Any explanation for move to Deferred Shading in 2.5.1 from Devs?

I've spent weeks and weeks playing with settings trying to optimise performance of 2.5.1 beta version to try and get both performance and image quality anything like as good as was before DS was introduced and especially in VR. Going back to V1.58 is like a breath of fresh air and wow for quality and immersion. With this in mind, I'm sure I'm not the only devoted player wondering why the move? Has this been explained anywhere please? :helpsmilie:

EDIT: For reference, observations of Deferred Shading so far.

1) Detail in image quality worse, cockpit text illegible without zooming in.

2) FPS performance down by 30%

3) Cockpits either too dark or rest of world too bright.

4) Sea is almost black except where suns reflections are.

5) Night lights barely visible

6) Delay re-shading scene when changing views. Especially from map back to cockpit view.

7) Plane surface appear over-shiny

However much some people may sympathise with your view, I’m afraid you’re stuck with DS, and 1.5.8 will be unsupported some time soon, if it isn’t already.

Don’t know if there has been a proper explanation, but it would be interesting to know the reasons. I’m sure it’ll be a technical reason, not just a preference.

 

1) Detail in image quality worse, cockpit text illegible without zooming in.

2) FPS performance down by 30%

3) Cockpits either too dark or rest of world too bright.

4) Sea is almost black except where suns reflections are.

5) Night lights barely visible

6) Delay re-shading scene when changing views. Especially from map back to cockpit view.

7) Plane surface appear over-shiny

All of the above especially the loss of definition in cockpit and the "luminous" colours of aircraft outside, since when has olive drab been illuminous?

I know we are stuck with it I've tried to mitigate it in some places, flying at different times etc.

but if you are on a server you have no control over time or weather, it could be midday easily.

Normandy has been completely unplayable for me since 2.5 release I don't see it ever getting any better. It was a complete waste of money.

Why are ED testing lighting fixes for a rural forest area on a desert map ? it doesnt make sense

I like the look of the F-18 but i'm not spending any more money until the basic game and what I have already purchased is fixed. ED is fast losing my support im sorry to say

I COMPLETELY AGREE...

You are NOT the only devoted player wondering why the move.

https://forums.eagle.ru/showthread.php?t=202460

[...]

EDIT: For reference, observations of Deferred Shading so far.

 

1) Detail in image quality worse, cockpit text illegible without zooming in.

2) FPS performance down by 30%

3) Cockpits either too dark or rest of world too bright.

4) Sea is almost black except where suns reflections are.

5) Night lights barely visible

6) Delay re-shading scene when changing views. Especially from map back to cockpit view.

7) Plane surface appear over-shiny

i can understand your frustration, but i think it is important to point out, that none of the issues are an inevitable consequenze of deferred shading, but probably more just bugs and problems from an major engine update that maybe has been pushed to release a bit too early.

for those that may not know: deferred shading is a fairly new technique, but used to great effect in many other games for some years now. it has many potential benefits while it's only major drawbacks are performance tolls on MSAA and very high resolutions (while giving huge potential performance increase in other areas, such as complex real-time lighting).

i think especially for a simulator that has many dynamic elements that affect the shading (seasons, weather, angle of sun/moon) a move to DF and PBR makes sense in the long run, since it should need less manual adjusting for assets/terrains/weather etc. once the lighting is dialed in correctly.

i myself am a little bit dissapointed too, because i feel that 2.5 was not quite ready when it got pushed out, with all the problems concerning night lighting especially.

however i think it does not make sense to argue against deferred shading in general. i would advise to see the problems that exist as bugs and politely demand for them to get fixed in a timely manner.

I agree, and do see the potential of deferred shading. BUT...

Not "maybe has been pushed to release a bit too early" Instead:

MOST CERTAINLY the major engine update has been pushed to release

FAR too early and in a rushed, erroneous and incomplete manner,

killing the cockpits and a lot of other stuff, especially at night.

So many people complain “We want 2.5, what’s taking so long?” And now you people complain that they released it to early. People already complaining about features on the ER of he F/A-18. That’s not even out yet. If you don’t like bugs, wait until full release. Stop complaining, report bugs as you find them and let the devs do their job. Thank you.

From what I have read, Deferred Shading's advantage over Pre-shading is when there are hundreds or even thousands of light sources to contend with, so the added time penalty of the second shading process more than pays off. As there is usually only one light source out there with us and our attention is the detail of cockpit dials and other information, choosing a shading means which negatively affects MSAA and resolution seems a bit odd to say the least. That's why I wondered if the Devs have given their explanation anywhere? I am guessing the time extra time now used for shading is the time the process used to have for carrying out MSAA.

So many people complain “We want 2.5, what’s taking so long?” And now you people complain that they released it to early. People already complaining about features on the ER of he F/A-18. That’s not even out yet. If you don’t like bugs, wait until full release. Stop complaining, report bugs as you find them and let the devs do their job. Thank you.

