Flighter

Nice post SkateZilla, thanks for your time. I have one question : Do you say that the Deferred Shading was the ONLY way to go for using the PBR material model ? If so, I strongly disagree. Agreed, the DS method is "less compromising", one small step closer to ray tracing. The alternative is a less accurate approach (for sane, short rendering times) but with significantly better performance IF WELL IMPLEMENTED/BALANCED. The produced PERCEIVED quality with present top notch rendering hardware is - for SCREEN projection identical for both methods. - for VR HMD projection the DS method is a killer no good rascal that eats away all your candy and you are left crying. Emphasis on the word PERCEIVED. As I am sure you are well aware, 3d rendering, starting from prefab pseudo 3d overlays, all the way to full blown ray tracing, reflection, refraction, caustics with infinite bounces and sugar on top, is a series of balancing between complexity/speed of render and optical precision/image quality. If we bring psycho optics in the equation we get very, very interesting results considering that : - the average person can hardly identify, let alone verify the effects of more than 10 point lights in a room (see cockpit). - a dynamically changing scene makes the accuracy of the results 99% impossible to verify in real time. - the irrelevance of the accuracy is multiplied if the lighting/shading itself is not the main subject. - Accuracy of ambient lighting is even harder to grasp (next to impossible). Human brain is easily pleased by very small tricks. Our brains only need to be convinced about a scene, they will never analyze it for geometrical precision. This is what DS brings us, do we need it ? And at what price ? How many fps and how much more shimmering is worth paying for the shadow and the lighting of the 30th little indicator that gets lit under my arm ? Is it going to make a perceivable difference ? Do I care if it is mathematically and analytical geometrically correct ? The way ED took, I think, is not the realistically wise but the theoretically impressive. At least for people with VR setups, this is major step backward. I believe that the one phrase that needs more focus is "Perceived Quality" as it holds many solutions that maybe were left unexplored. At the end of the day perfection is the use of the best compromises under given circumstances. Cheers

Hi Hippo. I cannot agree with you more. I think people went for DF as it was the new kid on the block and did not compare/evaluate it properly. Plus the priority in VR is the frame rate and the maximization of the perceived image quality (because of the low resolution of out HMDs). These two things got a deadly hit from the new developments. I tried last night to play on line in 2.5.1. I wanted to convince myself that there is some setting that is OK, even with less overall quality. I am afraid that I was left in disappointment and frustration. Cheers !

2.5.1 update 1 no improvement... sad... I have just installed and tested today's update (2.5.1.16987) I am afraid that the miracle I was expecting did not happen... :cry: Performance is identical with previous beta. I am sad.

lemoen, MSAA makes a hell of a difference on my system. Try on 2.5.0 with DS off, PD 1.5 and MSAAx4.

Guys, going from 2.5.0 to 2.5.1 with identical settings is not producing any noticeable difference in performance. The whole issue is that 2.5.1 does not let you turn off Deferred Shading and the FPS penalty for that is gigantic (increasingly so with MSAA on x2 and x4). Check what difference it makes using 2.5.0 and switching DS on and off (restart of the sim is required). Cheers

Hi there. Do you have Oculus ? Are you sure that some external utility (OculusTray for example) is not messing about with your settings ? Do you have 2.5 or 2.5.1 ? Changing deferred shading with MSAA x4 should make a hell of a difference. Or you have somehow an active fps limiter. Cheers

From latest ED update newsletter : Could this be it ? Anybody has any more info to share ? Cheers !

Wolf, it does look nice indeed. I have attached my measurements. See what performance you would be getting with your settings... The comparison says it all really. The new version is slower on every single combination (of interest to me at least). Let's hope the implementation is faulty and we get a nice surprise soon. Or the old rendered as an option. MSAA is also very important to my taste. Cheers

I also have it a "Use the 3D application Setting".

I was under the same impression until I saw Jef's post. Unfortunately after testing with NVIDIA profile inspector, nothing changed. Unless I am still missing something, I can verify that AA at least, cannot be modified outside DCS (DCS + Oculus runtime). If anyone knows any dirty tricks talk now !

Hi Jef. Many thanks for the file. I have imported your file into NVIDIA profile inspector and activated it. Unfortunately it has no effect on my Oculus system. If there is any other trick that I have (for so long) missed, on how to change the oculus AA method outside DCS, let me know. Cheers

Jef, I cannot force the AA settings using the NVIDIA Profile inspector. If I use the screen it seems to be working (for example I can see the green square if I set the FXAA on). But in the oculus, it completely ignores me. Can you please provide some tips or a guide ? Could you also post the exported profile form the nvidia inspector ? Cheers.

