Jump to content

6900XT vs 3080 Ti on DCS, no debate


dureiken
 Share

Recommended Posts

Hi guys

I just changed from 6900XT to 3080Ti because of FPS problem in VR for DCS, and I make the good choice I think.

Here are my results, with @speed-of-heat track, DCS 2.7.7 OB and no mod, no FSR, no shaders, no mirrors :

https://1drv.ms/u/s!Ar2cEt5yMycgm6skZyUPKfQMjm4mKw?e=so7zI5

and settings under DCS 2d and VR :

image.png DCS avant-apres 3080ti-TEST SS.png

image.png

So 3080Ti is about 40% more powerful than 6900XT under DCS.

I have Win 10, 8600k@5.1Ghz, 32GB of DDR4 XMP

I want to add that my 6900XT made extremly good results under benchmarks like 3dmark, superposition, Heaven. I think the card isn't the problem.

If it can help some people to choose their GPU 🙂

Have a nice day

PS : I post this twice in VR and PC forum

image.png


Edited by dureiken
  • Like 6
  • Thanks 1
Link to comment
Share on other sites

37 minutes ago, Lostaviator said:

Well crap I was just thinking of buying a 6900xt.  Have a 2070 now.  Nvidia is more expensive thought the AMD might be the answer.  The frames don't look bad was it only the frame rate that caused the switch?

frame rate compared to 3080ti/3090

and more important is the Motion vector problem

With nvidia I can force MV, and my 45fps is steady with MSAA*4, PD1.2 and high settings in DCS

Link to comment
Share on other sites

This result surprises me, such a big delta.

PC: 9900K/3090, 11700K/2080Ti, 8700K/1080.

Joystick bases: TMWx2, 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: TMWx2, Winwing Super Taurus, Logitech Throttle Quadrant, Realsimulator RS Linear Throttle (soon)

VR: HTC Vive/Pro/Pro 2, Oculus Rift/Quest 2, Valve Index

Link to comment
Share on other sites

4 minutes ago, Supmua said:

This result surprises me, such a big delta.

Was surprised too, but a friend with 6900XT and reverb G2 have nearly same results than me with same resolution and DCS settings.

If someone else have a 6900XT and Reverb, we could try

Link to comment
Share on other sites

Very interesting. When the new generation cards came out me and my brother made a lot of testing and came to the conclusion that AMD was way faster.
Id like to test but i dont have the F18 module.

 

Also what are your PC specs?

 

 


Edited by JayRoc

FC3 | F-14A/B | F-16C | SA342  R5 5600X | 32GB | RX 6800XT | TM Cougar | T-Rudder mk.IV | HP Reverb G2

Thrustmaster Cougar Throttle USB standalone MOD

Link to comment
Share on other sites

16 hours ago, dureiken said:

Was surprised too, but a friend with 6900XT and reverb G2 have nearly same results than me with same resolution and DCS settings.

If someone else have a 6900XT and Reverb, we could try

I like doing these types of comparison myself but have no real desire to switch camp, especially now that I’m getting Varjo Aero which doesn’t support AMD GPU.

PC: 9900K/3090, 11700K/2080Ti, 8700K/1080.

Joystick bases: TMWx2, 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: TMWx2, Winwing Super Taurus, Logitech Throttle Quadrant, Realsimulator RS Linear Throttle (soon)

VR: HTC Vive/Pro/Pro 2, Oculus Rift/Quest 2, Valve Index

Link to comment
Share on other sites

Hello. Just to add this test may not represent what you will see in real DCS gameplay.  When I have 80 fps when playing the track, I have 45 fps in a real mission., with CPU and GPU at very low usage. So I'm not sure changing your GPU will change your life...

See here : https://forums.eagle.ru/topic/207154-3dmigoto-vr-mod-for-dcs-label-masking-color-enhancement-sharpen-fxaa-copypaste-of-radio-msg/?do=findComment&comment=4810959

 

P8700K @4.8 GHz, 1080ti, 16 GB RAM, HP reverb Pro.

I spend my time making 3dmigoto VR mods for BoS and DCS instead of flying, see https://www.patreon.com/lefuneste

Link to comment
Share on other sites

5 hours ago, JayRoc said:

Very interesting. When the new generation cards came out me and my brother made a lot of testing and came to the conclusion that AMD was way faster.
Id like to test but i dont have the F18 module.

 

Also what are your PC specs?

 

He's using a 3080ti, not a 3080.  That's probably the difference.  😎

ASRock Z370 Professional Gaming i7, Intel i7 8700K Coffee Lake processor @4.7GHz, Corsair H105 Liquid Cooler, G.SKILL Ripjaws V Series 64GB DDR4 3400 memory, EVGA RTX3090 FTW3 Ultra 24GB video card, Samsung 840PRO 256MB SSD for Windows 10 64-bit OS, Samsung 850PRO 2TB SSD for program files, LG WH14NS40 Blu-Ray burner. HOTAS Warthog, Saitek Pedals, HP Reverb G2. Partridge and pear tree pending. pilotfly.gif

Link to comment
Share on other sites

6 hours ago, JayRoc said:

yeah but it still seems too slow for a 6900xt. And as i said, to draw conclusions someone would need more Infos.

All of his hardware info is right there in his post, as well as the track file that he used for his testing.  There are even screenshots of his DCS graphics settings and what mods he does or doesn't use.  He was comparing two different cards, in exactly the same rig, with everything else being the same.  You can't get much better than that for a true, apples to apples comparison between the cards.