The forum and in particular the thread 'Open Beta Bugs and Problems' is here for us to give feedback is it not? I am doing just that and asked if the devs have given an explanation for the reasons for the change that's causing a lot of 'issues in this beta.

"4) Sea is almost black except where suns reflections are."

try details "water on high" it make a huge difference for water/sea rendering.less setting also make it black on my rig.

very good work on that point from ED in high settings.

for the other just think "WIP"

As a skinner i would like sometime they stop one day experiment and now close the rules of rough and spec files regarding defferred shading tweaking as soon as possible. i'm fed up to do and redo my skins, with all different way and rules from one module to an other. but anyway. let's be comprehensive on constant evolution process.

It will need time for third party module to update new rules and settings also.

but for me the most important and quick fix to find now is antialiasing vs deffered. What's a pity to still have those horrible AA partial result in 2018, not yet resolved, with overexpensive Nvidia. on XP11, deffered shading and AA are well integrated. it's a huge improvement how light works with different materials, but it's still expensive in perf.

Sorry guys but the old system is dead, at least if you have seen the screenshots from the new map, even at night. There is some work left regarding the lighting issues but Wags said they will port the new feature also to the other maps.

While there may be a reason to complain from the performance side we will see the integration of Vulcan and let's be honest, we are playing a top level flight sim so if anybody gets here and asks why it doesn't run on his 5 years old notebook without stuttering :music_whistling:

Because it's 2018 and things change, just like the move away from Doom's DOS4GW to 3d engines killed many PC's, but we got rid of sprites.

Why did they do it? Because it needed to be done and it's not feasible to continue maintaining multiple iterations of the game and its assets. That was the whole point of 2.5, remember?

This is not the first time a game has launched with a new engine, nor is it even the first time DCS has undergone engine overhaul. Each time, the old gets ditched and the new is taken up. Flaws will eventually be rectified.

The only problem was it could have stood some extra time in the code oven, but everybody was beating the damn doors down so they didn't have to have multiple installs. Well, it's all unified now, and AS WAS THE IDEA ALL ALONG, it brought a lot of modern tech with it. Also AS COULD BE EXPECTED it pushed out lower end PCs.

1.5.8 will remain available alongside other legacy versions of the game like 1.2 and LOMAC1/2. In 10 years or so, we'll probably get another overhaul, and anybody not running something made since 2025 will get pushed out. It's life.

@Naked350, my reply wasn’t aimed at you it was aimed at people complaining about EA.

A lot of passion about driving the game forward which is awesome, but my original point is have the Devs given an explanation why they opted for DS over other rendering methods/ engines etc considering the negative impacts and mountain of work to do. There must be a technical reason, so because it's progress doesn't really answer the question.

"4) Sea is almost black except where suns reflections are."

 

try details "water on high" it make a huge difference for water/sea rendering.less setting also make it black on my rig.

 

Awesome! That's a lot better, thanks @snowsniper :thumbup:

The forum and in particular the thread 'Open Beta Bugs and Problems' is here for us to give feedback is it not? I am doing just that and asked if the devs have given an explanation for the reasons for the change that's causing a lot of 'issues in this beta.

What are the issues relating to this beta that you mention? The issues discussed here are present in the release version as well, but significantly improved in 2.5.1.

Please state what issues are specific to 2.5.1.

What are the issues relating to this beta that you mention? The issues discussed here are present in the release version as well, but significantly improved in 2.5.1.

Please state what issues are specific to 2.5.1.

The presence of the deferred shading 'off' button primarily.. There is no choice for users going forward, despite the huge negative impact on game quality and performance. That's a bold step for a technology that's no where near ready. It's not good for PR, especially for the customer base on a budget with lower end spec machines? I also have a gaming notebook, barely a year old cost £800 will just about run with DS on at 60 FPS with no MSAA. So I ask again from the Devs , why the move to DS?

It would always be a hard call in the direction to go in, it's a long way back for a new direction. I'm guessing they took quite a long time before deciding the direction of the graphic engine and moving to the Vulkan api. ED will get this settled, hopefully it won't take too long.

i can understand your frustration, but i think it is important to point out, that none of the issues are an inevitable consequenze of deferred shading, but probably more just bugs and problems from an major engine update that maybe has been pushed to release a bit too early.

 

for those that may not know: deferred shading is a fairly new technique, but used to great effect in many other games for some years now. it has many potential benefits while it's only major drawbacks are performance tolls on MSAA and very high resolutions (while giving huge potential performance increase in other areas, such as complex real-time lighting).

 

i think especially for a simulator that has many dynamic elements that affect the shading (seasons, weather, angle of sun/moon) a move to DF and PBR makes sense in the long run, since it should need less manual adjusting for assets/terrains/weather etc. once the lighting is dialed in correctly.