Hi ! 2.5.1 is in no way able to produce anything playable with these settings. Thus this whole thread. In 2.5.0 (current stable) as well as in 1.5.8 are all playable with 45 fps stable with ASW (very pleasant also). Plus our HDMs do not have the same pixels : Oculus : 2160x1200 = 2.592 Mpixels Odyssey : 2880x1600 = 4.608 Mpixels To match your pixel density I would have to start at 1.33 ! If you multiply your HMD's native pixels with increasing PD you run very fast to huge amounts of pixels. Do not forget that PD multiplies vertical as well as horizontal, so it's effect on GPU workload and memory requirements are squared ! Take also in consideration that one big disadvantage of Deferred Shading is big memory requirements (another reason why DS is bad for VR and high PD). The card you need probably does not exist yet. My system : i7 6700k at 4.6G Gigabytes G1 GTX1080 / 8GB ram / 2Ghz GPU clock 16GB ram / 3200 Mhz Separate SSD disks for DCS and windows WIN10/64 pro.

Dear Jef, an intriguing post. Totally agree with you. Good example was X-Plane vs FSX. The irregularity in frame presentation (jitter) of FSX was ruining it for me, while x-plane, although in the 25 to 30 fps range (on my rig at the time) was much more consistent and looked much better. I am very curious to try your setup. I am under the impression that we cannot (I have tried and failed) override in-game settings other than PD from any external tool (at least with oculus). AA level and it's method, including MFAA, even post filtering with FXAA (sacrificing all texture crispiness) etc. Is nVidia inspector able to do that ? I will give it a try and report back. Two visible reasons : Looking at dials, and looking on scenery with straight outlines (see buildings) flashing all over the place as you move. The in-game MSAA at x4 is a must for me for these two reasons. They vastly improve the perceived quality of the admittedly limited capabilities of the current mainstream HMDs. The same goes for Pixel Density (I find 1.8 ideal, 1.5 good compromise). As I said, I do not mind how, as long as the result is good. It was near perfect in 2.5. If I can get the same overall quality, I do not give a damn about the fps. I will try your recipe, I hope you will provide some tips if I run into problems. Thanks again !

Sun-screen advice Hi All. Always use sunscreen. For the tl;dr : - with DS, my VR experience is bad no matter the tweaking. - On line streaming is out of the question (Oculus mirror). - Deferred Shading does not produce better results. Bad implementations produce bad results. - D.S. may be the wrong selection for a flight simulator. Creates more problems that solves. - We have more than enough computing power to have excellent VR experiences and performance. - DCS seems in a permanent state of "work in progress"; including all (?) modules; prepaid W.I.P. that never get finished. Still awake ? read on, we'll fix that. I spent the entire weekend trying to find a set-up that will let me use the new scheme. Unfortunately, although I tried to see only the positive side, I was left with disappointment. The previous "experience quality" (for VR) is not comparable to the new one. I was toying with on-line streaming. I did a couple of tests and all was fine. I use Oculus mirror to create another window with proper ratio, let's say 16:9. Then capture, compress and stream that; while playing is VR. Well,,, you can forget that now. Plus the sacrificed visual quality is BAD on a monitor. While searching the forums for any missed tips, I noticed an interesting trend. People defend Deferred Shading because "it produces better image quality". I do not know how on earth this came about, at least on the technical side. Both available rendering pipelines (and many many more), IF WELL IMPLEMENTED, should be able to produce indistinguishable results. And like any tool, each one is better at one thing or the other. The strong point of deferred shading is the ability to produce lighting and shadows from many many light sources quite fast. That's about it. Now think, is this an FPS game, let's say Battlefield, where multiple light sources illuminating the scene (and producing shadows), really make a difference in the scene itself (and immersion) ? Is THIS the one thing that is missing from DCS ? Why on earth do you want that on a flight simulator ? We cannot even use "Cockpit Global Illumination" because of the extra FPS hit. And the biggest irony of all is that in order to have a decent frame rate in VR, you should turn off shadows completely (has even been suggested above !!!!!) Why the hell do I need DS in the first place then ? half of it's job is gone wasted. Another misconception is that at the moment there is not enough computing power to have a decent VR experience. This is very far from the truth. We have two key components, massive processing power and parallelism. If used properly the results may be amazing. I saw Aerofly FS2 mentioned elsewhere. I have this title and although of a totally different nature, is amazing in visual quality. IL-2 (I have almost all of them) are very, very good when it comes to VR quality. Even prepar3d v4 is getting very good at VR. And they have one unique and valuable feature : you can switch between monitor/ VR headset mid-game. But the title that is the golden standard (for me) of VR performance is Combat Air Patrol 2 (another title I bought). It demonstrates the capability of VR. Has to be flown to be believed. The fluidity at PD of 1.5 and everything maxed out, makes for an unforgettable experience. So, everybody is improving and DCS is going in reverse ? It's my favorite flight simulator for God's sake. I pay and run two servers for them, I talk with actions not simply words. Anyway, I still hope that they will rethink their decision and will at least give people the choice. Even if it costs them extra man-hours. I also hope that for a change we start seeing things finished. Anything I see in DCS is half done. Pre-paid and half done that is. There is no sense or "production grade" product; the type that you do not touch before you have tested thoroughly. Let's hope that they realize this never ending trouble pattern and put a stop to it. And as the famous song says, the sunscreen advice is the only sure thing about this post; everything else is my personal opinion and I reserve the right to be very, very wrong. Keep smiling !

