*******C.A.R.S - Community Assisted Race Sim*********

It's not like the game has suddenly come from nowhere and been released! It has been in development for years so I would have thought Amd would have optimised it if possible. Those benchmarks are just so wrong it's laughable!
 
Gameworks will get all the blame for poor AMD performance, when on other titles like Dying Light AMD cards perform fine with those settings on.
 
http://www.reddit.com/r/pcars/comments/351ht0/pcars_benchmarks_from_final_build/

seems like amd performance is still far from 'ok' now.... Shocking really.

That benchmark is distorting the comparison between AMD and Nvidia, look at the components used for the test. ;)

pCARS' physics engine, running at 600Hz (twice that of some other sim titles, such as AC), needs a lot of CPU power and excellent communication with memory.

I get a framerate increase of ~10fps (~20%) by running my system (spec in sig) with the Corsair DDR2 overclocked to 1000MHz, with my Q9300 overclocked to 400MHz FSB.

You don't have to run all pCARS "performance" settings at "ultra" to have amazing graphics in the game, just like the Trackmania series, pCARS has graphics options on the pc version that are beyond typical (not high end enthusiast) gaming machines.

I use a mix of "low" and "medium" settings and even with a low end AMD gaming graphics card and pCARS looks better than any race sim of the last 15 years, it plays well too. ;)

Given I don't game that much these days as a 40-something, I'm torn whether to spend ~£600-£1000 on a new system, simply to turn up pCARS graphics to a mix of "high" and "ultra."
 
Doesn't that depend on what's been implemented? Two different games with two different engines etc.

It could play a part but not the massive differences we're seeing in those benchmarks. It tells me the devs have either completely overlooked AMD hardware (which would be incredibly surprising considering the game on consoles runs on AMD) or AMD haven't worked with them much if at all on the PC version.
 
It could also be down to certain settings like detailed grass for example. That kills fps -in earlier builds at least- on AMD cards with no noticeable gain in looks.

They really do need to post screens of their exact settings, rather than just give a vague 'high/ultra'.

Anyway, I've been enjoying better performance on earlier builds with higher settings than what's being shown on those charts (and an earlier build performance chart from the same site I think). My preorder got shipped today, so we'll see what's what soon enough.
 
It's not a case of turning some settings down to get better performance. The problem here is that the top end amd gpus are performing on par with lower end nvidia gpus. This shouldn't be the case. The game has been delayed so much already and there shouldn't be this type of gap in performance between red and green. Assetto isn't far off graphically than pcars and I get a superb frame rate just by dropping shadows and smoke down a notch. This is with a full grid.
 
or..nvidia gave them aload of money :p

lol, the banners all over the tracks pretty much make that a given :p

It's not a case of turning some settings down to get better performance. The problem here is that the top end amd gpus are performing on par with lower end nvidia gpus. This shouldn't be the case. The game has been delayed so much already and there shouldn't be this type of gap in performance between red and green. Assetto isn't far off graphically than pcars and I get a superb frame rate just by dropping shadows and smoke down a notch. This is with a full grid.

If you think it's that simple, then I don't know what to tell you.

You really think a 660ti beating a 290x is purely down to crappy driver implementation?
 
Last edited:
I can pretty much max the game at 1080p with a GTX 970, 8GB RAM and an i5 4690k @ stock and maintain 60 fps.

AA, high detail grass and shadows have the biggest impact on performance.
 
Back
Top Bottom