 

i myself am a little bit dissapointed too, because i feel that 2.5 was not quite ready when it got pushed out, with all the problems concerning night lighting especially.

however i think it does not make sense to argue against deferred shading in general. i would advise to see the problems that exist as bugs and politely demand for them to get fixed in a timely manner.

The explanation they gave when they originally locked it down was effectively what I said, because keeping the option for both meant doing everything twice (when DS first came out, things that were ready looked fine, things that weren't looked weird, and vice versa, it required overhauling assets).

As for why Deferred Shading SPECIFICALLY was chosen, Google is your friend, as always.

::: (copy paste incoming):::

In the field of 3D computer graphics, deferred shading is a screen-space shading technique. It is called deferred because no shading is actually performed in the first pass of the vertex and pixel shaders: instead shading is "deferred" until a second pass.

The primary advantage of deferred shading is the decoupling of scene geometry from lighting. Only one geometry pass is required, and each light is only computed for those pixels that it actually affects. This gives the ability to render many lights in a scene without a significant performance-hit.[3] There are some other advantages claimed for the approach. These advantages may include simpler management of complex lighting resources, ease of managing other complex shader resources, and the simplification of the software rendering pipeline.

::: (end) :::

Basically, from that I gather it lets them shortcut to avoid spending so much resources on rendering terrain and related effects. In other words, it's more suitable for a large scale, open world where you're rendering hundreds/thousands of objects and light sources at a time. Disadvantages seem to be related to difficulty in rendering transparency (probably why the cockpits are f'ed up on multiple aircraft) and multiple materials rapidly increasing the draw on memory (the large increase in system requirements for RAM and especially VRAM). As always, the biggest hit to DCS comes from it's nature : a complex physics engine coupled with trying to render respectable modern graphics out to a potential range of a couple hundred miles (remembering at altitude you can see a LOOOOOONG way, and aircraft are often attempting to locate targets and landmarks from 20+ miles away).

DS does not appear to be related to PBR, which is a method of shading textures to more closely resemble real world materials. When it's done properly, it produces photo-realistic imagery with minimal overhead. BUT if resources have not been updated yet (as is the case with many things here at present) it causes noticeable color/material aberrations because they haven't been prepared properly for the rendering. A "texture" is not a simple image, but rather an image, with many, effectively subfiles, within it that the GPU interprets as to how to display that image and with what effects. That's one way you get "rough, textured" surfaces, without there actually being model geometry. It's conveyed in the texture and the GPU extrapolates it on the fly.

As for the issues it's causing... yeah, they pushed it out a few months early, looks like. Growing pains. People would complain if they didn't, they're complaining because they did. The technical reasons for why they chose it can be taken from the above, though. That's how it works, so it's pretty easy to see what their motivation in doing it was. Were there other options? Probably. I'm sure they looked at lots of stuff, and determined this was the best route for their application.

As for new blood, let's be honest 99% of people are going to flush out in the first couple hours because of DCS's complexity in the first place. The ones with the patience to read a 400+ page manual aren't likely to be deterred by graphical anomalies or temporary bugs and performance issues lol This was never a poor man's hobby :)

The presence of the deferred shading 'off' button primarily.. There is no choice for users going forward, despite the huge negative impact on game quality and performance. That's a bold step for a technology that's no where near ready. It's not good for PR, especially for the customer base on a budget with lower end spec machines? I also have a gaming notebook, barely a year old cost £800 will just about run with DS on at 60 FPS with no MSAA. So I ask again from the Devs , why the move to DS?

Indeed, so it’s not issues relating to 2.5.1, it’s DS that you are complaining about. Opinions on DS aside, 2.5.1 is a big improvement on 2.5 release version with DS on.

Indeed, so it’s not issues relating to 2.5.1, it’s DS that you are complaining about. Opinions on DS aside, 2.5.1 is a big improvement on 2.5 release version with DS on.

Well as 2.5.1 has singled out a rendering method that negatively affects both performance and image detail amongst the other issues I started my post with, (and I have read a fair bit about the science behind DS before posting) then the issue is 2.5.1 and why I asked the question if the reasoning to use DS has been given by ED anywhere?

Well as 2.5.1 has singled out a rendering method that negatively affects both performance and image detail amongst the other issues I started my post with, (and I have read a fair bit about the science behind DS before posting) then the issue is 2.5.1 and why I asked the question if the reasoning to use DS has been given by ED anywhere?

Your question is a very reasonable one, and I am interested in the answer as well. However, my point is that it is DS you have the issue with, and not 2.5.1.

When this beta is made the release version, you will have to live with DS whether you like it or not. That is, sadly from your point of view, a fact.

FWIW, I personally agree that cockpits are better with DS off at the moment, but the best thing to do is for everyone here to focus on raising issues to help ED improve the DS rendering over the next few months. Nobody, including ED, is saying this is the finished state of DS in DCS.

