I can’t tell you often enough that the 5800X3D DOES NOT perform like that in a real world scenario. Like mentioned before, it is WAY behind a 13900K. So don’t interpret too much into Benchmakrs like that.
I did the same “Benchmark” that @xxYUNIORxx did with my 5800X3D and got roughly 70% compared to his 7950X3D results. So, I would argue it is much more of an improvement than just 15%.
Can you share a bit more about your settings? I’m in VR (Index) with a new 7950x3d and my gpu (4080) is about half the number of my cpu timings. Cpu timings were between 24-28 ms in the Fenix.
Specifically windows 11 or 10, HAGS on or off, DX 11 or 12, and anything else you can maybe think of. I just learned from reading here that balanced in power settings is better. New to Ryzen and Intel loved more power 
. I’ll also try upping the LODs to see what happens.
I do have all drivers updated with the two specific to the new 7000x3d chips. I haven’t been able to mess around a lot because I’m traveling for work right now, but any tips would be nice for when I get back.
I did the exact same thing that you did and noticed the exact same performance uplift
I was just getting ready to ask you if you got your setup sorted out.
How you like it so far? It’s like you in MSFS heaven now huh? Lol
Are you also on a 4090?
So for all of you wondering about the 7900X3D in comparison to the 5800X3D.
I think it’s definitely worth the switch. I created several test scenarios for SU11 and used them to test the difference between the two generations. I couldn’t get my hands on a 7950X3D so I took the little brother and feared - since it only has 6 cores with extra cache instead of 8 - that it wouldn’t be much of an improvement compared to the 5800X3D. Still I took the plunge and was pleasantly surprised
I tested with Ultra settings at 1080p. Flying over PG and normal landscape, as well as sitting on the ground. Didn’t change anything about the LOD values and tested with the TBM and an empty Community folder.
To cut a long story short: FPS were at least 30% higher across the board.
Can’t give you any screenshots this time, since I did it only for my own curiosity. With the 7950X3D being so elusive and the 7800X3D still several weeks away (and probably similarly popular than the 7950X3D) for me the 7900X3D turned out a good solution.
Congrats dude. I’ve found that not only does it increase performance significantly, but also makes everything smoother with no stutters.
Using MSFS right now feels like heaven.
Thanks. So you also opted for the 7900X3D
About smoothness: I haven’t flown very much. Still trying to find DIMM modules that work well with my new mainboard.
No, I was able to get the 7950X3D on release day. It’s made MSFS and flying Fenix A320 smooth with no stutters and high FPS. No need for frame generation or DX12.
congrats on your luck. 
This is definitely promising thanks for this as I’ll be coming from a 13700k, I also returned my 7900XT to grab an XTX for practically MSRP(saw a price drop) I’m definitely stoked, just a waiting game now
I can tell you one thing. The 7950x3D is still not enough to bottleneck the 4090.
My testings are always the same settings in the same conditions.
Handcrafted airport, Fenix A320 (set to READY), clear skies, all ULTRA (airport life and traffic at 35 to simulate online vatsim situation / LOD 200) and native 4K (Dx 11).
Ryzen 7950x & 7900XTX → 74.0 fps 13.52ms
Ryzen 7950x & RTX 4090 → 83.2 fps 12.23ms
Ryzen 7950x3D & RTX 4090 → 111.9 fps 8.94ms
In all these situations, I was CPU limited due to the LOD and some resources I can afford.
Mind that these results are all on the ground with all those cpu bound settings. After take off and CRZ the fps go all the way up (~30fps more). At the point that I don’t even care about them anymore xD
If you are interested in the 7950x3D behaviour:
The cores used by the simulator are the CCD 0 (v-cache) and the CCD 1 are parked. But, as we add more programs running in the background (vPilot, Acars, SPADnext, Simlink etc) these parked cores get activated and used for these programs. I see no FPS penalizations.
In my opinion the 7950x3d is the best CPU for MSFS at the moment.
Also, for MSFS, as a hard CPU bounded software we need 4K benchmarks (for those online reviews) with high end GPUs (to avoid gpu bottleneck). You can see there is a substantial difference in fps compared to the non 3d cpu. Around 30% performance increase.
Interesting. In my tests, I’m averaging 80-88 fps with Fenix A320 and same settings as you, except I’m at 3440x1440p which I’m assuming would give me more fps than your results.
In what airport did you do these tests to get 111 fps?
Any other tweaks that I might be missing to get 100+ fps? Windows settings or BIOS settings?
My test was done in LVFR KMIA.
It’s done in LVFR’s LEMD, stand 348. Is not a very optimized airport.
Not many tweaks actually. Windows up to date. Latest nvidia driver (ncp controlled by application), chipset driver and bios. Windows set to default (a few tweaks in registry but nothing important or relevant for games)
Something I have always done is to disable the cpu igpu in the BIOS.
PBO enabled with disabled limits (cpu control) and negative 30 curve except -27 -26 in preferred and best cores for both CCD. Ram is 6000 CL30.
I read somewhere about the the performance impact that the CL from 5200 and 6000 modules may have in fps depending of their capacity to OC. Numbers varying until 17% from CL 36 to CL30 at 6000mhz in some games (over 6000mhz not so much of a difference for zen 4).
Fenix set to quality and render on gpu (no multithreading).
But, those fps numbers are in a particular stand with all started but not moving around more than the cameras. I set it like that to reduce any possible variations from the sim and check the hardware differences.
In real flight situation with all loaded and all my background software I could lose around 5-10fps while taxiing and take off roll and moving my view around.
Also try to keep your fenix copy up to date. If you see anything weird do a clean reinstall. I remember one day last year after loaded in it gave me hard stutters and like 30fps for no reason. Fresh reinstall and everything worked like always.
Hi all. Not sure where to post this as there are so many threads about this CPU now, but I’m needing a bit of a “sanity check” on this spec if anyone can please take a look before I order (whole new system). Does it look fine?
Hello. I have a more or less similar configuration but I’m still waiting for my 7950x3d for pre-order 
That’s my opinion, but I will directly put a power supply of minimum 1200. I bought 1500. 1000 seems little to me (RTX4000 and AMD7000) and if you change after a more efficient component, you will not have to change the power supply and redo all the cables… 
1000w is fine for that setup - even with a 4090.
Not sure what you mean here. You should never reuse PSU cables when replacing a PSU. This is a very good way to fry yoir entire build.
Here is the update of my configuration:
• Corsair 5000D airflow (7 rgb fans)
• Asus TUF GAMING X670E-PLUS
• AMD 7950X3D
• Psu Corsair HX1500i 360mm (3 rgb fans)
• Corsair DDR5 Vengeance RGB 2x32GB 6000
• Asus GeForce RTX 4090 TUF
• SSD M.2 1T (It’s my old one but I’m waiting for the release of the pcie 5 (2x faster) to change and switch to 2T)
• Corsair iCUE H150i RGB PRO XT
Sorry I’m French 
Yes, you have to change the cables with the new power supply. I meant, if you already buy a higher power supply, later you don’t have to change all the cables 
That’s the biggest job.
Thanks for the reply.
I only went with 4 fans. Can you even fit 7 in that case? 
I did change to 1200w - for £60 extra and it is Platinum so gives some headroom if I add more drives.
Guy on phone said getting 6000Mhz stable is still potential problem so he recommended the Dominator 5600Mhz. I think it’s enough for me!
I placed the order so we will see - so surprised to find all the items I wanted in stock so I jumped on it!
