-
Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.
Digital Foundry Project Cars 2 Performance Analysis
- Thread starter danowat
- Start date
blastprocessor
The Amiga Brotherhood
LOL good driving DF. Disappointing they can't lock FPS on Pro.
adamsapple
Or is it just one of Phil's balls in my throat?
Watched the video. Good Pro upgrades.
The worst is 59~58 and that too in stress locations with 32 cars and weather effects. That's a pretty good turnaround. Not sure why you'd count that as a disappointment.
The worst is 59~58 and that too in stress locations with 32 cars and weather effects. That's a pretty good turnaround. Not sure why you'd count that as a disappointment.
Bitch Pudding
Member
Money well spent.
BigEmil
Junior Member
Exactly talk about hyperbole
matrix-cat
Member
Really pleased with that performance on Pro. Literally had to max out the car count and drive through a thunderstorm to get it to drop to 58FPS 
sems4arsenal
Member
The Pro doing well here. The OG PS4 not so much.
soundscream
Member
So that other thread showing a locked 60fps on og ps4 has shown that candyland is fakenews.
BigEmil
Junior Member
No it's just DF mostly focuses on the more demanding stressed scenes while CandyLand focuses on normal moments in gameplay
matrix-cat
Member
I imagine it's much like the first game, where performance is great in clear weather with maybe eight cars or fewer, but cranking up the car count and weather is a recipe for tanking the framerate.
Looking at the graph that looks more like 40fps (and feels like it when watching the video) than the 59.1 from the fps counter
adamsapple
Or is it just one of Phil's balls in my throat?
On OG PS4 and XB1, yeah. Pro is very good at maintaining 60 FPS 99% of the time.
soundscream
Member
Regular ps4 was dropping below 60 in the non stress test parts, candyland showed a locked 60.
I'd have said it looks way better than the first game in terms of performance.
Liabe Brave
Member
No, it's not the footage they chose to analyze; there's crowded, wet conditions in there, and it does drop to ~45fps.
The problem is with Candyland's framerate analysis tool. It's clearly looking for unique frames by testing the top portion of each against the previous. But in a non-synced title like PCars 2, at lower fps the top of each frame is still unique...but then somewhere in the middle it tears, and the lower part isn't redrawn. Thus their tool counts 60 new frames all the time, but some frames aren't wholly new.
The VG Tech channel is better for this stuff. Their analysis tool keeps track of torn frames as well as base framerate. (Of course, they don't cover all the same games.)
Jack The Nipper
Banned
That driving most things do not make me laugh but the ideas here...wow.
Dude being so serious and bam bam,you. Haha
This needs a review of the review video. Gifs please
What an intro.
Dude being so serious and bam bam,you. Haha
This needs a review of the review video. Gifs please
What an intro.
Zombie James
Banned
Agh, I didn't see this thread. Anyways, here's my summary if you want to add it to the OP:
- PS4 Pro: Dynamic 1360-1440p.
- PS4 Pro: Holds 60fps with 16 cars.
- PS4 Pro: Switches to adaptive v-sync when things get demanding with screen tearing.
- PS4: Dynamic 864-1080p.
- PS4: Performance similar to first Project Cars. Around 40-60fps with frequent or constant screen tearing.
- XBOX: Dynamic 864-972p.
- XBOX: Performance around 30-55fps with constant screen tearing.
- All: Replays capped at 30fps with bad frame pacing.
- PS4 Pro: Dynamic 1360-1440p.
- PS4 Pro: Holds 60fps with 16 cars.
- PS4 Pro: Switches to adaptive v-sync when things get demanding with screen tearing.
- PS4: Dynamic 864-1080p.
- PS4: Performance similar to first Project Cars. Around 40-60fps with frequent or constant screen tearing.
- XBOX: Dynamic 864-972p.
- XBOX: Performance around 30-55fps with constant screen tearing.
- All: Replays capped at 30fps with bad frame pacing.
My bad, it's 12GB patch [on PS4].
Yautja_Warrior
Member
Yeah strange, I noticed it too. The graph definitely hit the 40fps line for a second or two but the fps number in the top right was reading high 50's. Bug in the software or something else ?
Would be great if DF / dark10x could shed some light on this.
NullPointer
Member
Probably for the best at any rate, to see if there is hitching on collisions or weirdness when slipping off the track.Watching now, the person driving is so bad
Looks pretty solid, except for the tearing under weather conditions or full grids. Really do wonder whether a "performance" preset at 1080p would help on the Pro.
I've never understood their graphs in these videos. I just go by the voiceovers and footage.
I get the impression, but am not absolutely sure, that every race start would be considered demanding - what with the full grid all clumped up and all the bumping on the first turn.
FIREKNIGHT2029
Member
Nice to see the Pro handling it well. Staying 60fps a majority of the time is excellent news. I'll be getting this on my PC however. My PS4 Pro will get to shine when GTSport releases.
Dictator93
Member
Those low resolution half refresh rate planar reflections in rainy conditions look straight garbage on console. I understand the need to make them that way given the console specs, but man, that looks rough.
Dave made an error on one of the exports. Our software has a lot of different modes and adjustments that need to be made based on the content and he simply used the wrong combination of settings so the graph appears incorrect. The actual frame-rate number is right, though. I think the error is only present in one of the Pro clips.
Yautja_Warrior
Member
Ok thanks, good to know the framerate number is correct and performance really is great on Pro.
adamsapple
Or is it just one of Phil's balls in my throat?
Thanks for the clarification, good to know. I'm sure if the frame rate was dropping in the mid 40's, Dave would have pointed it out even if the frame rate counter was showing 60.