jparker36 Posted October 15, 2022 Posted October 15, 2022 I've been chasing after VR perfection for a long time, and having installed my 4090 last night I can finally say I've got it. 10850k @ 5ghz 32gb ddr4 3600 RTX 3090, now 4090 Reverb G2 using OpenXR Test scenarios I use: BF109 Channel Bomber Intercept Mission BF109 Channel Low Level Hell Mission AH64 Marianas Free Flight F16 Syria Free Flight I can run 100% locked 45fps ASW with all settings maxed other than shadows on low/flat, MSAA 4x. No drops below 45 on any test scenario. Only downside as anyone who uses ASW knows is that BF109/Spitfire props and AH64 and KA50 rotors suuuuuck using reprojection, so I tried to see how I could get to 90fps. Alternatively, I can run 90fps 95% locked on the same settings, but with shadows and cockpit global illumination off and MSAA at 2x. The drops that do occur are CPU bottlenecks. I will be getting the 7800X3D as soon as AMD releases it early next year to resolve the CPU issue. In order to get to this level I had to turn on FSR @75% resolution, 80% sharpness. I don't really notice any quality degradation. If I were to choose, I do prefer the locked 45 reprojection as it felt smoother without the CPU bottlenecks, but I imagine I'll switch full time to 90fps using FSR once I get a new CPU. In either case, the GPU is never at 100%, only drawing ~300 watts, so for sure the 4090 is the first cpu that can actually pull off high resolution HMDs. With the 3090 I had to turn all shadows off, clouds to low, terrain distance to low, FSR on, msaa 2x and was only able to get 45fps with reprojection, which would often drop to 33 in low level flights such as the f16 syria and ah64 marianas above. 6 4
dburne Posted October 15, 2022 Posted October 15, 2022 Good to know thanks for that feedback. Don B EVGA Z390 Dark MB | i9 9900k CPU @ 5.1 GHz | Gigabyte 4090 OC | 64 GB Corsair Vengeance 3200 MHz CL16 | Corsair H150i Pro Cooler |Virpil CM3 Stick w/ Alpha Prime Grip 200mm ext| Virpil CM3 Throttle | VPC Rotor TCS Base w/ Alpha-L Grip| Point Control V2|Varjo Aero|
Burt Posted October 15, 2022 Posted October 15, 2022 Awesome ! But I’m 2 years with rig and wanna get 2 more so I’ll be ready for the 6090ti and DCS 3.0. I’m locked at 45fps MR and high settings running oxr and it’s the best I’ve seen VR performance ever, well at least when I started, it just keeps getting better and better. 1 ALIENWARE R11 - I9 10900KF @ 5.1 GHz - M.2 NVMe 2TB - RTX3090 - XFURY 64GB -3400 MHz RAM Monitor AW3420DW @ 120Hz - Virpil CM3 Throttle - TM TPR Rudder pedals - Virpil CM2 w/TM Hornet Stick Center - Monstertech Deck Mounts RealSimulator FSSB-R3 Lightning Base w/ F16SRGRH SideStick - VR user / Varjo Aero - Big Thx to mbucchia Start Date April 2020
Supmua Posted October 16, 2022 Posted October 16, 2022 1 hour ago, Burt said: Awesome ! But I’m 2 years with rig and wanna get 2 more so I’ll be ready for the 6090ti and DCS 3.0. I’m locked at 45fps MR and high settings running oxr and it’s the best I’ve seen VR performance ever, well at least when I started, it just keeps getting better and better. The next RTX will prob be in 2 years, and unless NVIDIA can make a big performance jump like this time again expect the usual 30-35% improvement from the 4090. The high-end GPU is getting up to a point that it's not really much of a factor in performance now, it's the CPU that really needs to catch up in single core performance as long as we don't have multi-core DCS engine. 1 2 PC: 5800X3D/4090, 11700K/3090, 9900K/2080Ti. Joystick bases: TMW, VPC WarBRD, MT50CM2, VKB GFII, FSSB R3L Joystick grips: TM (Warthog, F/A-18C), Realsimulator (F-16SGRH, F-18CGRH), VKB (Kosmosima LH, MCG, MCG Pro), VPC MongoosT50-CM2 Throttles: TMW, Winwing Super Taurus, Logitech Throttle Quadrant, Realsimulator Throttle (soon) VR: HTC Vive/Pro, Oculus Rift/Quest 2, Valve Index, Varjo Aero, https://forum.dcs.world/topic/300065-varjo-aero-general-guide-for-new-owners/
