April 18, 20251 yr I found this method pretty interesting and logical. Have any of you tried it already? Many people in the comments are reporting a very smooth experience after implementing it. https://youtu.be/651r5ZtJOuE Private Pilot | Windsor Flying Club | CYQG Intel i5 14400F | Gigabyte H610M | Manli RTX 3090 24GB | Kingston DDR4 32GB | DarkFlash 800W | Win 11 | MSFS2020&24 | Lossless Scaling & AutoFPS |
April 18, 20251 yr 5950x3d 5.4-5.7 GHz - Asus ROG 870 Crosshair Apex - GSkill Neo 2x 24 Gb 6000 mhz / cas 26 - MSI RTX 5090 Gaming Trio OC - 1x SSD M2 6000 2TB - 1x SSD M2 2800/1800 1Tb - Corsair 5400 case - Corsair 360 liquid cooling set - 3x 75’ TCL tv. 13600 6 cores @ 5.1 GHz / 8 cores @ 4.0 GHz (hypterthreading on) - Asus ROG Strix Gaming D - GSkill Trident 4x Gb 3200 MHz cas 15 - Asus TUF RTX 4080 16 Gb - 1x SSD M2 2800/1800 2TB - 2x Sata 600 SSD 500 Mb - Corsair D4000 Airflow case - NXT Krajen Z63 AIO liquide cooling - FOV : 200 degrees My flightsim vids : https://www.youtube.com/user/fswidesim/videos?shelf_id=0&sort=dd&view=0
April 18, 20251 yr Limiting fps to a fraction of the monitor frequency (and applying FG, if possible) always was the way to achieve smooth frame times. The new setting in the config file might make this easier. I am going to try.
April 18, 20251 yr Tried this briefly this morning before work. Dont see any reduction in vram use during a quick test sitting in Fenix at gate of LEBB. Frame limited doesnt appear to kick in during the initial sim load, my 3080ti is 'screaming' due to the high fps. Before this test I was using RTSS to limit frames but had to set it as the total inc frame gen eg 60 Chris Ibbotson AMD Ryzen 7 7800x3D / MSI AMD MPG x670E Carbon Motherboard / Gigabyte GeForce RTX 5090 Gaming OC 32GB GDDR7 PCI-Express Graphics Card / Corsair DOMINATOR Titanium RGB Grey 64GB 6000MHz AMD EXPO DDR5 / NZXT C1200 Gold ATX 3.1 - Fully Modular Low-Noise PC Gaming Power Supply - 1200 Watts - 80 PLUS Gold / 48” UltraGear™ UHD 4K OLED Gaming Monitor / 40" Philips 4K LED Monitor / Honeycomb Alpha Yoke / Honeycomb Bravo Throttle Quadrant / WinWing Orion Rudder Pedals c/w dampener / WinWing Ursa Minor Airline Joystick / WinWing Airbus MCDU
April 18, 20251 yr Currently testing this "FrameLimiter 30" modification in my C:\Users\<User Name>\AppData\Local\Packages\Microsoft.Limitless_8wekyb3d8bbwe\LocalCache RTX 5070ti, FrameGen x2 (max fps 60 and vsync ON in Nvidia App), DLSS. First flights seems to show that it reduce the FG artefact (massive flickers and stutters) maybe by limiting load on GPU ? https://forums.flightsimulator.com/t/extreme-flickering-and-stuttering-with-frame-generation/678009 Edited April 18, 20251 yr by vbazillio Vincent B. Check my free MSFS sceneries : https://flightsim.to/profile/vbazillio/trending and my hardware configuration.