Your question is a very reasonable one, and I am interested in the answer as well. However, my point is that it is DS you have the issue with, and not 2.5.1.

When this beta is made the release version, you will have to live with DS whether you like it or not. That is, sadly from your point of view, a fact.

FWIW, I personally agree that cockpits are better with DS off at the moment, but the best thing to do is for everyone here to focus on raising issues to help ED improve the DS rendering over the next few months. Nobody, including ED, is saying this is the finished state of DS in DCS.

There are some great advances in 2.5.1, such as the off-line mode and memory manager seems to load missions a lot faster than 2.5. However, I guess what I am looking for is some reassurance from ED, that despite the initial set back in performance and quality of detail, that they have every confidence DS is the best choice available the end result will be better resolution, and run slicker than it was before.

I appreciate the comprehensive work that must be going on behind the scenes, but some sort of road map would be great if available. Luckily my PC will still run 2.5.1 in VR with reasonable results, but at the moment it's not as good experience as it is in 2.5. If ED wish to push ahead in the DS route that is their perogative, however they will alienate

/ discriminate against the existing lower end new entry users, who potentially have the most potential to spend on the offering of 20 or so planes plus the maps, whereas existing, elite players will possibly now be buying just the latest ones as they are released, which is what, 2 releases per year?

People will altimately vote with their wallet and feet, so if nice trees is what battle sim pilots are after there shouldn't be a problem lol.

however they will alienate

/ discriminate against the existing lower end new entry users

Yep, advancing technology tends to do that. Especially when said technology still has kinks to be worked out.

who potentially have the most potential to spend on the offering of 20 or so planes plus the maps

People always love to trot out market projections, about how this or that minority group has tons of money to blow, and how devastating some decision or other is to "the bottom line". Here's the thing : These people who can't/won't buy a $200-300 GPU made in the last 2 or 3 years are going to trot out hundreds of dollars on airplane modules for their antiquated PC/notebook? There are flight sims more appropriate to their hardware/price bracket than DCS. You don't get the perks of DCS, which is hardware intensive by nature, without having to keep a moderately powerful/somewhat modern build on hand.

Once the new engine has somewhat stabilised performance will improve a bit, like it already is beginning to as I understand it, but nothing is going to change the fact you can't buy the latest and greatest games year after year and expect they're always going to work on your PC. You have to upgrade, typically every 3 years or so, if you want to keep doing that. That, or when the new stuff starts outpacing you, you stop buying it and stick to the stuff you can run.

whereas existing, elite players will possibly now be buying just the latest ones as they are released, which is what, 2 releases per year?

My goal is to purchase every module, even the ones I don't like :) I'm down to that little Italian plane, the Spitfire... and the Viggen. And that weird GPS thingy only one helicopter can use. I'm doing my part to fund development :P

Yep, advancing technology tends to do that. Especially when said technology still has kinks to be worked out.

 

 

 

People always love to trot out market projections, about how this or that minority group has tons of money to blow, and how devastating some decision or other is to "the bottom line". Here's the thing : These people who can't/won't buy a $200-300 GPU made in the last 2 or 3 years are going to trot out hundreds of dollars on airplane modules for their antiquated PC/notebook? There are flight sims more appropriate to their hardware/price bracket than DCS. You don't get the perks of DCS, which is hardware intensive by nature, without having to keep a moderately powerful/somewhat modern build on hand.

 

 

 

Once the new engine has somewhat stabilised performance will improve a bit, like it already is beginning to as I understand it, but nothing is going to change the fact you can't buy the latest and greatest games year after year and expect they're always going to work on your PC. You have to upgrade, typically every 3 years or so, if you want to keep doing that. That, or when the new stuff starts outpacing you, you stop buying it and stick to the stuff you can run.

 

 

 

 

My goal is to purchase every module, even the ones I don't like :) I'm down to that little Italian plane, the Spitfire... and the Viggen. And that weird GPS thingy only one helicopter can use. I'm doing my part to fund development :P

I guess it all comes down to an individuals perception of advancement in technology. For me with respect to a game / sim would be to improve both the visual and combat experience whilst also improving performance and reducing hardware dependancy. I am yet to understand the 'advancement' DS brings that outweighs the negative impact the change has made. It's not a win win situation, which as a product developer myself, is what I always strive for when advancing a product, but with a cost saving too.

I have many of the planes and terrains of DCS, but for me Normandy was a waste of money due to the low performance with graphics settings needed to make it look anything like decent. DS is needed for one. You can have my Spitfire if you want. I never fly it in DCS as it's a better experience flying the MkVb over Kuban, Stalingrad etc. MkIX imminent too, which is where my money will be spent, again down to better gaming and graphical experience. Just my personal opinion.