Burt Posted October 16, 2022 Posted October 16, 2022 That totally makes sense 1 ALIENWARE R11 - I9 10900KF @ 5.1 GHz - M.2 NVMe 2TB - RTX3090 - XFURY 64GB -3400 MHz RAM Monitor AW3420DW @ 120Hz - Virpil CM3 Throttle - TM TPR Rudder pedals - Virpil CM2 w/TM Hornet Stick Center - Monstertech Deck Mounts RealSimulator FSSB-R3 Lightning Base w/ F16SRGRH SideStick - VR user / Varjo Aero - Big Thx to mbucchia Start Date April 2020
WipeUout Posted October 16, 2022 Posted October 16, 2022 12 hours ago, jparker36 said: Alternatively, I can run 90fps 95% locked on the same settings, but with shadows and cockpit global illumination off and MSAA at 2x. The drops that do occur are CPU bottlenecks. I will be getting the 7800X3D as soon as AMD releases it early next year to resolve the CPU issue. In order to get to this level I had to turn on FSR @75% resolution, 80% sharpness. I don't really notice any quality degradation. The AMD X3D is a good solution but the intel 12th gen also with its high single core speed. The LGA1700 mobos are much cheaper and you can get DDR4 version and re-use your memory... And you will be able to upgrade to 13th gen also. ------------------------------------------------------------------------------------------------------------------------------------------------------------ 9800X3D, RTX 4090, 96GB DDR 5, MSI Tomahawk 870E, Crucial 2TB x 2, TM WARTHOG COMBO + PENDULAR RUDDER PEDALS, THE AMAZING PIMAX 8K X, Sony 5.1 Spks+SubW | DCS, A-10C_II, AH-64D, F-14/15E/16/18, F-86F, AV-8B, M-2000C, SA342, Huey, Spitfire, FC3.
Voyager Posted October 16, 2022 Posted October 16, 2022 @WipeUout I'd be careful of Alder Lake/Raptor Lake with DDR4. The 5800X3D is memory insensitive, but the non-cache chips really want the bandwidth, especially in flight sims. Right now the top performers, excluding Zen 4, are Alder Lake +fast DDR5, 5800X3D, the Alder Lake + fast DDR4. If you want to go DDR4, you're better of going 5800X3D, otherwise the memory is going to end up being your limiting factor. On the AM5 boards, they've all got a 20% early adopter markup right now, because there is literally nothing out to compete with them. That will change once Raptor Lake starts going live. And definitely change once B650's hit shelves. We've already seen specials with AMD offering free DDR5 ram with certain CPUs. That's only going to accelerate as the market goes sideways.
WipeUout Posted October 16, 2022 Posted October 16, 2022 37 minutes ago, Voyager said: If you want to go DDR4, you're better of going 5800X3D, otherwise the memory is going to end up being your limiting factor. Memory speed (DDR4 vs DDR5) does not play a significant role in DCS or most games. If you have infinite money, well go for the DDR5 and 13th gen or AM5. If you are looking at a cost effective solution then re-using your RAM is possible with intel 12th and 13th gen. What counts above anything in DCS(VR) is a fast single core speed and fast GPU. DDR5 vs DDR4 will not make a huge difference. With my i7-12700kf I have a CPU frame below 9 ms most of the time. It will jump over 13 ms on loaded supercarrier deck but once I'm airborne, no problem and no re-projection needed! 1 ------------------------------------------------------------------------------------------------------------------------------------------------------------ 9800X3D, RTX 4090, 96GB DDR 5, MSI Tomahawk 870E, Crucial 2TB x 2, TM WARTHOG COMBO + PENDULAR RUDDER PEDALS, THE AMAZING PIMAX 8K X, Sony 5.1 Spks+SubW | DCS, A-10C_II, AH-64D, F-14/15E/16/18, F-86F, AV-8B, M-2000C, SA342, Huey, Spitfire, FC3.
