Graphics improvements suggestions

Madcowie

 

Awesome. Cool. Top notch. Jolly good. Ruddy marvellous. Smashing, old chap.

Or a coating put on aviation fabrics. I'm unsure.

 

To avoid inconsistencies completely, the only method would be to have not delivered either an update or any additional content using updated core components between 2012 and 2014 - or - to have done similar for shorter gaps. This would have really made things slower, even if just from a learning standpoint. And that is both with the developers and the content creators. Luc himself looks at his old rF2 tracks pretty negatively.

I don't have issue with anything anyone is really posting here, bar those who just seem to come across as offensive as Tuttle was responding to (whether that is a language barrier issue or not - the ban button will work the same). Most people have an opinion that whether I agree with it or not, I can certainly respect that they have one.

Edit: It could be a huge headache I have, but I am having trouble following your posts.

 

Okaaaay...

(I think lol)

 

j

yes I see, but surely there must be a time when its decided that core shaders/tech is at a stage where all content must be updated/sync'd & it held in that state

.....unless there's some development which shows a nice step up in visual quality/performance gain that another total update
(-that's core&content release) needs to be done.

otherwise it seems that we're stuck in this ongoing process of inconsistencies between core & content
not to mention 3rd party mods falling out of sync some of which will never be updated

Rfactor 2 will not seem like a close to finished product untill this happens in my opinion
perhaps we're at the stage where it needs to feel this way

sorry for the shopping list :

core/shaders (any optimisations/updates that may be pending)
re-sync of all content to latest shader&core
weather visual overhaul (is this in the core?)
UI update

the actual overall feel of rf2 should hopefully feel complete or close to complete then,
the line has to be drawn somewhere (I'm not saying no future development but perhaps a very important mile marker has to be passed ...then the future handled a tad differently)
with any core updates to be together with previous content update/sync
(do we need many more core updates??? )
future updates mainly minor bug fixing & any already planned content

the all new rf2 would have landed then, good news for users, good news for modders,
hopefully good news for rf2 / ISI, bit of promo too

 

hehehe

Yeah what Tim about said old vs new components was " the dope" .


Like Yin and Yang p

RF2 will get it's yin yang don't you worry about that.





I think of it as another way tj , addiction


rF2 is the dope !@!

lool

 

lol...

 

Actually, I have a difficulty with newer core not handling older content properly Monaco is a great example.

I don't expect older content to magically take advantage of new features and I don't feel the need for it. They could use the same features they did on the day of their release. The problem is that they don't

True, but I think we need to look at two aspects of this:
- overall quality (textures, polygons, etc.) and taking advantage of features in newer builds
- backward compatibility

I have no problem with the first aspect. I enjoy Portugal the way it is. I don't need higher resolution textures nor more detailed objects. For me, it's an example of well balanced track that is a pleasure to drive. Sure some close up screenshots from recent tracks look nice, but I don't see these details while driving so it's not such an added value.

But as far as backwards compatibility goes things don't look so good.

What you would usually expect from developer is to add new features or making major changes to existing features in one release, not building each feature up through multiple releases.
If things keep shifting and changing all around, then product as a platform is unstable.
I will use iRacing as example - when NTM was introduced, old content did not become obsolete or started acting weird, because backward compatibility was provided. Neither I've seen older tracks or cars in iRacing start to look broken when new shadow system was introduced.

And one more point - I perfectly understand why you don't update your content. And this is why you should understand why modders don't update theirs and in some cases - abandon working on new. They're simply waiting for stable platform. And it seems you do, too That's the paradox of it, I guess. Updating old content is not a question of how much work it will require, but how long until that content requires another update.

But non-updated cars should show old raindrops just fine.
Otherwise you're in for a snowball effect - with growing number of content available, each build update without backward compatibility will require more content update work.

Even better alternative is to separate content from engine development. Content should describe physical properties of objects and not contain engine configurations (for example - materials should not refer to shaders nor provide transparency group settings).
In some cases it cannot be avoided (advanced things, like calculations for tires must be done offline) but in many cases it could (most of the graphical stuff, which does not require offline processing).

 

I think that's exactly what Tim meant, there will always be backwards compatibility, which means none of the old content will show the new raindrop effect. So it will be limited to newly released cars. It's the same approach as used with this upcoming tire contact patch update, afaik it will only apply for new cars. So I don't see it as reason not to introduce new effects to graphics.

 

This. Is. So. True.
I would quote this 100 times, but I don't want to get banned

 

+100

 

Well, since it would require changes to tools, engine and content, I don't suppose ISI will be making such change, but if we keep mentioning it and let the idea linger in their heads for a while who knows what may happen

Well, at least this is how the industry leaders do it.

 

Yeah.

I did make a proposal for a materials override file per mod where we could override each material with custom properties (without the need to re export the GMTs), but no feedback yet on that matter.

 

Well, that would affect the workflow, but not backwards compatibility.

To make backwards compatibility easier to achieve, we would need to change the way materials are described, not just where descriptions are stored. You only need to look at some tutorials for physical based materials in Unreal Engine 4 to see that artist does not work with such low level parameters like specular power anymore. He is not bothered with transparency order, shader selection and such. He is merely describing the object as it is in real world. It's up to game engine to decide how this appearance would be achieved.

So if some day programmer will rewrite, let's say, reflections from scratch, no one will have to update their content. That's the beauty of it - it makes life easier for both artists and developers.

 

Yeah, indeed.

Then again, it would be possible to implement such system with the material override file, where one could specify exactly what you just said. That however, requires more support from the game engine side.

I don't think anything like this will happen in rF2.

I just wish all the small annoying graphics bugs (HDR oil/black patches, envmap rendering issues, ambient car shadow issues etc) would be fixed.

 

It is kinda funny

I never here people talk about AC or pCARS " graphical glitches" which from my account there are many i have seen ?

It is like it is used as a excuse not to say " what does a few graphics glitches matter when you have superior feel and physics"



Personally I think you should be happy you have the best physics in a sim, ever.


I mean you say it is not the best graphics, so if it is not the best physics either what are you doing here ?


Be happy one out of two ain't bad.


If it comes to the crunch what do you want the nicest graphics or the best physics ?


If the answer is not the latter you need to shift sims. No offence meant

 

Durge, there has been great discussion between K Szczech, Tim, and Some1. The discussion is linked only to the betterment of rFactor 2. Please don't detract from it.

 

Erm...





Both...

 

No, there's no need to sit and be content as we have great physics, forgoing graphics and easier/better systems. The idea bandied about by ISI seemed to be constant improvements.
What I think it needs though is just the time to come around to where the devs focus is on those systems. Right now it looks like finalising the tyre model for this stage and sorting out the rules implementation and the new UI is their focus. Arguably those things could be seen as more important to them, a successful NASCAR and indeed any other oval type racing implementation would be seen as a big boost. Then a more complete tyre model and maybe tweaks to realroad to polish up. The new UI you think will be a long term thing, once it's done, it should be sussed for a few years to come.

So after these core things are done which are more important to the simulation, maybe then they will look at refining their graphical side of things.

 

Well said Mini.