreeProcessorCount 0
Thread.JobThreadPriority 0
GstRender.Thread.MaxProcessorCount 16
Thread.ProcessorCount 8
Thread.MaxProcessorCount 8
Thread.MinFreeProcessorCount 0
Thread.JobThreadPriority 0
GstRender.Thread.MaxProcessorCount 16
Phylop
Posts: 204 +279
Thanks for sharing these settings. I really do appreciate people who take the time to dig into things like this, because sometimes these tweaks can lead to real improvements. That said, I’m always a little skeptical about which ones are truly effective and why they’re needed. Why do you think these particular settings make a difference?
Often when a new game launches, there’s a wave of advice telling players to edit config files, like adjusting core mappings or other low-level settings. I’m not saying those tweaks are useless, because some of them probably do make a real difference, but I’m curious about why certain settings have to be enabled manually in the first place. Even when they do help, I wonder if there are trade-offs we don’t always consider, like gains in one area coming at the cost of stability or performance elsewhere.
What I find most interesting is why this kind of tweaking is necessary at all. Shouldn’t the engine already detect your hardware and configure itself in an optimal way? The developers know the engine better than anyone, so you’d think they would build that into the game by default. I can understand why manual tweaks like this might be helpful on older hardware, where the engine might not be as well-optimized for the CPU, but I often see them suggested even for the latest processors. Why would that be necessary on modern hardware? Shouldn’t the game already be tuned to take full advantage of current CPU designs? It makes me wonder how many of these changes are truly impactful, and how many might just feel that way because of a bit of placebo effect.
maxxcool7421
Posts: 863 +1,235
whateversa
Posts: 306 +356
Excellent article. My i7-9700k runs at 100% on BF6... I think it was also so high with BF2042.
My RTX 3070 is running somewhere below 80% usage at just over 100fps (I limit it to 110fps) with almost all settings set to low or off and DLSS set to performance.
My RTX 3070 is running somewhere below 80% usage at just over 100fps (I limit it to 110fps) with almost all settings set to low or off and DLSS set to performance.
Eh? How did you come to that conclusion?
The 285k is faster then every AMD chip sans the x3d models on low, and only "loses" (within margin of error) to the 9950x on high. The 265k is similarly fast on low, and both have higher 1% lows.........
It's so weird to me that the 9700k is now an "old" CPU. It's performance is still impressive if you consider it is a Skylake core that is nearly 10 years old now.
amstech
Posts: 2,801 +1,963
Fanboys see what suits their projection.
So weird you omitted the 7800x3d... Would have been really curious to see that, especially since it's probably the most popular enthusiast CPU in the world right now.
The 9700K released in October 2018. It was slower than the Ryzen 2700X which was released in April of that same year, six months earlier. When AMD released the 3000 series, the only non-APU within reach was AMD's slowest product in the series, the 3500. The 3600 was significantly faster than the 9700K.
There is absolutely nothing impressive about that chip. In fact, it was probably an early sign that Intel could no longer compete.
The 9700k has significantly better single core performance and better multi core performance in applications that can't utilize hyper threading as well.
The 10700k was better than the 3600x in every way from a performance perspective, aside from power consumption.
Not sure why people love to bash on Intel and praise AMD like either of these corporations care about you as an individual. Intel has never been as uncompetitive as AMD was during the Bulldozer days
PierreJG
Posts: 108 +34
After upgrading my I5 3340 to a Ryzen 5 5500 (sadly, the pc died. Can fix it, just not worth it). Good to see it can still handle some newer games. I am not one for FPS anymore, as way over my prime, other than Borderlands. It is still good to see it can reach 60fps at least, depending on settings.
PS: Battlefield 2 and Call of Duty 2 are still the best FPS in their respective series (My opinion)
PS: Battlefield 2 and Call of Duty 2 are still the best FPS in their respective series (My opinion)
In gaming? You must be crazy, the 9700k was much faster than the 2700x and still is.
In fact according to this review the 9700k is significantly faster in both gaming and applications compared to the 3600 you also brought up. Even an overclocked 3600x falls short. Where are you getting your missinformation from?
AMD Ryzen 5 3600 Review
Ryzen 5 3600 is the most affordable Zen 2 processor in AMD's lineup. At just $200, it offers six cores and twelve threads, yielding a significant advantage in applications against the competition from Intel. Gaming performance is also improved nicely as it is around 10% higher than with previous...
Pastuch
Posts: 166 +199
Terrific content as always Steve, I think your 5800x3d results are lower than expected because you had Smart Access Memory (SAM) enabled. That feature doesn't really work properly with the 5800x3d in most games. I saw a massive performance improvement in BF6 for the 5800x3d when I disabled SAM. I'm running a 7900xtx.
Its absolutely wild how everyone just takes these tests without a single doubt or double check of their own. I just did my own test on my PC with a 13900KS clocked at a locked 5.6/4.4Ghz (which is about stock) with DDR4 at 4000Mhz CL16 and a 4090 and did the EXACT same tests as Steve did and got significantly different results.
On Operation Firestorm 64 players Conquest with Low preset, native TAA 1080p I got Avg 224fps and 1% Lows at 162fps. I though maybe its the map. Tried Liberation Peak, same 64 player Conquest configuration and got Avg 212fps and 1% Lows 154fps. I said ok, lets also test some Overkill preset, just be be sure so I tested on Mirak Valley Conquest 64 players full server 1080p Native TAA and with the Overkill preset I got Avg 182fps and 1% Low 135fps.