Voyager Posted October 16, 2022 Posted October 16, 2022 2 hours ago, WipeUout said: Memory speed (DDR4 vs DDR5) does not play a significant role in DCS or most games. If you have infinite money, well go for the DDR5 and 13th gen or AM5. If you are looking at a cost effective solution then re-using your RAM is possible with intel 12th and 13th gen. What counts above anything in DCS(VR) is a fast single core speed and fast GPU. DDR5 vs DDR4 will not make a huge difference. With my i7-12700kf I have a CPU frame below 9 ms most of the time. It will jump over 13 ms on loaded supercarrier deck but once I'm airborne, no problem and no re-projection needed! 32GB of DDR5-4800 is $140 USD. It's faster than DDR4 3600 in pretty much all conditions. Things have changed considerably since this time last year. 1
WipeUout Posted October 16, 2022 Posted October 16, 2022 Ok, but how much for a AM5 mobo compared to an LGA1700? Using your ''already paid for'' DDR4 instead of buying new memory will always be cheaper nevertheless. ------------------------------------------------------------------------------------------------------------------------------------------------------------ 9800X3D, RTX 4090, 96GB DDR 5, MSI Tomahawk 870E, Crucial 2TB x 2, TM WARTHOG COMBO + PENDULAR RUDDER PEDALS, THE AMAZING PIMAX 8K X, Sony 5.1 Spks+SubW | DCS, A-10C_II, AH-64D, F-14/15E/16/18, F-86F, AV-8B, M-2000C, SA342, Huey, Spitfire, FC3.
Voyager Posted October 16, 2022 Posted October 16, 2022 1 hour ago, WipeUout said: Ok, but how much for a AM5 mobo compared to an LGA1700? Using your ''already paid for'' DDR4 instead of buying new memory will always be cheaper nevertheless. Ask me in January. Right now everything AM5 has a 20% markup. The BOM costs are on par with the LGA1700 motherboards, probably less, but the vendors are in full milking mode, at least until Raptor Lake finally shows up, and probably then some. But if you're talking using DDR4, you're still better off getting a 5800X3D with an X570, than even a 13700K with an LGA1700 motherboard. That's that little red bar than Intel is mostly hiding on their charts, and it does not care whether you're using DDR4-4000 or running it at 2366, unlike Alder Lake, which does care about RAM speed. That's where we're at right now: if you want to keep DDR4, then the 5800X3D is the way to go, but if you're aiming for Alder Lake, Raptor Lake or Zen 4, then go DDR5. Buying a new LGA1700 motherboard for DDR4 is just a bad deal right now. 3
DirtyMike0330 Posted October 17, 2022 Posted October 17, 2022 On 10/15/2022 at 2:49 PM, jparker36 said: I've been chasing after VR perfection for a long time, and having installed my 4090 last night I can finally say I've got it. 10850k @ 5ghz 32gb ddr4 3600 RTX 3090, now 4090 Reverb G2 using OpenXR Test scenarios I use: BF109 Channel Bomber Intercept Mission BF109 Channel Low Level Hell Mission AH64 Marianas Free Flight F16 Syria Free Flight I can run 100% locked 45fps ASW with all settings maxed other than shadows on low/flat, MSAA 4x. No drops below 45 on any test scenario. Only downside as anyone who uses ASW knows is that BF109/Spitfire props and AH64 and KA50 rotors suuuuuck using reprojection, so I tried to see how I could get to 90fps. Alternatively, I can run 90fps 95% locked on the same settings, but with shadows and cockpit global illumination off and MSAA at 2x. The drops that do occur are CPU bottlenecks. I will be getting the 7800X3D as soon as AMD releases it early next year to resolve the CPU issue. In order to get to this level I had to turn on FSR @75% resolution, 80% sharpness. I don't really notice any quality degradation. If I