Jump to content

NVIDIA 3090 BFGPU announced. Will DCS benefit from this brute force vs 2080ti?


Recommended Posts

I wrote this once before and still think it holds true.

 

Getting a 30xx will only increase max fps.

 

But it will not raise the min fps, since those come from the CPU (System). When testing I had CPU-frametimes of ~10-15ms in light load but up to 50ms in heavy load like MP or missions with lots of units. In such a case the strong GPU wont help nothing.

 

But with the new RTX one can probably use higher SS to get finally rid of those terrible jaggies (aliasing, shimmering).

 

Anyway, Im waiting for RDNA2, which should be competitive in the 3070 and 3080-case (not 3090 though). My 2070S with 225W is already heating up my small office-room quite a bit and going to something like 320W (with the 3080, which would be the logical step for me for the upcoming Reverb G2) would be too much. But RDNA2 should be more energy-efficient, if leakers are correct.

Win 10, I5-9600K @5,0Ghz, RTX2070S, 32GB Ram, 2x 512GB SSD (one .M2 exclusivly for DCS), TrackIRpro 4, TM Warthog, VKB T-Rudder Mk IV

Link to post
Share on other sites
Go for it! Grinding away with that rusty old 2080 Ti has really gotta suck. It's only money, time for an upgrade! Chuck in another 128 GB of RAM while you're at it and you'll really be cooking with gas.

 

(In short, No)

 

AD

 

I agree...as the bottleneck is more on the CPU side of things.

Of course there can be some benefit but not a whole changing difference

Vincent "Virus" DThe

PC: R9 5900x/RTX2080Ti, 64GB RAM.

Joystick bases: Virpil T-50CM2 with 20cm extension (Center)

Joystick grips: Realsimulator F18

Throttles: Winwing Super Taurus

Hardware: Saitek Combat Rudder, 4x Thrustmaster Cougar MFD, Logitech G13, Winwing Panels

VR: Valve Index 

Monitor: Samsung Odyssey G5, TrackIr v5

Link to post
Share on other sites

One more bottleneck is coming too.

 

PCIe 4.0.

 

It seems, that RTX3090 will only be able to use its full potential with PCIe 4.0.

 

Since its rumored that current LGA1200 boards might not support PCIe 4.0 (even RocketLake is not confirmed me thinks), a move to an AMD-based system might also be in order……

 

If a PCIe 3.0 based 3090 with a 10900K would loose say 10% compared to a PCIe 4.0 based one witha Ryzen 3900X and a 3080 looses the expected 15% to the 3090 it could be more economic to get a 3080 and with the saved money invest in a new AMD CPU(Zen3)/board, since overall performance could be quite similar…..

 

But economical reasons will not be the most important for someone considering a RTX3090. Maybe a new mainboard (after RocketLake is released) will do also.

 

[just made a funny typo: the above sentence at first was: But economical reason will not be…..:-)]


Edited by Wali763

Win 10, I5-9600K @5,0Ghz, RTX2070S, 32GB Ram, 2x 512GB SSD (one .M2 exclusivly for DCS), TrackIRpro 4, TM Warthog, VKB T-Rudder Mk IV

Link to post
Share on other sites
FYI everyone:

 

Not DCS (of course... too niche for general gamers) but seven games: CoD, BF5, Assassin's, Just Cause, Anthem, Fortnite. (I have no idea what most of these games are). But anyways.

 

"Executive summary"

4K (as that is all I am interested in as I have a 5K monitor and have a 4K G2 on pre order). On 10900K test system.

 

3080 vs 3070 average +29-31% fsp increase. +29-33% low 1% fps increase.

3090 vs 3070 average -49-51% and 1% low 50-53% increase.

Pretty consistent improvements.

 

3070 supposed to be close to 2080 Ti performance.

 

So £649 for a 3080 is a no brainer for me. £750 extra for a 3090 on top of the asking price of the 3080 very much diminishing returns with only 14-18% extra performance. Call it +£750 for 15% fps as a median across the titles tested in 4K. But you know what people are like with hobbies and money...

 

Watch the video yourself for details.

 

Enjoy.

Pretty much a no brainer for me too. Go with the 3080. In 2 years, the 4080 will be out, and it'll perform better than the 3090, and at a lower cost rather than spending $1,500 for the 3090 up front.

Gigabyte Z390 Aorus Ultra, i7-9700k (4.9ghz all cores), NH-D15 cooler, 64 GB CL-14 3200mhz RAM, RTX3090FE, Soundblaster Z, Yamaha 5.1 A/V Receiver, 2x 512GB Samsung NVMe, 1x 1TB Samsung NVMe, 2TB Samsung SSD, Win 10 Pro, TM Warthog, Virpil WarBRD, MFG Crosswinds, 43" Samsung 4K TV, 21.5 Acer VT touchscreen, TrackIR, JetSeat KW-908, Reverb G2, Wheel Stand Pro Super Warthog, Phanteks Enthoo Pro Full Tower Case, Seasonic Prime 750W Titanium PW Supply, PointCTRL, Buttkicker 2 , K-51 Helicopter Collective Control

Link to post
Share on other sites

FYI I deleted my post because it looks like these are only synthetic benchmarks, not actual tested benchmarks. Plus some of the results from his 2080 Ti vs 3080 videos are not consistent. Possibly fake news. Sorry for misleading / posting this guys.

[sIGPIC][/sIGPIC]

Link to post
Share on other sites
  • 1 month later...

I wonder how 3090 will do it @8k for DCS World.

What do you think, guys ? 20 - 25 fps average, with max settings?

I7 9700K@5.0 GHz | 32 GB DDR4 | Gigabyte Z390 AORUS Ultra | RTX 2080 Ti | 6 TB SSD + 8 TB HDD | 4K Samsung Q90R 55" | VKB MK III PRO L | Virpil Throttle MONGOOST-50 | MFG Crosswind | TrackIR5

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...