Jump to content

Flying near trees causes great fps loss.


OttoPus

Recommended Posts

  • ED Team
I think there should be different shadow options for users to enable or disable it and currently the shadow options are very confusing.

 

What do you mean? There are different options right now... Off, Flat Only, Low, Medium, High, Ultra?

64Sig.png
Forum RulesMy YouTube • My Discord - NineLine#0440• **How to Report a Bug**

1146563203_makefg(6).png.82dab0a01be3a361522f3fff75916ba4.png  80141746_makefg(1).png.6fa028f2fe35222644e87c786da1fabb.png  28661714_makefg(2).png.b3816386a8f83b0cceab6cb43ae2477e.png  389390805_makefg(3).png.bca83a238dd2aaf235ea3ce2873b55bc.png  216757889_makefg(4).png.35cb826069cdae5c1a164a94deaff377.png  1359338181_makefg(5).png.e6135dea01fa097e5d841ee5fb3c2dc5.png

Link to comment
Share on other sites

  • Replies 162
  • Created
  • Last Reply

Top Posters In This Topic

I have done several tests!

OC and non OC -> no change

3 different drivers -> slightly different FPS but the same behavior as before

The interesting part is, that 1.2.6 has the same behavior but not so extrem.

 

With 1.5 I have some more FPS as with 1.2.6 at the start of the quick intercept D-9 mission (1.5 = 125 FPS max versus 1.2.6 97 FPS max).

If I turn 180° to the Forrest, the behaviors of both engines are the same.

The more I close the range to the Forrest, the more I can notice a FPS drop.

If I set both engines at the same game settings, the VRAM use is the same.

All max. like in 1.2.6 the VRAM use is for both engines ~1650 MiB.

If I disable the shadows in 1.5 the loss of FPS is per percent the same as with 1.2.6 with disabled shadows.

So I think the shadow thing is only a point over the real reason... the trees themselves!

Two different engines with the same behavior (more or less), that seems to be a program (Direct3D?) problem.

I would really like to see which VRAM use the nVidia GC have.

Max. 2100 VRAM use is extreme low for a DX11 engine.

And because the behavior is per percent the same with both engines, it seems so that not all is DX11 and there could be a DX9 fallback which is the reason for this behavior.

Looks more like an old behavior in a new dress!

;)


Edited by Nedum

CPU: AMD Ryzen 7950X3D, System-RAM: 64 GB DDR5, GPU: nVidia 4090, Monitor: LG 38" 3840*1600, VR-HMD: Pimax Crystal, OS: Windows 11Pro, 2*2TB Samsung M.2 SSD, HOTAS: TM Warthog, Paddles: MfG.

Link to comment
Share on other sites

I think there should be different shadow options for users to enable or disable it and currently the shadow options are very confusing.

 

I have run many test missions and have come to the conclusion that flat only with flat terrain shadows on looks great with a minimal amount of flashing and maximum FPS.

Link to comment
Share on other sites

I am also experiencing micro stutters regardless if the FPS are steady or not and in all situations. With all possible combinations around Vsync, Nvidia control panel, inspector...I have been trying virtually every advice and combination there is (except the 144Hz Gsync monitor because I do not own one) for about 5 days and the issue just wont go away. My opinion is that the new 1.5 engine still needs more/better optimization and I hope that this will be solved. I expect that with my PC specs DCS runs smoothly...

 

EDIT: Wow! I am blown away!!! I highly suggest trying out this tip...it solved my micro stuttering issue for a good amount! Happy times :)


Edited by cro_mig_21
Link to comment
Share on other sites

After studying the issue made some adjustments (after which I had completely disappeared micro stuttering):

 

In BIOS:

HPET (High Precision Event Timer): Disabled

(In ASUS motherboard BIOS label it as "ACPI 2.0")

 

As administator CMD:

bcdedit /deletevalue useplatformclock

bcdedit /set disabledynamictick yes

 

More info: http://www.overclock.net/t/1487105/want-higher-fps-what-timer-do-you-use#post_22209166