were to choose, I do prefer the locked 45 reprojection as it felt smoother without the CPU bottlenecks, but I imagine I'll switch full time to 90fps using FSR once I get a new CPU. In either case, the GPU is never at 100%, only drawing ~300 watts, so for sure the 4090 is the first cpu that can actually pull off high resolution HMDs. With the 3090 I had to turn all shadows off, clouds to low, terrain distance to low, FSR on, msaa 2x and was only able to get 45fps with reprojection, which would often drop to 33 in low level flights such as the f16 syria and ah64 marianas above. I hate to break it to you, but this sounds like exactly what I am running with a 5800X3D and a 3090. All settings high, shadows medium (I haven't tried MSAA 4, usually keep it on 2) and get smooth performance all the time with OpenXR. I would love to be proven wrong, but the 4090 is going to be minimal gains until the multi threading issues are sorted out in these flight sims. "With the 3090 I had to turn all shadows off, clouds to low, terrain distance to low, FSR on, msaa 2x and was only able to get 45fps with reprojection, which would often drop to 33 in low level flights such as the f16 syria and ah64 marianas above." With OpenXR? Why? Again, running all high settings with 3900/5800X3D. Maybe your CPU is the limiting factor there since it is a couple years old at this point? I'm happy you are getting better performance with the 4090, but it is NOT "the answer" for our VR flight sim woes overall. Disclaimer: I don't fly Marianas because it is a POS lol 1 PC: ASUS TUF 4090oc - Ryzen 7950X3D - 32gb DDR5 6000 - Quest Pro Sims: DCS, IL2, MSFS Pilot Skill: Drunk guy from Independence Day RIO Skill: Goose (post neck-break)
Dragon1-1 Posted October 17, 2022 Posted October 17, 2022 30 minutes ago, DirtyMike0330 said: Maybe your CPU is the limiting factor there since it is a couple years old at this point? It doesn't really matter whether it's a couple years old or brand new, progress on single core performance had been slow, and DCS uses only one for everything except audio (which isn't exactly a huge CPU hog). CPU is the bottleneck in almost every practical scenario, particularly with high end GPUs. I'm not touching my PC's configuration until multicore comes out. It's perfectly flyable as it is, and while I'd like more eyecandy, I want to build my new rig to get the most of the new tech. 6
WipeUout Posted October 18, 2022 Posted October 18, 2022 (edited) 22 hours ago, Dragon1-1 said: It doesn't really matter whether it's a couple years old or brand new, progress on single core performance had been slow I would have a different opinion on this since I made the move from a 9900k to an 12700kf. CPU frametime dropped significantly and now I can use my 6900X to its full potential. I think that the 4090 offers a great performance leap but you also need to have a top CPU to give it breathing room. Granted that the actual DCS graphic engine using a single core is a serious limiting factor but you can mitigate it and get 100% out of your GPU by running your mission on a dedicated server, having a fast (or hugely cached) single core CPU and being careful with settings that are CPU taxing such as shadows. Edited October 18, 2022 by WipeUout ------------------------------------------------------------------------------------------------------------------------------------------------------------ 9800X3D, RTX 4090, 96GB DDR 5, MSI Tomahawk 870E, Crucial 2TB x 2, TM WARTHOG COMBO + PENDULAR RUDDER PEDALS, THE AMAZING PIMAX 8K X, Sony 5.1 Spks+SubW | DCS, A-10C_II, AH-64D, F-14/15E/16/18, F-86F, AV-8B, M-2000C, SA342, Huey, Spitfire, FC3.
