Released DX11 and VR Open Beta, Is Here!

Isosceles Kramer said:
That's simply not true. It doesn't matter what content I use, I have lost 30-45 FPS.

EDIT: Just tested, without single setting changed.
SuperGT Nissan (ISI), Matsusaka GP, sitting in pits
DX11: 59 FPS EDIT2: Actually it's 52, not 59.
DX9: 99 FPS
Click to expand...

Matsusaka uses a lot of GPU VRAM and is not built by ISI/S397, and it hasn't been optimized for DX11 either. I would advise to only tests tracks/cars that appear in Studio 397 DX11 workshop, those have been updated for DX11.
 
stonec said:
Matsusaka uses a lot of GPU VRAM and is not built by ISI/S397, and it hasn't been optimized for DX11 either. I would advise to only tests tracks/cars that appear in Studio 397 DX11 workshop, those have been updated for DX11.
Click to expand...
I have 4GB GPU VRAM, plenty for a single car around that track. I tested ISI tracks as well, same result, I will post numbers after LM qualifying is over.
 
More news at 11.

Post-processing set to "none"
ISI SuperGT Nissan
Nola
DX11: 61 FPS
DX9: 133 FPS

Mills
DX11: 57
DX9: 116

Laughable, and pathetic.
 
Isosceles Kramer said:
It can't tell me that because there's nothing to look for within. It's not a rant either. Performance drop is in fact laughable and pathetic.
Click to expand...


re-install rf2 fresh , then run on min settings at first ,
I'd still run dx11 even if your current hardware is much better on running dx9

just looked at your spec, you should be fine on low settings, do put an overclock on your cpu if you have good cooling , i overclock my cpu the difference in performance is night & day
 
Isosceles Kramer said:
More news at 11.

Post-processing set to "none"
ISI SuperGT Nissan
Nola
DX11: 61 FPS
DX9: 133 FPS

Mills
DX11: 57
DX9: 116

Laughable, and pathetic.
Click to expand...

In my effort for trying help you out with your FPS loss, I had to dig deep into historical archives to find information about your hardware.

CPU: Core 2 Quad Q9550@ 2.83 GHz, L2 cache 2x6 mb, fsb 1333 MT/s, released in March 2008
MBO:
PCIe 2.0 x16
Dual-channel DDR2 1200*/1066 MHz

I am sorry to have to say this but, It is obvious to all who has knowledge about hardware that your CPU, ram speed and PCIe 2.0 x16 support is bottleneck to your GTX 960. I would never use anything better then GTX 480-580 with hardware like this and would never dream about to use PP in rFactor 2.

You may find rFactor 2 DX11 Beta laughable and pathetic but in my mind 57 fps is impressive with your specifications.

;)
 
davehenrie said:
Are you overclocking anything? (cpu? video? memory?) If so, at least try with everything set back to default settings.
Click to expand...
Just the CPU, overclocking GPU and memory is, from my experience, usually a waste of one's time as gains are not significant enough.

Flaux said:
Hello @Isosceles Kramer can you please run a perfomance benchmark and send it in? Usually S397 tries to help individuals to get stuff like that ironed out.

Check this thread for more informations. https://forum.studio-397.com/index.php?threads/official-dx9-vs-dx11-performance-difference.55385/
Click to expand...
Thank you, I'll do that.

Ari Antero said:
In my effort for trying help you out with your FPS loss, I had to dig deep into historical archives to find information about your hardware.

CPU: Core 2 Quad Q9550@ 2.83 GHz, L2 cache 2x6 mb, fsb 1333 MT/s, released in March 2008
MBO:
PCIe 2.0 x16
Dual-channel DDR2 1200*/1066 MHz

I am sorry to have to say this but, It is obvious to all who has knowledge about hardware that your CPU, ram speed and PCIe 2.0 x16 support is bottleneck to your GTX 960. I would never use anything better then GTX 480-580 with hardware like this and would never dream about to use PP in rFactor 2.

You may find rFactor 2 DX11 Beta laughable and pathetic but in my mind 57 fps is impressive with your specifications.

;)
Click to expand...
Difference between PCIe 2.0 x16 and 3.0 x16 in games is negligible (on single card systems at least). I'm not looking for a lecture on hardware.

I don't find rF2 DX11 beta laughable and pathetic, I find performance drop (30-70 FPS just by changing one variable) laughable and pathetic, huge difference. And please don't put words in my mouth.
 
Isosceles Kramer said:
Just the CPU, overclocking GPU and memory is, from my experience, usually a waste of one's time as gains are not significant enough.


Thank you, I'll do that.


Difference between PCIe 2.0 x16 and 3.0 x16 in games is negligible (on single card systems at least). I'm not looking for a lecture on hardware.