I cannot disagree with most of the previous posts as they mostly express personal preference. But regardless of preferences the fact is that : We had the option to play at almost 90fps with MSAAx4 and PD at 1.5. This is not there anymore. We are forcefully downgraded to 33 fps. This is almost 1/3 of the initial performance. So I think, it is important to KEEP the choice rather than not. And we should ask for that because it is to our benefit as users, even if some may not be using it. Regarding the MSAA, the sensation of pulsating scenery that is caused by the moving jaggies, totally destroys the immersion factor of VR for me. Sorry guys but I (we) have payed a lot of money in software and hardware (server owners even more...) to enjoy this game. To have things that were there, taken away, simply does not feel good. We may at least ask them back, polite and with reason. And I hope ED is reading this and working on a solution. I hope. Cheers

Nessunio, most probably you went for the "deferred shading on" from the beginning and did not experience the fluidity and picture quality without it. I speculate that if you had, there is no way you'd be saying this. But again, there are more opinions that people, and all have a valid point. Especially when it comes to preference. Cheers

Hi guys. Unfortunately deferred shading cannot be controlled anymore. According to the change-log : Deferred Shading render locked in the game and can not be turned off. DCS ceased support of old graphic render without Deferred Shading.I don't think they realized the implication of this decision in this (small ?) user group. It's a shame really as the last version was amazing in VR. Instead of going forward we go back. At the moment I am very disappointed. It's too big of a setback and sacrifice of visual quality to accept it. Still, what can you do ? I hope they realize that so many ppl with VR and 1060s and 1070s are thrown out of the game (or to very lesser quality of it) and maintain the older rendering path a little bit longer. Cheers guys.

