Here is it, maybe I forgot something relevant also. (Could you do that with the 14900 to compare?)
9800X3D stock, X870-i
Standard FS20 F-18, initial VC position
Start on default KJFK RWY 31L
10:00 AM local hour,
Vsync on,
Nvidia Low Latency off, Nvidia control panel default
Framerate limit 100% monitor refresh,
1440p 144Hz,
Traffic AI off,
Weather preset few clouds,
online users off,
TAA,
DX11
LODs 200,
all maxed out,
Lens correction off
(Wait some second till CPU is stabilized in the Task manager graphic, after being more maxed loading the flight).
Do you remember to have some cores full load like me with the 13900, and not like the 7/9800x3d?, and like always was normal in my all former intel and in other Flight Simulators, first time I am in AMD.
None of the occupied M.2 slots will limit the primary PCI_E1 graphics card slot. Only the drive in the M2_3 slot will reduce the PCI_E3 slot to x2 speeds. If you select x4 for PCI_E3 in the BIOS, the M2_3 slot will be disabled.
It appears 9950X3D will not have 3D V-Cache on each chiplet, same as 7950X3D.
However, the new generation 3D V-Cache is now located below the processor cores, and that allows better cooling efficiency.
We donāt know yet what the advertised clock speeds will be on 9950X3D. This is where it could potentially bring a gaming performance uplift compared to 9800X3D, thanks to the āreversedā location of the V-Cache compared to Ryzen 7000.
Weāll see whether or not that affects performance in MSFS.
First time seeing this. This is an interesting graph for several reasons. Seeing a 14900K match the 7800X3D in 1% lows is surprising and unexpected, as prior benchmarks for 2020 showed massive gains. It would seem to suggest the performance uplift from X3D chips in MSFS2024 isnāt quite as strong as MSFS2020.
I always speculated that the massive performance gains from having huge L3 cache would probably be less pronounced once the MSFS engine is optimized more in terms of multithreading. Once you start alleviating that CPU bottleneck that was notorious in MSFS2020, that L3 cache becomes less helpful.
However, itās only one benchmark set. It would be interesting to see more trickle in.
Who games at 1920x1080? Show me some results in 4K. I have two systems, both with 4090 and both using 4k, and I know from my own tests the non-3D cache CPU is performing better.
CPU benchmarks are done at 1080p to force a CPU bottleneck. CPU benchmarks at 4K are largely useless, as usually itās a GPU bottleneck controlling performance.
Iām surprised people still donāt understand thisā¦
I do understand the point, itās you thatās missing the point. Itās of no use for those in 4K. Sure if youāre in 1080p, but who is these days? Canāt imagine many.
