ISI Planning to convert RF2 from DX9 to DX10 or DX11 like iRacing did?

Status
Not open for further replies.
Well its not an attack but i really like the way rf2 is but its those stupid graphics stuff.

Long loading times i dont care. If i need to wait 5 min for a track load its ok. But the game needs to be smooth.
 
Tim, i understand what u mean. But a league is hard to convince taking rf2 right now.

And i know that AC is not RF2 but it is, imo, a big gap. Again my opinion.
I think the performance gap will reduce, both because of our work, and because of their work. When they implement some of the things we already have, I am also sure your FPS will go down there as well. It's a fact of life. :)
 
People see "DX11" and "performance" and assume double+ framerates if ISI would just pull their finger out and move to DX11.

Read this guy's posts properly, and try to actually understand what's being said:

You need to design your game engine an your tools towards these new features.
That's why it's not something I believe would happen with rF2. ISI is a small company and is using "cheap" modding tools (just some plugins for 3DS) and will not be able to take advantage of all DX11 features. It's rather a priviledge of big companies or companies making non-moddable games.

In the end, offering no visual improvements and only some performance improvements, DX11 version of rF2 would not be as much different to DX9 version, as some of you may expect.

Here's something of an analogy: say you're a current bottom-feeder F1 team, 3-4 seconds a lap off the leading pace. Because of the recent engine rules your power is pretty good, but your aero is letting you down.

Suddenly, somehow, you're offered completely free use of Red Bull front wings. You know your current front wing sucks, and here's your opportunity to have quite possibly the best there is. Great, right?

Well, no. Those wings on your car will make it worse, because they won't work to shape the airflow over the rest of your car the right way. You'd have to go and redesign every other piece of bodywork in order to try and make use of the new front wing. You can't just plug in something that works better elsewhere and expect a proportionate gain.

And that's not even a good analogy, because in aero there's an absolute tangible difference between best and worst.

You can make a Dx11 game look fantastic without actually utilising any of the Dx10+ features, and get it looking exactly the same with nearly the same performance in Dx9. For sure if you make a game use Dx11 features and then try to emulate them in Dx9 you'll get a massive performance hit, but that's another matter entirely.

If rF2 doesn't use any of those features - and looking at what K has said, you'd have to say it seems unlikely at least in the short term - there'd literally be no point in moving to Dx11. It would just further delay more flexible qualifying formats, tyre set histories, a finished contact patch model, fixed collision modelling, better safety car behaviours, better AI, better sound, .... and as Tim points out, in the end people will still wrongly compare rF2 to other driving games (or games in an entirely different genre) and compare framerates while completely ignoring anything else that's going on.
 
Yeah, old-assed thread.

I also dabble in R3E, and just last night noticed that it's also running DX9.
And that game looks absolutely fantastic.

So there is PLENTY of room for rF2 to grow, visually.
 
i think its been demonstrated that given the right pair of eyes (several modded tracks in addition to some isi ones) or just a nice ToD/weather condition rf2 can look gorgeous. thats not the issue with dx9 vs dx11. oddly enough, the issue is dx12. and it sucks to hear rF2 wont be updating, because dx12 will offer a huge performance boost when it arrives.

my performance with rf2 isn't bad...at least judging by framerate...(im going to have to google 'microstutter' bc last night i had some weird issues) but its not where i feel it should be. and certainly not where it would be if the dx12 bottlenecking optimizations were magically in place already.
 
There is little difference visually between 9 and 11...effort is much better spent elsewhere. The only real advantage is that a copy of vram isn't kept in main memory for 11, as it is for 9. That is a big issue for 32-bit but as we are entering the 64-bit era for rF2 that is no longer an issue really.
 