April 18, 20251 yr I have max fps set to 60 in NCP using the frame gen mod with super DLSS quality and monitor FPS using the nvidia experience overlay. FPS show 60 or just a little below When I turn down max fps to 30 in the NCP I get 30 FPS or just below and a choppy msfs. TrackIR is smooth with the former but horribly jerkey with the latter, so I have to stick with 60 in the NCP. I have noticed that with the latest driver I am automatically using NVIDIA profile K even though I have not set it manually. I'm not on the SU2 beta though I have used DLSS swapper to update to the latest Nvidia and AMD dlls Edited April 18, 20251 yr by cianpars Ryzen 5800X3D, Nvidia RTX5080 - 32 Gig DDR4 RAM, 1TB & 2 TB NVME drives - Windows 11 64 bit MSFS 2024 Premium Deluxe Edition Resolution 2560 x 1440 (32 inch curved monitor)
April 18, 20251 yr 5 hours ago, vbazillio said: First flights seems to show that it reduce the FG artefact (massive flickers and stutters) maybe by limiting load on GPU If there is truly a significant redux in FG artifact that would be music to my ears, but doubtful it's related to limiting load on the GPU because that is already what I do, limited load to 30 or so native FPS and that had no impact on the artifact problem. I stopped using FG of any kind because of this and the 2nd screen judder. That being said if indeed this lock method truly improved artifact I would be better with coping with the judder issue. Noel System: 9900X3D Noctua NH-D15 G2, MSI Pro 650-P WiFi, G.SKILL 64GB (2 x 32GB) 288-Pin PC RAM DDR5 6000, WD NVMe 2Tb x 1, Sabrent NVMe 2Tb x 1, RTX 4090 FE, Corsair RM1000W PSU, Win11 Home, LG Ultra Curved Gsync Ultimate 3440x1440, Phanteks Enthoo Pro Case, TCA Boeing Edition Yoke & TQ, Cessna Trim Wheel, RTSS Framerate Limiter w/ Front Edge Sync. Aircraft used in MSFS 2024: Fenix A320, Aerosoft CRJ, FBW, WT 787X, I-Fly 737 MAX 8, Citation Longitude.
April 18, 20251 yr Yes, Frame generation quality is much much better. No artefacting. I belevie it is because sim is producing stable limited frames on gpu, exactly half from target with FG. gpu is not utilized as much, running cooler also. Ofcourse that is if you limit to something rational that your system can handle. I limit it to 45, and with FG I get 90 smooothh fps on Gsync 360Hz monitor. no stuttering, lower power consumption and better quality. I havent been so excited since FSX days.... 🙂 Ryzen 9 9950X3D, 64Gb DDR5@6000Mhz cl30, PCIe 4 NVMe SSD, RTX4090 GPU
April 18, 20251 yr The only scenario that will show a reduction in artifact is if you are moving from a driver based or external framegen to an in game framegen with motion vectors, which is now available in beta SU2 for AMD cards and possibly older Nvidia GPUs that could not previously take advantage if the 40 series in game framegen. For me personally the suggestion to lock the frames to improve frame pacing and conserve gpu and cpu cycles is not new information. What is new is altering the user config file to do the lock and whether there is any additional benefit to doing this vs using the tried and true external solutions (RTSS, Radeon Chill, or whatever the Nvidia driver based counterpart is to do the lock. The person in the video is asserting that locking the frames in the game gives a VRAM benefit. I am not seeing this vs my preferred method (Radeon Chill) Edited April 18, 20251 yr by Maxis AMD Ryzen 9800X3D/ Asus ROG Strix B650E F Gaming WiFi / Asrock Taichi 9070XT / 32GB G.Skill Trident Z5 Neo DDR5 6000 / 2x ADATA XPG 8200 Pro NVME / Arctic Liquid Freezer II 280 / Seasonic Vertex 1000w PSU / Lian Li LanCool II Mesh Performance / Asus VG34VQL3A / Topping E70 Velvet DAC & L70 Amp /Sennheiser HD660s2 Thrustmaster Boeing Yoke + TCA Sidestick + TFRP Rudders
April 18, 20251 yr 6 minutes ago, Maxis said: The only scenario that will show a reduction in artifact is if you are moving from a driver based or external framegen to an in game framegen with motion vectors, which is now available in beta SU2 for AMD cards and possibly older Nvidia GPUs that could not previously take advantage if the 40 series in game framegen. For me personally the suggestion to lock the frames to improve frame pacing and conserve gpu and cpu cycles is not new information. What is new is altering the user config file to do the lock and whether there is any additional benefit to doing this vs using the tried and true external solutions (RTSS, Radeon Chill, or whatever the Nvidia driver based counterpart is to do the lock. The person in the video is asserting that locking the frames in the game gives a VRAM benefit. I am not seeing this vs my preferred method (Radeon Chill) Person in the video is known for his invalid statemants. But he was close this time 🙂 It is not the same. External solutions will limit frames at the end of pipeline. This is limiting renering of "real frames" by the simulator. VRAM will be the same, the sceen is still the same all the textures and objects are the same, so..... ignore the vram part. Ryzen 9 9950X3D, 64Gb DDR5@6000Mhz cl30, PCIe 4 NVMe SSD, RTX4090 GPU