Edited by Alexx78

ASUS Rampage III GENE X58 \Intel i7 920 (3623 MHz OC) \ASUS DirectCU Radeon HD 7970 \3 to 4GB Patriot PSD34G16002 \Enhance ENP-6610GA 1000W \SSD Intel 730 240GB, WD Caviar Black 2TB, Seagate 7200.11 1.5TB \Thermaltake Element G \42'' LG LE8500 (LED-TV) \steel floor-mounted stick \steel pedals AC-1x/FW-190 \Free Track (PS3 Eye) \Windows 10 Pro x64

Link to comment
Share on other sites

I've mentioned that for this "tree problem of low FPS" there must not be many trees. In the BF109 k-4 free flying map even a simple row of 20 trees are enough to slow my rig completely down. From high 100 FPS down too low 20 FPS.

The micro stutter seems to be another problem.

Tried 4 different AMD driver 15.7 up 15.10 and the problem is still there.

I cannot say if this has something to do with the drivers or with the trees but if nV and AMD have the same problem, then I would think it's something with the trees.

A single line of 20 trees and the FPS will be eaten like there is no tomorrow anymore... seems to be a little too much for me.

;)

CPU: AMD Ryzen 7950X3D, System-RAM: 64 GB DDR5, GPU: nVidia 4090, Monitor: LG 38" 3840*1600, VR-HMD: Pimax Crystal, OS: Windows 11Pro, 2*2TB Samsung M.2 SSD, HOTAS: TM Warthog, Paddles: MfG.

Link to comment
Share on other sites

I have done several tests!

OC and non OC -> no change

3 different drivers -> slightly different FPS but the same behavior as before

The interesting part is, that 1.2.6 has the same behavior but not so extrem.

 

With 1.5 I have some more FPS as with 1.2.6 at the start of the quick intercept D-9 mission (1.5 = 125 FPS max versus 1.2.6 97 FPS max).

If I turn 180° to the Forrest, the behaviors of both engines are the same.

The more I close the range to the Forrest, the more I can notice a FPS drop.

If I set both engines at the same game settings, the VRAM use is the same.

All max. like in 1.2.6 the VRAM use is for both engines ~1650 MiB.

If I disable the shadows in 1.5 the loss of FPS is per percent the same as with 1.2.6 with disabled shadows.

So I think the shadow thing is only a point over the real reason... the trees themselves!

Two different engines with the same behavior (more or less), that seems to be a program (Direct3D?) problem.

I would really like to see which VRAM use the nVidia GC have.

Max. 2100 VRAM use is extreme low for a DX11 engine.

And because the behavior is per percent the same with both engines, it seems so that not all is DX11 and there could be a DX9 fallback which is the reason for this behavior.

Looks more like an old behavior in a new dress!

;)

 

I have to say I agree. The engine is like a hybrid of the old with the new. Frame rate drops occur when smoke and explosions occur (old & new engine), trees cause frame rate problems (old & new) and shadows cause frame rate problems (old & new). These issues effect frame rate greatly and so far in the beta have not been addressed. Nvidia Sli as of now is not working properly and I am starting to get worried that the Edge engine will not deliver. I just saying how i feel with the Beta build, still love the game, will always buy modules and play it, just feel confused with what Edge is actually doing.

4.8 I7, 1080, TMW&T, SSD, VKB MK.IV.

Link to comment
Share on other sites

I've mentioned that for this "tree problem of low FPS" there must not be many trees. In the BF109 k-4 free flying map even a simple row of 20 trees are enough to slow my rig completely down. From high 100 FPS down too low 20 FPS.

The micro stutter seems to be another problem.

Tried 4 different AMD driver 15.7 up 15.10 and the problem is still there.

I cannot say if this has something to do with the drivers or with the trees but if nV and AMD have the same problem, then I would think it's something with the trees.

A single line of 20 trees and the FPS will be eaten like there is no tomorrow anymore... seems to be a little too much for me.

;)

 

Yep, was just about to post this.

 

I have reduced the trees to 2000m, and monitored the vram usage.

 

Well in some areas, with some type of trees, it is around 1800mb (so not saturated), and with very few trees in sight I have a tremendous fps drop.

 

There's for sure something wrong with trees (some type of).