Well something has to change. rf2 isn't exactly flooded with players, graphics been the number one reason.

that opinion could be changed with marketing & i suppose a little 'artistic vision' from ISI with the HDR. i maintain that rf2 can look as good as any sim of the moment, static or in motion, save probably pcars. a lot of this has to do with the support of these things like WEATHER, and THE SUN GOING DOWN, that we experience here on earth. make an ad, a video, stock it w some nice, intricate, high poly cars. use a familiar circuit. but of course yr not just showing off car porn, youre showing off an unpredented degree of dynamic racing available to the player. its huge; maybe pcars might really wow us but i doubt they'll have, say, realweather like we do.

i mean, i watched some of the 24 HR le mans stream & my god the sun rising after a good rainfall, seeing those cars go by in the wet, knowing it was all just happening there because it -really was all just happening there-, at that time....think it was the day or 2 after i picked up rf2 but what an unforgettable experience & one that easily would've sold me.
 
There is little difference visually between 9 and 11...effort is much better spent elsewhere. The only real advantage is that a copy of vram isn't kept in main memory for 11, as it is for 9. That is a big issue for 32-bit but as we are entering the 64-bit era for rF2 that is no longer an issue really.

DX9-DX11 has many improvements (some found in 10 and improved since 10) from efficiency to lighting techniques.
 
Well something has to change. rf2 isn't exactly flooded with players, graphics been the number one reason.

Different strokes for different folks. Vast majority of rF2 owners aren't here for graphics. If someone chose for example Grid Autosport as his most favourite game I guarantee you that there is no way for him to like rF2. rF2 aims at certain group of players for whom graphics are lower on priority ladder.

Things are changing all the time but perhaps in not way it pleases you. I can get that from your youtube comments about rF2. ;)
 
the thing it seems to me everyones missing is that the cards youre using to run this game -right now-, cards you probably will have pitched in 2015 for shinier, faster cards, even these cards are FULLY dx12 compatible. if it worked with 11, itll work with 12. the reason to upgrade to dx11 isnt marketing; thats just a side effect. the reason is pure performance. AMD fired a shot at microsoft with mantle. microsoft quickly fired back, & here we have DX12. notice you your single GPU is outputting only 70-80% of power or whatever? dx12 aims to greatly ease the chokehold the cpu can have on the gpu. this is -realliy- relevant stuff for simulators. i dont feel liike googling msfts estimates, i feel like theyd be mostly shrugged off anyway, but yeah, dx12 is real, its coming late 2015 (?) and its gonna kick ass. and its really, really ****ty that rf2 isnt gonna be able to climb aboard.


im not surprised bc i dont imagine overhauling an engine to such a degree is remotely simple. but, if we look at rf2 3-5 year project...it just starts looking very long in the tooth to me saddled with dx9, a platform that was dated back when multicore was barely a thing..
 
Not dramatically different differences...and that effort might be better spent elsewhere.

again, its not about making the game look better. if they want to market it t hat way too, fine. its simply about bringing it into the sa me place that iracing, assetto corsa, & hopefully pcars will occupying, which is a pretty much instant across the board massive gain in performance for everyone as soon as dx12 sdk hits. its raw performance thats the issue; this isnt about a beauty pageant.
 
iracing is dx9. They have been working towards dx11 for close to 2 years now and still haven't finished, so yeah, like c3po says, that effort might be better spent elsewhere.
 
theres no rush to get dx11 out. as said, it doesnt offer much over 9, until you glance around the bend. dx11 will allow for an easy transition to dx12 (still not due till, i think, late next year), which means simulators running dx11 & making that transfer are going to see a massive performance increase due to the cpu & especially gpu being able to run at a much cleaner, higher capacity than before.

mantles already proven this concept works, now add a years worth dev time by msft. your card doesnt have to change. suddenly youre gaining, idk, 10-20 FPS if your target was 60. bc now instead of being cockblocked the gpu is now able to run at 98-99% while the CPU does its thing as strong as ever;. eeez a big deal. it will ultimately allow for better graphics, of course...but as simracers i assume the main focus is on the performance gains we should see.
 
Well those are the claims for dx12. Call me skeptical but I remember these images for dx10, we all know they were a bunch of bs.


directx-9-vergleich-directx-10.jpg
 
Status
Not open for further replies.
Back
Top