April 18, 20251 yr 15 minutes ago, Dankovic said: Person in the video is known for his invalid statemants. But he was close this time 🙂 It is not the same. External solutions will limit frames at the end of pipeline. This is limiting renering of "real frames" by the simulator. VRAM will be the same, the sceen is still the same all the textures and objects are the same, so..... ignore the vram part. Fair .. at least you understand the process and thus my skepticism. I appreciate that. But as someone who also likes to see the "graphs" i would have to do some measurements to see if there are any actual measurable benefits or if its just placebo. My hardware configuration is not typical so cannot be fully relied on when i run my own tests (AMD CPU+GPU) so i don't speak up much and do my own thing but i usually test out my options before deciding on what to use. Will run a few tests over the weekend to fully validate the assertions in his video. Edited April 18, 20251 yr by Maxis AMD Ryzen 9800X3D/ Asus ROG Strix B650E F Gaming WiFi / Asrock Taichi 9070XT / 32GB G.Skill Trident Z5 Neo DDR5 6000 / 2x ADATA XPG 8200 Pro NVME / Arctic Liquid Freezer II 280 / Seasonic Vertex 1000w PSU / Lian Li LanCool II Mesh Performance / Asus VG34VQL3A / Topping E70 Velvet DAC & L70 Amp /Sennheiser HD660s2 Thrustmaster Boeing Yoke + TCA Sidestick + TFRP Rudders
April 18, 20251 yr It would be nice if there was a way to smooth out things with FG. I get triple digits but sometimes there is stuttering. Are they saying to also set a dynamic fps target along with enabling FG in sim? | My Liveries | FAA ZMP | PPL ASEL | | Windows 11 | MSI Z690 Tomahawk | 12700K 4.7GHz | MSI RTX 4080 | 64GB 6000 MHz DDR5 | 500GB Samsung 860 Evo SSD | 2x 2TB Samsung 970 Evo M.2 | EVGA 850W Gold | Corsair 5000X | HP G2 (VR) / LG 27" 1440p |
April 18, 20251 yr 24 minutes ago, Dankovic said: Yes, Frame generation quality is much much better. No artefacting. I belevie it is because sim is producing stable limited frames on gpu, exactly half from target with FG. gpu is not utilized as much, running cooler also. Sounds promising. I launched 2024 after the edit to Frame Limit to 30, but when I then enabled Nvidia DLSS Frame Gen in 2024 settings the rate does not double. Why might that be? Noel System: 9900X3D Noctua NH-D15 G2, MSI Pro 650-P WiFi, G.SKILL 64GB (2 x 32GB) 288-Pin PC RAM DDR5 6000, WD NVMe 2Tb x 1, Sabrent NVMe 2Tb x 1, RTX 4090 FE, Corsair RM1000W PSU, Win11 Home, LG Ultra Curved Gsync Ultimate 3440x1440, Phanteks Enthoo Pro Case, TCA Boeing Edition Yoke & TQ, Cessna Trim Wheel, RTSS Framerate Limiter w/ Front Edge Sync. Aircraft used in MSFS 2024: Fenix A320, Aerosoft CRJ, FBW, WT 787X, I-Fly 737 MAX 8, Citation Longitude.
April 18, 20251 yr 11 minutes ago, Maxis said: Fair .. at least you understand the process and thus my skepticism. I appreciate that. But as someone who also likes to see the "graphs" i would have to do some measurements to see if there are any benefits or if its just placebo. My hardware configuration is not typical so cannot be fully relied on when i run my own tests (AMD CPU+GPU) so i don't speak up much and do my own thing but i usually test out my options before deciding on what to use. Will run a few tests over the weekend to fully validate the assertions in his video. I am running a similar set up 9800x3d and 9070xt I tried to limit frames and then frame gen using the Radeon Experience app but it didn’t do anything sticking with FSR3 and in a frame gen Mark CYYZ
April 18, 20251 yr Just now, MarkW said: I am running a similar set up 9800x3d and 9070xt I tried to limit frames and then frame gen using the Radeon Experience app but it didn’t do anything sticking with FSR3 and in game frame gen Edited April 18, 20251 yr by MarkW Mark CYYZ
Create an account or sign in to comment