Link to comment
Share on other sites

 

EDIT: Wow! I am blown away!!! I highly suggest trying out this tip...it solved my micro stuttering issue for a good amount! Happy times :)

 

How is the input lag after this 'fix'? By input lag I mean the joystick and trackIR responsiveness.

  • Like 1

Current specs: Windows 10 Home 64bit, i5-9600K @ 3.7 Ghz, 32GB DDR4 RAM, 1TB Samsung EVO 860 M.2 SSD, GAINWARD RTX2060 6GB, Oculus Rift S, MS FFB2 Sidewinder + Warthog Throttle Quadrant, Saitek Pro rudder pedals.

Link to comment
Share on other sites

If Mustang's tree shadow removal could be be an option in the settings it would be nice. Tree shadows worked fine in 1.2 but for now it is better to remove them in 1.5.

P-51D | Fw 190D-9 | Bf 109K-4 | Spitfire Mk IX | P-47D | WW2 assets pack | F-86 | Mig-15 | Mig-21 | Mirage 2000C | A-10C II | F-5E | F-16 | F/A-18 | Ka-50 | Combined Arms | FC3 | Nevada | Normandy | Straight of Hormuz | Syria

Link to comment
Share on other sites

If Mustang's tree shadow removal could be be an option in the settings it would be nice. Tree shadows worked fine in 1.2 but for now it is better to remove them in 1.5.

 

I don't think that the shadows are the problem, because even you disable the shadows the problem is still there, but not a so hard FPS hit anymore, but it's still there!

Take the quick free flight mission with the D9. Turn 180° and fly to the lake near the shore. At the sea side there is a big city with many trees inside and on the other side of this lake is the big forrest.

I have around 85+ FPS at the city but as soon as I heading to the forrest, the FPS went down.

It's like many "meshes" are only CPU generated (soft) and not by the GPU (hard). This could explain why the VRAM is most of the time less than 50% filled.

A single row of 20 trees without shadows kills 60% of my FPS (free flight quick mission with the BF109). That has nothing to do with the shadows.

I believe that the "kill shadows mod" will really help but not fix the problem. For me the mod didn't change a bit. The FPS lost behavior is still there.

I would like to have a mod whic´h would change the trees in such a way that they will become a 3D object. So we could see the difference of VRAM usage and FPS drop. Becuase if they are all 3D objects the GPU should use them and should fill the VRAM.

Now it's looks like the drops coming from the CPU tree math and later the CPU must "translate" them to the engine also by software to mix them up with the 3D rednerer. And this slows down my system.

The shadows are no problem for my rig the difference between off and Ultra is like it should be, nothing special. The trees alone are the problem for me.

CPU: AMD Ryzen 7950X3D, System-RAM: 64 GB DDR5, GPU: nVidia 4090, Monitor: LG 38" 3840*1600, VR-HMD: Pimax Crystal, OS: Windows 11Pro, 2*2TB Samsung M.2 SSD, HOTAS: TM Warthog, Paddles: MfG.

Link to comment
Share on other sites

I don´t have this FPS drop at all in the Dora quick mission or the K4 quick mission over this places.

 

I have the Mustang mod installed and my FPS are steady like rock in 60 FPS all around this cities, forests etc.

 

It´s true that the trees shadows on cities are still there but it seems this casted shadows doesnt affect at my FPS at all.

 

Maybe difference between Nvidia and ATI

" You must think in russian.."

[sIGPIC][/sIGPIC]

 

Windows 7 Home Premium-Intel 2500K OC 4.6-SSD Samsung EVO 860- MSI GTX 1080 - 16G RAM - 1920x1080 27´

 

Hotas Rhino X-55-MFG Crosswind Rudder Pedals -Track IR 4

Link to comment
Share on other sites

I don´t have this FPS drop at all in the Dora quick mission or the K4 quick mission over this places.

 

I have the Mustang mod installed and my FPS are steady like rock in 60 FPS all around this cities, forests etc.

 

It´s true that the trees shadows on cities are still there but it seems this casted shadows doesnt affect at my FPS at all.

 

Maybe difference between Nvidia and ATI

 

Maybe your system is costantly rendering higher than 60 fps (idk, like 80-120) and you cannot notice the fps drop.

