BANANA Benchmark Results Comparison

yep that will be in one my upcoming projects; to build a full AMD only build.
I still have not confirmed the parts, but might be looking at 7950X3D or 7800X3D, both have big performance differences. (from physics etc)The GPU will be the RX 7900 XTX as it matches the 4080.
when that time comes, I will post something on the forum or on reddit.

 

Yes please! 7950x3d is the price of a 13900k at this current time and has longer lasting motherboard compatability. So I HOPE it does better but with what i seen (for Now) 13900k is currently the winner. But we will see what the 7950x3d or even 7800x3d can do.

 

Laptop Dell G15 5511 i7 11800H + undervolt + turbo up to 4.6ghz

Been using this laptop since 2021 to play BeamMP, trying to upgrade to a laptop with 13th gen, but the prices here...

Also @Alex [ITA] I saw that there's a 7700X with "26" threads on the spreadsheet, I believe it should be 16?

 

Anybody know the % of realtime to FPS Translation? Since I'm still having trouble understanding it. May of asked this before and forgot but just asking.

 

There is no direct translation, as the % of realtime only shows the performance of the physics engine. If it's less than 100% the engine will be sluggish and the game physics may not be realistic.
That said, you can sort of guess how many cars a pc can run at a playable framerate by looking at the number of (physical) cores of the cpu: usually beam uses one core per vehicle (simplified cars are obviously lighter to run) so a 4 core cpu will be able to run 4 vehicles smoothly.
I'm not sure how well this comparison holds up with performance/efficiency cores on modern Intel cpus.

 

yeah banana bench doesnt tell the full story of performance in game, be it with cars moving or multiple cars parked, getting crumpled, so on so forth.. so many tests has to be done

 

Ah well that sucks. Maybe in the future it will be different but I'm guessing not any time soon.
--- Post updated ---

Huh makes sense. Though i heard that was with threads too or something? Idk But apperently intels Hybrid core are great but then just drop off in performance somewhere near their thread limit. Saw a video with a 12900K benchmark. Did 20 cars at around 60-50+ Fps but then when 30 cars spawned it was pain. Looked like 30 cars was around 15 - 25 fps. So even if 20 cars are better on 12900K i think the 5950x has a more stable fps since it seemed to do better at 30 cars.

That also makes me wonder. Has anybody tested with the Intel I9 10980XE. For some reason i just feel like it can do really good. Would it beat something like the 5950x or 12900K in this since the 18 P core and 36 thread design seem pretty strong even if its a lot older

 

unfortunately might not be as powerful even with that much core I'm just grabbing the number from 3dmark timespy, the gpu cant be fully utilized for like even 4080.

 

Ahhh that sucks. I really like the 10980XE too. even if its overpriced. But im guessing its not going to beat out the 5950x even if i overclocked huh

 

Hey you seem to know a lot about beamng. Do you know if DDR5 Increases FPS in beamng? Like would DDR5 6000 work better than lets say DDR4 4000 or even 4400?

 

I’ll have to research a bit, please give me roughly a week or so. I’ll try to get back to you asap

 

Wow thanks! Since im thinking the price i get 64gb of ddr5 i can get 128 of ddr4 so its cheaper for ddr4

 

Hey I do have to get back to you:
I did ask someone on the beamcord (thank you Excalibras and CK for assisting me!), We both have the same cpu (i5-13600KF), cooler(Arctic Liquid Freezer II 360mm), same ram size but different platform (ddr4 and ddr5)
There is negligible difference in ddr4 and ddr5 for the most part (besides the usual performance uplift it brings for either ram on a case by case basis, look for higher speeds and lower latency (cl))
Excalibras: (2x16GB DDR4 3600Mhz CL18)

mine: (2x16GB DDR5 6400Mhz CL32)

I hope you find this useful! if you find something interesting, please do let me know

 

Wow. While it (May) just be silicon lottery. It looks like DDR4 may be slightly better which i didn't expect at all. Probably due to cl18. But thanks a lot. But i do agree with Car_Killer. A real test would probably be much more reliable since Banana Benchmark seems to be unreliable at times with some information. And i do want to see those 1% and 0.1% lows since ddr5 does seem to improve them most of the time.

But i may just get ddr4 if they end up being pretty much the same

 

I ran the test as CK said, and it might be a little more interesting:
So during testing, when I finish loading 57 traffic simplified vehicles, I run the the test at the checkpoint area, onto the highway, reason being that the default area might the only area that is heavy; not really representative of actual gameplay,
tested on other areas and its better so I decided to pick the highway is physics movement is constant while most of them will be rendered, likewise, geometry isnt too heavy and pretty representative of the performance in wcusa.
Initial test was worse stuttering and lagging from 11 to 26 fps, and constantly stuttering. respawning the vehicle alleviates this issue. likewise stuttering or lagging does not occur as often while driving along with the same traffic.
WCUSA @ 1440p Ultra + Dynamic Reflections (1024p, update rate 2, 0.8 details, 256 dist, usually it would be at 500)
1 normal vehicle + 57 simplified vehicles without police at 26.7 fps avg, 15.8fps 1% low, 6.9 fps 0.1% low, pretty rough pacing but alot better compared to initial run. Anymore cars added will lose the fps pacing and may become very distracting and or unsatisfactory

 

Alright. Thanks a lot.

Guessing DDR5 Does help on those 1% May as well get some DDR5 5600 CL40 since im able to get 96gb since the difference between this and 64gb of 6200 cl32 is not huge so may as well get that extra memory.

 

Hey Your the person who is doing the Vulkan API Right? My 5600XT For some Reason gets less frames In Vulkan than Normal. Im running Safe mode and non Safe mode and even used DDU but get less fps. Is their a fix for this?

 

Forgot to say, im getting 15fps in vulkan which is a lot less