tomcat_driver Posted October 18, 2022 Posted October 18, 2022 1 hour ago, WipeUout said: I would have a different opinion on this since I made the move from a 9900k to an 12700kf. CPU frametime dropped significantly and now I can use my 6900X to its full potential. I think that the 4090 offers a great performance leap but you also need to have a top CPU to give it breathing room. Granted that the actual DCS graphic engine using a single core is a serious limiting factor but you can mitigate it and get 100% out of your GPU by running your mission on a dedicated server, having a fast (or hugely cached) single core CPU and being careful with settings that are CPU taxing such as shadows. Your argument is true if you're flying solo and just flying. If the variable AI is in place, specially ground units, the situation tends to change considerably, regardless of what CPU you have. The most taxing of all the computing cycles in DCS is the simulation itself. If you press FPS assigned key twice you can get a detailed view of all the computing cycles. 2
Dragon1-1 Posted October 18, 2022 Posted October 18, 2022 3 hours ago, WipeUout said: I would have a different opinion on this since I made the move from a 9900k to an 12700kf. CPU frametime dropped significantly and now I can use my 6900X to its full potential. It's interesting because the specs between the two are very similar. The 4 additional cores on the 12700kf are E-cores, and single core performance on P-cores seems to be exactly the same (both hit 5Ghz on turbo, 9900 has a slightly lower base clock). There doesn't seem to be anything there that would cause a significant performance boost. That said, those two support different sockets, so you likely changed more than just the CPU. Sometimes there's a bottleneck somewhere else, or even something stupid like thermal paste on the old CPU going bad and causing thermal throttling (although it shouldn't happen after such a short time). Or maybe one of those fancy additions that were put into the 12700kf did it.
WipeUout Posted October 18, 2022 Posted October 18, 2022 2 hours ago, Dragon1-1 said: single core performance on P-cores seems to be exactly the same (both hit 5Ghz on turbo, Core clock speed is not core performance. To compare single core performance it is better to use benchmark such as Cinebench. 5 hours ago, tomcat_driver said: If the variable AI is in place, specially ground units, the situation tends to change considerably, regardless of what CPU you have. Not regardless, core performance is not equal for all CPUs. 1 ------------------------------------------------------------------------------------------------------------------------------------------------------------ 9800X3D, RTX 4090, 96GB DDR 5, MSI Tomahawk 870E, Crucial 2TB x 2, TM WARTHOG COMBO + PENDULAR RUDDER PEDALS, THE AMAZING PIMAX 8K X, Sony 5.1 Spks+SubW | DCS, A-10C_II, AH-64D, F-14/15E/16/18, F-86F, AV-8B, M-2000C, SA342, Huey, Spitfire, FC3.
nikoel Posted October 19, 2022 Posted October 19, 2022 (edited) Who has enough money for a RTX 4090 but then worries about a few hundred for DDR5?! Anyway, early next year we will have 7000 series X3D. For the time being 5800X3D is the best performing processor for simulator type games (which are going ham on draw-calls) by a loooooong way over Alderlake/7000 series The 10K chip is choking that 4090 so hard this thread should be labeled NSFW - at this stage unless you're on AM4 already, it's probably better to wait for the 7K-X3D though Edited October 19, 2022 by nikoel 1
Doncho Posted October 19, 2022 Posted October 19, 2022 13 hours ago, WipeUout said: Core clock speed is not core performance. To compare single core performance it is better to use benchmark such as Cinebench. Not regardless, core performance is not equal for all CPUs.