 

But even if it drops from 120 to 80, there's a 33% of fps lost due very likely to a bug/optimization issue. It means that there are system resources used without reason that can be used to calculate other things (i.e.: IA battles).

 

In my opinion is well worth investigation, if it turns out that is due to the old map though, I'd not bother trying to fix it :)

Link to comment
Share on other sites

shadows are an issue in 1.2.16 as well. In that version if I go from shadows medium to shadows low I see an immediate 15 to 20% improvement. Its a hefty price to pay for shadows when the overall ambience is undiminished, subjectively speaking of course.

Link to comment
Share on other sites

I'm experiencing the same drop with a GTX Titan 6GB. I did some screenshots where the difference is clear:

 

Object count is similar

Frame time is similar

Shadow time is similar

FPS is 83 vs 29 (I expect the value to be over 60 considering the frame time)

 

This makes me think more of a CPU rather than GPU issue.

 

System specs:

i5-3570k @ 4.5 GHz - 8GB DDR3 1600 RAM - Win10 64 - Geforce Titan 6GB

Hotas Warhog - Saitek Combat Rudder Pedals - Cougar MFCDs - Custom head tracker

32" 1440p main display + 22" MFCD/Helios display

256 GB 840Pro SSD - 2Tb Caviar Green

 

No trees:

NoTreesValues.PNG

 

Trees:

TreesValues.PNG

 

Full images are attached to this post.

Screen_151017_183508.thumb.jpg.e6172ab939bb3e157b1dd23b28610cfd.jpg

Screen_151017_184756.thumb.jpg.70c7dbc633301317994cbedbd71dc1dd.jpg


Edited by AMVI_Rider
typo

[sIGPIC][/sIGPIC]

Author of DCSMP and VRK

 

Ryzen 5 3600X - 32GB DD4 3200C14 Win10 64 - Geforce GTX 1080Ti

Hotas Warhog + Virpil T-50 base - Saitek Combat Rudder Pedals - Cougar MFCDs - Custom head tracker

35" UWQHD main display + 22" MFCD/Helios display / Rift S

2x256 GB SSD - 2Tb Caviar Green

 

Link to comment
Share on other sites

I'm experiencing the same drop with a GTX Titan 6GB. I did some screenshots where the difference is clear:

 

Object count is similar

Frame time is similar

Shadow time is similar

FPS is 83 vs 29 (I expect the value to be over 60 considering the frame time)

 

This makes me thing more of a CPU rather than GPU issue.

 

System specs:

i5-3570k @ 4.5 GHz - 8GB DDR3 1600 RAM - Win10 64 - Geforce Titan 6GB

Hotas Warhog - Saitek Combat Rudder Pedals - Cougar MFCDs - Custom head tracker

32" 1440p main display + 22" MFCD/Helios display

256 GB 840Pro SSD - 2Tb Caviar Green

 

No trees:

NoTreesValues.PNG

 

Trees:

TreesValues.PNG

 

Full images are attached to this post.

 

Nice to hear that the nV users have the same (not that you've got the same problem), so this can give this story a new direction!

It seems so, that this is a prob with the 3D renderer. Some things are now not "made for the GPU" and so the CPU must do all the work.

 

:thumbup:


Edited by Nedum

CPU: AMD Ryzen 7950X3D, System-RAM: 64 GB DDR5, GPU: nVidia 4090, Monitor: LG 38" 3840*1600, VR-HMD: Pimax Crystal, OS: Windows 11Pro, 2*2TB Samsung M.2 SSD, HOTAS: TM Warthog, Paddles: MfG.

Link to comment
Share on other sites

 

~

 

EDIT: Wow! I am blown away!!! I highly suggest trying out this tip...it solved my micro stuttering issue for a good amount! Happy times :)

 

 

Cheers...

 

First entry came back as "element can't be found" (searchings find it is a High latency fix for Windows 8/ 10)

Second entry came back as "successful" (now my X58 based PC feels like my new X99 based PC... lets see what happens there :) )

 

Windows 7 64bit

 

 

 

iEdit Mmmm, nice

 

platformclock = HPET Enabled


Edited by Wolf Rider

City Hall is easier to fight, than a boys' club - an observation :P

"Resort is had to ridicule only when reason is against us." - Jefferson

"Give a group of potheads a bunch of weed and nothing to smoke out of, and they'll quickly turn into engineers... its simply amazing."