ASRock Z370 Professional Gaming i7, Intel i7 8700K Coffee Lake processor @4.7GHz, Corsair H105 Liquid Cooler, G.SKILL Ripjaws V Series 64GB DDR4 3400 memory, EVGA RTX3090 FTW3 Ultra 24GB video card, Samsung 840PRO 256MB SSD for Windows 10 64-bit OS, Samsung 850PRO 2TB SSD for program files, LG WH14NS40 Blu-Ray burner. HOTAS Warthog, Saitek Pedals, HP Reverb G2. Partridge and pear tree pending. pilotfly.gif

Link to comment
Share on other sites

6 minutes ago, eaglecash867 said:

All of his hardware info is right there in his post, as well as the track file that he used for his testing.  There are even screenshots of his DCS graphics settings and what mods he does or doesn't use.  He was comparing two different cards, in exactly the same rig, with everything else being the same.  You can't get much better than that for a true, apples to apples comparison between the cards.

Edited 3 hours ago by dureiken <--- this does mean something to you mr. smart, right?

image.png


Edited by JayRoc

FC3 | F-14A/B | F-16C | SA342  R5 5600X | 32GB | RX 6800XT | TM Cougar | T-Rudder mk.IV | HP Reverb G2

Thrustmaster Cougar Throttle USB standalone MOD

Link to comment
Share on other sites

1 hour ago, JayRoc said:

Edited 3 hours ago by dureiken <--- this does mean something to you mr. smart, right?

image.png

 

Yes, it means that he edited his post.  That's all it means.  Not sure what posting a quote from a thread other than this one has to do with anything.  Anyway, just pointing you to the specs in his post.  Sometimes people miss stuff.  Not sure why you're getting offended.

ASRock Z370 Professional Gaming i7, Intel i7 8700K Coffee Lake processor @4.7GHz, Corsair H105 Liquid Cooler, G.SKILL Ripjaws V Series 64GB DDR4 3400 memory, EVGA RTX3090 FTW3 Ultra 24GB video card, Samsung 840PRO 256MB SSD for Windows 10 64-bit OS, Samsung 850PRO 2TB SSD for program files, LG WH14NS40 Blu-Ray burner. HOTAS Warthog, Saitek Pedals, HP Reverb G2. Partridge and pear tree pending. pilotfly.gif

Link to comment
Share on other sites

Hi, don't make a drama 🙂 I add my rig spec in a post and in OP 🙂

 

but information about track were already there. And yes I think it's relevant because all my rig is the same except the GPU, just 10 min of difference between tests.

 

And my 6900XT beats 3080TI on a lot of benchmarks

Link to comment
Share on other sites

  • 2 weeks later...

Dureiken can you share your steam vr settings?  I see your game setting at the top.  I went with the 3080 ti and i'm trying to optimize.  I see some artifacts if I use motion smoothing but good frames so trying to figure that out.  Thanks  Apologies if i missed your steam settings some where. 

Link to comment
Share on other sites

On 10/29/2021 at 6:09 AM, dureiken said:

Hi guys

I just changed from 6900XT to 3080Ti because of FPS problem in VR for DCS, and I make the good choice I think.

Here are my results, with @speed-of-heat track, DCS 2.7.7 OB and no mod, no FSR, no shaders, no mirrors :

https://1drv.ms/u/s!Ar2cEt5yMycgm6skZyUPKfQMjm4mKw?e=so7zI5

and settings under DCS 2d and VR :

image.png DCS avant-apres 3080ti-TEST SS.png

image.png

So 3080Ti is about 40% more powerful than 6900XT under DCS.

I have Win 10, 8600k@5.1Ghz, 32GB of DDR4 XMP

I want to add that my 6900XT made extremly good results under benchmarks like 3dmark, superposition, Heaven. I think the card isn't the problem.

If it can help some people to choose their GPU 🙂

Have a nice day

PS : I post this twice in VR and PC forum

image.png

 

For any of these tests were you forcing DX11 mode?  The AMD cards appear to be sensitive to this setting with WMR in DCS.

R5-5600X @4.8GHz, 32GB DDR4@3400, 6800 xt, Reverb G2, Gunfighter/TMWH

Link to comment
Share on other sites

These kinds of posts worried me because I have a G2 and was planning on getting a 6900 XT. I decided to get the 6900 XT anyway. I don't have a 3080 to directly compare fps, but I have not having any problems using motion smoothing: generally stable fps and no artifacts. Compared to the 1080 I have been using for years, I can now choose between a lower fps with insanely high quality (60 fps motion smoothed down to 30 fps with the steam VR slider set to 110% and MSAAx2 in-game) or keep the same or better quality I had before and shoot for 45 fps in multiplayer or even turn off the forced motion smoothing and go for 90 fps in Nevada type missions where there are few clouds.

When I first go the new gpu, I was having performance issues that resembled the complaints, but I found that I had accidently let the Winwing software creep into my export lua while doing firmware updates a while ago. As soon as I commented out the Winwing exports, my performance and stability shot up to match what I had expected to see. However, when using higher quality settings, my VRAM gets pegged to 16 Gb, so I would still be interested in a 3090 to get more bandwidth (faster RAM and a bigger bus) and capacity. Will DCS code chew up all the VRAM on a 3090?

The 3080 and even the 3080 Ti seem to have too little RAM for DCS. My 1080's 8 Gb was always pegged and that was using SteamVR at 52% (roughly 1:1 pixel density) and at most MSAA x 2 and a PD of 1.0 in-game.

[sIGPIC][/sIGPIC]

Link to comment
Share on other sites

 Share

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...