Supmua Posted October 19, 2022 Posted October 19, 2022 (edited) The 5800X3D trounces everything at 1080p, but single and multicore performance for day to day use is abysmal by next gen’s standard. I’m not a part of AMD religion but have one coming my way soon. Will crunch some CPU frametime numbers when the rendering resolution is around 6.5 to 7K pixels. I have some reservation about it pushing my rig from average 45 fps to 90 fps tier in VR but let’s see what happens. Edited October 19, 2022 by Supmua 1 PC: 5800X3D/4090, 11700K/3090, 9900K/2080Ti. Joystick bases: TMW, VPC WarBRD, MT50CM2, VKB GFII, FSSB R3L Joystick grips: TM (Warthog, F/A-18C), Realsimulator (F-16SGRH, F-18CGRH), VKB (Kosmosima LH, MCG, MCG Pro), VPC MongoosT50-CM2 Throttles: TMW, Winwing Super Taurus, Logitech Throttle Quadrant, Realsimulator Throttle (soon) VR: HTC Vive/Pro, Oculus Rift/Quest 2, Valve Index, Varjo Aero, https://forum.dcs.world/topic/300065-varjo-aero-general-guide-for-new-owners/
ralch Posted October 20, 2022 Posted October 20, 2022 On 10/15/2022 at 12:49 PM, jparker36 said: ... I can run 100% locked 45fps ASW with all settings maxed other than shadows on low/flat, MSAA 4x. No drops below 45 on any test scenario. Only downside as anyone who uses ASW knows is that BF109/Spitfire props and AH64 and KA50 rotors suuuuuck using reprojection, so I tried to see how I could get to 90fps. ... What happens if you set 45fps and ASW=OFF? That way you get rid rid of the reprojection issues (props/rotos/etc)?
jparker36 Posted October 21, 2022 Author Posted October 21, 2022 On 10/18/2022 at 10:09 PM, nikoel said: Who has enough money for a RTX 4090 but then worries about a few hundred for DDR5?! Anyway, early next year we will have 7000 series X3D. For the time being 5800X3D is the best performing processor for simulator type games (which are going ham on draw-calls) by a loooooong way over Alderlake/7000 series The 10K chip is choking that 4090 so hard this thread should be labeled NSFW - at this stage unless you're on AM4 already, it's probably better to wait for the 7K-X3D though Yep, I was waiting until raptor lake reviews to decide what to do CPU wise. Since 5800x3d still dominates ACC and MSFS, I can assume it does in DCS as well. So I wait until Q1 for 7800x3d and that will unlock the 4090 in DCS a bit more I think. 20 hours ago, ralch said: What happens if you set 45fps and ASW=OFF? That way you get rid rid of the reprojection issues (props/rotos/etc)? If I turn off reprojection and use those same settings, I get 55-70fps. Not worth it for me since it still has judder when low and looking to the side past trees, and I tend to enjoy CAS AH64 and KA50 missions. One interesting thing is that I've found for whatever reason, steamVR with the G2 is actually much much better in reprojection than openxr now. Rotors and props on ww2 jets (mostly the bf109) no longer make the entire screen a wobbly mess, so my optimal right now is actually 45fps settings with shadows on high using steamvr. Rock solid 45fps and minimal reprojection issues. Looks amazing! 1 1
CptBligh Posted October 21, 2022 Posted October 21, 2022 On 10/18/2022 at 10:37 AM, Dragon1-1 said: It's interesting because the specs between the two are very similar. The 4 additional cores on the 12700kf are E-cores, and single core performance on P-cores seems to be exactly the same (both hit 5Ghz on turbo, 9900 has a slightly lower base clock). There doesn't seem to be anything there that would cause a significant performance boost. That said, those two support different sockets, so you likely changed more than just the CPU. Sometimes there's a bottleneck somewhere else, or even something stupid like thermal paste on the old CPU going bad and causing thermal throttling (although it shouldn't happen after such a short time). Or maybe one of those fancy additions that were put into the 12700kf did it. Single core IPC on the 12700k is ~30% faster than 9900k. CPU frequency comparisons across generations/manufacturers hasn't been adequate for a long time. MSI Z690 Edge | 12700k | 64GB DDR4 3200 | RTX 4080 Super | Varjo Aero