EVGA X99 FTW, EVGA GTX980Ti FTW, i7 5930K, 16Gb Corsair Dominator 2666Hz, Windows 7 Ultimate 64Bit, Intel 520 SSD x 2, Samsung PX2370 monitor and all the other toys

-

"I am a leaf on the wind, watch how I soar"

Link to comment
Share on other sites

I am also experiencing micro stutters regardless if the FPS are steady or not and in all situations. With all possible combinations around Vsync, Nvidia control panel, inspector...I have been trying virtually every advice and combination there is (except the 144Hz Gsync monitor because I do not own one) for about 5 days and the issue just wont go away. My opinion is that the new 1.5 engine still needs more/better optimization and I hope that this will be solved. I expect that with my PC specs DCS runs smoothly...

 

EDIT: Wow! I am blown away!!! I highly suggest trying out this tip...it solved my micro stuttering issue for a good amount! Happy times :)

 

Cheers...

 

First entry came back as "element can't be found" (searchings find it is a High latency fix for Windows 8/ 10)

Second entry came back as "successful" (now my X58 based PC feels like my new X99 based PC... lets see what happens there :) )

 

Windows 7 64bit

 

iEdit Mmmm, nice

 

platformclock = HPET Enabled

 

Thanks guys.

 

Tried this trick after reading a bit about it.

 

Running windows 7 64-bit, so only used the first option, "bcdedit /set useplatformclock true" that forces HPET timer only in Windows.

Running a i7 2600K OC to 4.4 GHz max, and a GTX 670 GPU.

 

I am, like cro_mig_21 blown away..

 

Never have I seen DCS, or any simulator for that matter, run so smoothly..

Everything feels like a live movie now, and I can follow even the tiniest adjustments in my flight controls.

 

Zoom in and out and smoke from ground attacks, that both used to stutter terribly, are now almost completely smooth, with only a slight hiccup from time to time.

 

Though your results may vary, I highly recommend people to at least test this tweak, and see if you gain something by it.

I certainly did.

 

Now I just hope that it does not have too big adverse effects on my PC, but future will tell.

 

PS: Even if it does not solve the low FPS issue near trees, I can at least fly smoothly even with only ~30 FPS.

System specs:

 

Gigabyte Aorus Master, i7 9700K@std, GTX 1080TI OC, 32 GB 3000 MHz RAM, NVMe M.2 SSD, Oculus Quest VR (2x1600x1440)

Warthog HOTAS w/150mm extension, Slaw pedals, Gametrix Jetseat, TrackIR for monitor use

 

Link to comment
Share on other sites

Very interesting thread about this topic ( dont want to call it issue as it is very specific if it is a problem or gift... it depends)

 

http://www.overclock.net/t/1567745/windows-10-disable-hpet-before-install-and-enjoy-low-latencies

 

dont take the link's message as the "lemme say ONLY" truth, read before you work on your Bios/OS

Gigabyte Aorus X570S Master - Ryzen 5900X - Gskill 64GB 3200/CL14@3600/CL14 - Asus 1080ti EK-waterblock - 4x Samsung 980Pro 1TB - 1x Samsung 870 Evo 1TB - 1x SanDisc 120GB SSD - Heatkiller IV - MoRa3-360LT@9x120mm Noctua F12 - Corsair AXi-1200 - TiR5-Pro - Warthog Hotas - Saitek Combat Pedals - Asus PG278Q 27" QHD Gsync 144Hz - Corsair K70 RGB Pro - Win11 Pro/Linux - Phanteks Evolv-X 

Link to comment
Share on other sites

Very interesting thread about this topic ( dont want to call it issue as it is very specific if it is a problem or gift... it depends)

 

http://www.overclock.net/t/1567745/windows-10-disable-hpet-before-install-and-enjoy-low-latencies

 

dont take the link's message as the "lemme say ONLY" truth, read before you work on your Bios/OS

 

Well, it goes both ways: It seems that the issue is that either you enable HPET timer in both BIOS and Windows or you disable it both places.