Hi all. I hate to be negative but... The latest update really killed the FPS for VR players. I hope ED takes notice and comes up with a solution. The disaster comes from the "forced deferred shading". I used to get about 80 to 90 fps on my test flight which is the Nevada free flight with the KA50. Now I get 30 to 37 with terrible visuals (anything below 45 is catastrophic for VR as all hell breaks loose). As probably most of you know the deferred shading has a very big enemy called MSAA. Which (MSAA) is an absolute must (for me at least) for the low native resolution of the Oculus headset. So I had the deferred shading off, MSAA at x4, shadows at medium and pixel density at 1.5. And I had a nice 80 to 90 fps. With ASW on a SOLID 45. All this heaven is lost as the deferred shading is now forcibly on. I have to either turn off shadows completely or turn off MSAA. With MSAA at x2 I almost get my 45 fps with bad visuals that kill the VR immersion. I really hope the ED will reconsider their decision forcing this very questionably good / bad (depends who you ask) shading method on. My system: CPU : i7 6700k at 4.6G GPU : Gigabytes G1 GTX1080 / 8GB ram / 2Ghz GPU clock RAM : 16GB ram / 3200 Mhz HDD : Separate SSD disks for DCS and windows OS :WIN10/64 pro. Test procedure described here : https://forums.eagle.ru/showpost.php?p=3549074&postcount=218 LATEST MEASUREMENTS 3-APR-2019 - 2.5.4.29167 : One year later... Modified shaders by Kegetys : Just remarkable improvement ! Addition 1: I cannot disagree with most of your posts as they mostly express personal preference. But regardless of preferences the fact is that : We had the option to play at almost 90fps with MSAAx4 and PD at 1.5. This is not there anymore. We are forcefully downgraded to 33 fps. This is almost 1/3 of the initial performance. So I think, it is important to KEEP the choice rather than not. And we should ask for that because it is to our benefit as users, even if some may not be using it. Regarding the MSAA, the sensation of pulsating scenery that is caused by the moving jaggies, totally destroys the immersion factor of VR for me. Sorry guys but I (we) have payed a lot of money in software and hardware (server owners even more...) to enjoy this game. To have things that were there, taken away, simply does not feel good. We may at least ask them back, polite and with reason. And I hope ED is reading this and working on a solution. I hope. Cheers Addition 2: Hi All. Always use sunscreen. For the tl;dr : - with DS, my VR experience is bad no matter the tweaking. - On line streaming is out of the question (Oculus mirror). - Deferred Shading does not produce better results. Bad implementations produce bad results. - D.S. may be the wrong selection for a flight simulator. Creates more problems that solves. - We have more than enough computing power to have excellent VR experiences and performance. - DCS seems in a permanent state of "work in progress"; including all (?) modules; prepaid W.I.P. that never get finished. Still awake ? read on, we'll fix that. I spent the entire weekend trying to find a set-up that will let me use the new scheme. Unfortunately, although I tried to see only the positive side, I was left with disappointment. The previous "experience quality" (for VR) is not comparable to the new one. I was toying with on-line streaming. I did a couple of tests and all was fine. I use Oculus mirror to create another window with proper ratio, let's say 16:9. Then capture, compress and stream that; while playing is VR. Well,,, you can forget that now. Plus the sacrificed visual quality is BAD on a monitor. While searching the forums for any missed tips, I noticed an interesting trend. People defend Deferred Shading because "it produces better image quality". I do not know how on earth this came about, at least on the technical side. Both available rendering pipelines (and many many more), IF WELL IMPLEMENTED, should be able to produce indistinguishable results. And like any tool, each one is better at one thing or the other. The strong point of deferred shading is the ability to produce lighting and shadows from many many light sources quite fast. That's about it. Now think, is this an FPS game, let's say Battlefield, where multiple light sources illuminating the scene (and producing shadows), really make a difference in the scene itself (and immersion) ? Is THIS the one thing that is missing from DCS ? Why on earth do you want that on a flight simulator ? We cannot even use "Cockpit Global Illumination" because of the extra FPS hit. And the biggest irony of all is that in order to have a decent frame rate in VR, you should turn off shadows completely (has even been suggested above !!!!!) Why the hell do I need DS in the first place then ? half of it's job is gone wasted. Another misconception is that at the moment there is not enough computing power to have a decent VR experience. This is very far from the truth. We have two key components, massive processing power and parallelism. If used properly the results may be amazing. I saw Aerofly FS2 mentioned elsewhere. I have this title and although of a totally different nature, is amazing in visual quality. IL-2 (I have almost all of them) are very, very good when it comes to VR quality. Even prepar3d v4 is getting very good at VR. And they have one unique and valuable feature : you can switch between monitor/ VR headset mid-game. But the title that is the golden standard (for me) of VR performance is Combat Air Patrol 2 (another title I bought). It demonstrates the capability of VR. Has to be flown to be believed. The fluidity at PD of 1.5 and everything maxed out, makes for an unforgettable experience. So, everybody is improving and DCS is going in reverse ? It's my favorite flight simulator for God's sake. I pay and run two servers for them, I talk with actions not simply words. Anyway, I still hope that they will rethink their decision and will at least give people the choice. Even if it costs them extra man-hours. I also hope that for a change we start seeing things finished. Anything I see in DCS is half done. Pre-paid and half done that is. There is no sense or "production grade" product; the type that you do not touch before you have tested thoroughly. Let's hope that they realize this never ending trouble pattern and put a stop to it. And as the famous song says, the sunscreen advice is the only sure thing about this post; everything else is my personal opinion and I reserve the right to be very, very wrong. Keep smiling ! Addition 3: The measurements of the oculus performance tool. Old vs New renderer. Addition 4 : 2.5.1 update 1 shows no improvement (or change) in performance :cry:. Addition 5 : Update 2.5.2.17559.377 shows no improvement (or change) in performance The new update has almost identical performance with previous versions (well within the measurements uncertainty). The new Anti-aliasing method present in the settings (SSAA) is not compatible with VR as it produces distorted view. It is more or less identical in operation with the Pixel Density setting so it is no big deal. The brown curve (2.5.2 SSAA) is pretty pointless (especially in conjunction with PD other than 1) but included just for fun.