I don't find rF2 DX11 beta laughable and pathetic, I find performance drop (30-70 FPS just by changing one variable) laughable and pathetic, huge difference. And please don't put words in my mouth.
Click to expand...

Difference between PCIe 2.0 x16 and 3.0 x16 in games is negligible (on single card systems at least). I'm not looking for a lecture on hardware.

I don't find rF2 DX11 beta laughable and pathetic, I find performance drop (30-70 FPS just by changing one variable) laughable and pathetic, huge difference. And please don't put words in my mouth.[/QUOTE]

Forget difference between PCIe 2.0 x16 and 3.0 x16 in other games you are in rFactor 2 now o_O

TechAde has pointed the finger at the PCI-e bus as to the reason a performance disparity exists benchmarking rFactor 2 on identical GPUs.
GPU: EVGA GTX 980 SC OC #1 (1550MHz, Mem 1900MHz, TDP < 88%)

PCIe 2.0 @ x16
Time: 67875ms - Avg: 145.655 - Min: 113 - Max: 176
PCIe 3.0 @ x16
Time: 67627ms - Avg: 187.975 - Min: 153 - Max: 216

29% average increase, 22% max, and the most important ... 35% min increase on a single card.:rolleyes:
 
rF2 is indeed heavy on PCIe, but his midrange GPU should not bottleneck the PCIe 2.0 bus as much. I think it's a combination of CPU, game settings and software (drivers etc.), because I experience 10-15% less FPS in DX11 using the same settings with an old i5-750 & GTX 660 Ti rig. So something is not right for him.
 
Ari Antero said:
Forget difference between PCIe 2.0 x16 and 3.0 x16 in other games you are in rFactor 2 now o_O

TechAde has pointed the finger at the PCI-e bus as to the reason a performance disparity exists benchmarking rFactor 2 on identical GPUs.
GPU: EVGA GTX 980 SC OC #1 (1550MHz, Mem 1900MHz, TDP < 88%)

PCIe 2.0 @ x16
Time: 67875ms - Avg: 145.655 - Min: 113 - Max: 176
PCIe 3.0 @ x16
Time: 67627ms - Avg: 187.975 - Min: 153 - Max: 216

29% average increase, 22% max, and the most important ... 35% min increase on a single card.:rolleyes:
Click to expand...
That's a good info, thanks, rF2 then might be an exception to the rule. I'm not sure how it relates to my problem of losing so much performance just by switching from DX9 to DX11.
 
stonec said:
rF2 is indeed heavy on PCIe, but his midrange GPU should not bottleneck the PCIe 2.0 bus as much. I think it's a combination of CPU, game settings and software (drivers etc.), because I experience 10-15% less FPS in DX11 using the same settings with an old i5-750 & GTX 660 Ti rig. So something is not right for him.
Click to expand...

Just to info. I have tested how much ram has to say in rF2 with 1080Ti and hardware from 2013.
MBO: Z87, CPU: i7-4770k, GTX 1080Ti
Resolution: 5760x1080@ 120hz.

rFactor 2 > DDR3@2666 vs DDR3@1333 .> DDR3@2666 has average increase 13% .
 
Last edited:
Ari Antero said:
Build ID : 1879993 has about same performance as DX9 if original ISI/S397- DX11 content are used, no PP.
Some third party mods and tracks which are not updated to DX11 seem to have low GPU loads which is reason to low fps.
Click to expand...

That's not true. Many of us have much lower performance in DX11. S397 has been working since the release of the beta of DX11, but for the moment is not solved.

Many people have gone back to DX9 or stopped playing in hopes that this problem will be solved. In my case the performance is almost 50% higher in DX9, with other cars even more.

FPS.jpg
 
Isosceles Kramer said:
That's a good info, thanks, rF2 then might be an exception to the rule. I'm not sure how it relates to my problem of losing so much performance just by switching from DX9 to DX11.
Click to expand...

I'm not sure how much of the thread you have read in preparation to using DX11 so I will refer you to this post as it may make a difference if you haven't already followed this important information. https://forum.studio-397.com/index.php?threads/dx11-and-vr-open-beta-is-here.55270/ Hope this helps you out. Also if you have run any of the tracks in DX9 and they get updated to DX11 you have to empty the folders again to remove legacy data.

Having said that I must admit that performance is down slightly on my system while using DX11 but I do have everything on maximum setting.
Sorry.. Sys Spec
MBO Gigabyte H97
i5 4690 3.5g
16g DDR3
GTX970
1920 x 1080
average fps around 98

Pedals Fanatec v2
Wheel Fanatec v2
 
Last edited:
You must log in or register to reply here.
Back
Top