 

The reason seems to be, that if HPET is enabled in BIOS, Windows tries to synchronize between it and its other timers, and that costs performance.

 

It is described a little in this thread, but people get diverting results:

http://www.neowin.net/forum/topic/1075781-tweak-enable-hpet-in-bios-and-os-for-better-performance-and-fps/

 

and here:

http://www.overclock.net/t/1487105/want-higher-fps-what-timer-do-you-use#post_22209166

 

I can just say on my PC it worked wonders to have HPET enabled both in BIOS and Windows.

I got very smooth flying.

But it might not be the same for everyone.

 

That's why I recommend people to try if it makes a difference for them or not.

It might be worth the try for some. :)


Edited by Sporg

System specs:

 

Gigabyte Aorus Master, i7 9700K@std, GTX 1080TI OC, 32 GB 3000 MHz RAM, NVMe M.2 SSD, Oculus Quest VR (2x1600x1440)

Warthog HOTAS w/150mm extension, Slaw pedals, Gametrix Jetseat, TrackIR for monitor use

 

Link to comment
Share on other sites

I tried it with the MiG-21 FreeFlight.

 

Indeed, it was enabled in Bios and visible in DeviceManger as HighPrecisionEventTimer.

 

I must say, I have Gsync and ever since no stutter, neither 1.5 nor 1.2.x, I just did it because I am curious :)

 

Well, first thing it did upon rebooting back from Bios and thus changed hw was to have a new "Legacy Device" which iirc wasnt there before and the driver is machine.inf, same as HPET device, might be the same device disengaged, might be not.

 

Second thing, it killed my MiG-21 license completely, worse than ever before. Glad I just got a new license from LS as I lost pretty all of them lately when the module disbehaved.

This time, I wasnt able to reactivate it not because of none left but some other "cant run as long as another instance bla..."...OK, I did the registry edit, deleted LS entry, tried... no luck

 

I then deleted the Mig21 module, rebooted, installed Mig21 again but git stuck with same error which I then surpassed by activating via console and the new lic LS granted me.

 

Well, after finally getting into the Mig21 cockpit, same as ever mission, it hasnt changed to the lesser, I dare to say it "seems" to be smoother, also my overall windows seems more responsiv, my Asus Tool now opens in a splitt second, it never did that, it feels boosted.

 

Might be placebo, might be true, for now as far as I can say it didnt harm it, neither OS nor DCS.

 

There seem to be tools out that indicate which Timers Windows has currently engaged as Windows seems to be able to activate HPET and others at will, regardless of what you set in Bios or Software elsewhere. For now I have no clue what Timer device is used by the DCS program on my system.

 

One should do a 3Dmark etc. before and after HPET changes, curious to see what that does.

It could really be the cure for many of the stutters in here, not all, but a good margin maybe.


Edited by BitMaster

Gigabyte Aorus X570S Master - Ryzen 5900X - Gskill 64GB 3200/CL14@3600/CL14 - Asus 1080ti EK-waterblock - 4x Samsung 980Pro 1TB - 1x Samsung 870 Evo 1TB - 1x SanDisc 120GB SSD - Heatkiller IV - MoRa3-360LT@9x120mm Noctua F12 - Corsair AXi-1200 - TiR5-Pro - Warthog Hotas - Saitek Combat Pedals - Asus PG278Q 27" QHD Gsync 144Hz - Corsair K70 RGB Pro - Win11 Pro/Linux - Phanteks Evolv-X 

Link to comment
Share on other sites

I'm sorry to hear you got problems with your activation. :(

I got nothing of that sort.

And I even tried all combinations also.

 

Thanks for your feedback, that is of course a caveat for people wanting to test it.

System specs:

 

Gigabyte Aorus Master, i7 9700K@std, GTX 1080TI OC, 32 GB 3000 MHz RAM, NVMe M.2 SSD, Oculus Quest VR (2x1600x1440)

Warthog HOTAS w/150mm extension, Slaw pedals, Gametrix Jetseat, TrackIR for monitor use

 

Link to comment
Share on other sites

  • Recently Browsing   0 members

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