PHP snipet for testing TCP socket connectivity (server) Hi guys. Assuming you have a web server with php capability (almost all), use the code below either embedded in a page or by itself in a file (eg test.php). Adapt the ip, port and messages to your server and liking. Any html tags will work in the messages. Note that if the server is down you may have to wait for 2 seconds (last parameter on the fsockopen function, may want to experiment/trim it) before the rest of the page loads. Good luck <?php // Enter your ip, port and messages below // You can use any html format for the messages // for example : // $messageOK = "<img src='serverok.gif' alt='Server OK'>"; // $ip = "94.130.23.93"; $port = 10308; $messageOK = "JUST DOGFIGHT Server is UP <br/>\n"; $messageNR = "JUST DOGFIGHT Server is DOWN <br/>\n"; // // ---------------- $fpntr = fsockopen($ip, $port, $errnumber, $errstring, 2); if (!$fpntr) { echo $messageNR; } else { echo $messageOK; fclose($fpntr); } ?>

Auto server start with new login system Hi guys. The new login will work with the following modification of the MissionEditor.lua located in <DCS INSTALLAION FOLDER>\MissionEditor folder. It is based on the script of this thread. You need the dedicated.lua file as described on the first post. Search for the function onShowMainInterface() in the MissionEditor.lua Change it like the one below by adding the code from "-- START DEDICATED CODE" to "-- END DEDICATED CODE". function onShowMainInterface() --print("--onShowMainInterface()---") if tooltipSkin_ == nil then tooltipSkin_ = Gui.GetTooltipSkin() else Gui.SetTooltipSkin(tooltipSkin_) end prepareMissionPath() mmw.setLastWallpaper() openReturnScreen() -- START DEDICATED CODE - ADD FROM THIS LINE -- if grgFirstRun == nil then grgFirstRun = true local net = require('net') local lfs = require('lfs') local Tools = require('tools') local mpConfig = Tools.safeDoFile(lfs.writedir() .. 'Config/serverSettings.lua', false) local dediConfig = Tools.safeDoFile(lfs.writedir() .. 'Config/dedicated.lua', false) if dediConfig and dediConfig.dedicated ~= nil and dediConfig.dedicated["enabled"] == true then net.set_name(dediConfig.dedicated["name"]) net.start_server(mpConfig.cfg) net.log("Starting Dedicated Server...") end end -- -- END DEDICATED CODE - ADD UP TO THIS LINE end function setPlannerMission(pm) planner_mission = pm end Cheers

Excellent !! My clipboard has the string as I write this :) Once again many thanks !!!

No More Stutters !! Hi Rik 'n company, I am very glad to report that on 1.5.8.12823 (open Beta ATM) the stutters are gone. I also noticed on the change-log that the modal crash window is also gone (Do we need to disable it somehow ? ) Congratulations to the team. Best regards.

Hi Rik. I hope your devs can hunt it down and produce a quick fix. And once there, please please pass them on one more message : Please remove this modal dialogue box that stops a crashed server from exiting. Make it an option if you want. It prevents my little script that monitors the server process, from automatically restarting it. Please pass it on. Cheers

Adding a password fixes stutters. Hi Drex Very very good post. Very important point is the modal window about the crash report that throws my restart script to the trash. IT MUST BE REMOVED ASAP. Regarding the issues. See this thread : https://forums.eagle.ru/showthread.php?t=197916 Although ED was warned they went ahead and rolled out buggy as hell code. Anyway,it seems I solved (until a proper fix) these terrible stutters on my servers by (!!!!!!) using a password (!!!!!!). See linked images; same server, same mission, viewing same static ground object. The only change is the connection password. Try and please let me know. In an Oculus client I get an almost solid 90 fps. It was disgusting before the password voodoo. I hope it works for you too. Check the JUST DOGFIGHT server. It was jerky as hell... Cheers.