EightyDuce Posted October 22, 2022 Posted October 22, 2022 On 10/18/2022 at 10:37 AM, Dragon1-1 said: It's interesting because the specs between the two are very similar. The 4 additional cores on the 12700kf are E-cores, and single core performance on P-cores seems to be exactly the same (both hit 5Ghz on turbo, 9900 has a slightly lower base clock). There doesn't seem to be anything there that would cause a significant performance boost. That said, those two support different sockets, so you likely changed more than just the CPU. Sometimes there's a bottleneck somewhere else, or even something stupid like thermal paste on the old CPU going bad and causing thermal throttling (although it shouldn't happen after such a short time). Or maybe one of those fancy additions that were put into the 12700kf did it. 12th gen vs 9th gen aren't comparable by frequency alone. 12th gen has significant IPC improvement over 9th gen, so clock for clock, 12700kf will spank 9900 in gaming. Windows 11 23H2| ASUS X670E-F STRIX | AMD 9800X3D@ 5.6Ghz | G.Skill 64Gb DDR5 6200 28-36-36-38 | RTX 4090 undervolted | MSI MPG A1000G PSU | VKB MCG Ultimate + VKB T-Rudders + WH Throttle | HP Reverb G2 Quest 3 + VD
Thinder Posted October 23, 2022 Posted October 23, 2022 (edited) On 10/16/2022 at 8:20 PM, Voyager said: 32GB of DDR5-4800 is $140 USD. It's faster than DDR4 3600 in pretty much all conditions. Things have changed considerably since this time last year. What really matters isn't the RAM frequency, it is what your CPU controller can deal with. In some case, pushing the frequency will result in a performance loss, simply because the controller witll throttle down under load, in the case of Ryzen 5 for example, the maximum is 3200MHz and 4 banks, so instead of fitting 3600MHz RAM, you're better off using a combination of 4 X B.die 3200MHz sticks, 1 bank per stick, whatever capacity, i gained 6.04% at 4K this way. The point is good bounding pays more in term of performances under load than brute force or even overclocking, if this is possible with your combo, motherboard, CPU and DDR5 RAM, by all mean use DDR5, but in some case, the motherboard might not even be supporting it to the full, on top of which, if B.Die isn't available, I personally would pass until I can find a combo with all parts bounding properly. My tip: Check your system for full support (Motherboard and CPU) of any type of RAM you want to upgrade to (DDR 4 or 5), if you really are looking for improved performances at 4K (maximum load), go for B.Die RAM sticks and respect the CPU controller limits, 4 banks maximum and max frequency. In the pic below you can see that because the CPU didn't throttle back and shut the channels, improvments translated to Graphics as well, which is important if you intend to splash the sort of money a 4080 will take to buy, you cut corner with the RAM, you cut performances at 4K. CPU was my actual Ryzen 5 5600X. I don't even know if manufacturers have released B.Die DDR 5 but just in case here is the link. B-Die Finder Edited October 24, 2022 by Thinder Win 11Pro. Corsair RM1000X PSU. ASUS TUF Gaming X570-PLUS [WI-FI], AMD Ryzen 7 5800X 3D, Sapphire Radeon RX 7900 XTX Nitro+ Vapor-X 24GB GDDR6. 32 GB G.SKILL TridentZ RGB Series (4 x 8GB) RAM Cl14 DDR4 3600. Thrustmaster HOTAS WARTHOG Thrustmaster. TWCS Throttle. PICO 4 256GB. WARNING: Message from AMD: Windows Automatic Update may have replaced their driver by one of their own. Check your drivers. M-2000C. Mirage F1. F/A-18C Hornet. F-15C. F-5E Tiger II. MiG-29 "Fulcrum". Avatar: Escadron de Chasse 3/3 Ardennes. Fly like a Maineyak.
Recommended Posts