Not only are these results better than all his Low 1080p native TAA results and about the same as his top Overkill 1080p native TAA preset, they're a far cry from what he's getting on the intel platforms. I'm assuming this has something to do maybe with the Intel Default profile he's using, which is limiting the CPUs to a ridiculous 125w, which is quite a bit lower than what these older Intel chips need. For reference, on my system, the average CPU wattage was ~165W.
So take these results with a MASSIVE grain of salt because for me, they are straight up BS, at least for the intel platform, and don't come calling me a fanboy, because I'm a massive fan of the x3D CPUs. I only got this KS because back then it was the fastest gaming CPU, and that's what I'm gonna get next time I upgrade, regardless of Intel or AMD. I also have all the data to prove it. I can even record my screen with my phone as I set these exact settings and you can see for yourself how skewed those results are, likely due to the stupid low PL1 of 125w.
On Operation Firestorm 64 players Conquest with Low preset, native TAA 1080p I got Avg 224fps and 1% Lows at 162fps. I though maybe its the map. Tried Liberation Peak, same 64 player Conquest configuration and got Avg 212fps and 1% Lows 154fps. I said ok, lets also test some Overkill preset, just be be sure so I tested on Mirak Valley Conquest 64 players full server 1080p Native TAA and with the Overkill preset I got Avg 182fps and 1% Low 135fps.
Not only are these results better than all his Low 1080p native TAA results and about the same as his top Overkill 1080p native TAA preset, they're a far cry from what he's getting on the intel platforms. I'm assuming this has something to do maybe with the Intel Default profile he's using, which is limiting the CPUs to a ridiculous 125w, which is quite a bit lower than what these older Intel chips need. For reference, on my system, the average CPU wattage was ~165W.
So take these results with a MASSIVE grain of salt because for me, they are straight up BS, at least for the intel platform, and don't come calling me a fanboy, because I'm a massive fan of the x3D CPUs. I only got this KS because back then it was the fastest gaming CPU, and that's what I'm gonna get next time I upgrade, regardless of Intel or AMD. I also have all the data to prove it. I can even record my screen with my phone as I set these exact settings and you can see for yourself how skewed those results are, likely due to the stupid low PL1 of 125w.
I did my own testing on my locked 5.6/4.4 13900KS (about stock - 5.6/4.3) with DDR4 and a 4090 and got Avg 212-224fps and 1% Lows at 154-162fps depending on maps for the Low 1080p TAA preset, and Avg 182fps and 1% Low 135fps for the Overkill preset. So you can take these results with a massive grain of salt.
Why was dude singling out 265k users reporting they had much faster fps than his own testing for this review. I actually did the same tests and got better than the ryzen 9900x3d at high preset 1080p with an Intel 265k, 32gb V-color 8000mhz cl38 cudimm ram, Asus ROG Strix Z890-A mobo, and a Strix RTX 5080 and turned on 200s boost (Yes you're supposed to for better performance, its like not enabling XMP) I was in the middle of conquest using the same maps with 12fps higher 1% lows than the 9800x3d with an average fps 4fps below the 9800x3d. What kinda tards are these "tech reviewers" I even repped ryzen since my first Ryzen 5 2600 rx 580 build, then upgraded to a 3700x and then 5700x with a RX 5700xt before hopping to intel now.. serious lousy
I was worried my CPU had finally reached its limit. Sadly currently not in a position to upgrade my PC
I have an i9-9900k it did struggle to run the game during the BF6 beta without crashing. As in as soon as I would tab out the game would crash. Interestingly enough I can't run Battlefield 2042 due to the same issue
But fortunately EA changed something before launch and now in the launch build the game runs really well for me. I do have a lot of settings turned to low but the CPU is now sitting around 80% utilisation and no more crashes. My RTX 4070 Ti Super is hardly being used though. I'm running the game at 1440p native with most settings off or low and only a few things like textures to high
I tend to average around 100 FPS in most maps it does at times drop to 70 FPS though
I wasn't 100% sure the poor performance was due to the CPU but thanks to these benchmarks I'm now clear that my GPU can likely go higher FPS but is being bottlenecked by the CPU on BF6
Hopefully it is only BF6 for now that is this CPU intensive most other games tend to go more for the GPU and thus my old, but powerful, CPU doesn't cause a bottleneck there
I have an i9-9900k it did struggle to run the game during the BF6 beta without crashing. As in as soon as I would tab out the game would crash. Interestingly enough I can't run Battlefield 2042 due to the same issue
But fortunately EA changed something before launch and now in the launch build the game runs really well for me. I do have a lot of settings turned to low but the CPU is now sitting around 80% utilisation and no more crashes. My RTX 4070 Ti Super is hardly being used though. I'm running the game at 1440p native with most settings off or low and only a few things like textures to high
I tend to average around 100 FPS in most maps it does at times drop to 70 FPS though
I wasn't 100% sure the poor performance was due to the CPU but thanks to these benchmarks I'm now clear that my GPU can likely go higher FPS but is being bottlenecked by the CPU on BF6
Hopefully it is only BF6 for now that is this CPU intensive most other games tend to go more for the GPU and thus my old, but powerful, CPU doesn't cause a bottleneck there
You are indeed severely CPU bottlenecked. Anything over a 1080ti for that CPU would be limited. If you'd upgrade to at least a 5800x3d you would get MUCH better performance, idealy a 7800x3d and then you'd be amazed at how much more you can get out of your 4070ti Super.
AndrettiGTO
Posts: 29 +14
BF6 runs well on my 5950x/RTX3090 so looking at the charts, placing it around the 7700x performance.
Similar threads
Latest posts
-
Ford learned the hard way that AI can't replace experienced engineers- Nobodysfool replied
