• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Rocket League - Four Player Split Screen Performance

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Dlaw

Member
Joined
Apr 26, 2012
Location
New York, USA
Split from here by ATMINSIDE

Question:

I'm in the process of doing the same type of build. You say you're running 4k with 2x 290x in x-fire. Do you have the 4Gb versions? What FPS are you seeing? My understanding was that 4Gb wouldn't be very smooth @ 4k.

I can tell you that at 4K, the only time I've maxed my 4GB 980s is when running benchmarks. Playing 4 player split screen Rocket League maxed with 8xAA is the only thing that I play that gets close (~95%). I've heard some games (GTAV) can use enough vRAM to make 980Ti's page out, though, so it is a consideration.
 
Last edited by a moderator:
OP
D

Dlaw

Member
Joined
Apr 26, 2012
Location
New York, USA
What settings though? Are you disabling AA (as it isn't really needed at 4K)?

I turn AA down if I need to, but if I don't need to I leave it maxed. It's always the first setting I tweak to get more frames if I need them, though.

Also, AA does help at 4K to a point. I can see jaggies if I turn it off, particularly if an object is against a black background. I can't tell a difference between any level or type of AA, though, so I usually leave it at 2x or whatever the quickest setting is.
 
OP
D

Dlaw

Member
Joined
Apr 26, 2012
Location
New York, USA
Rocket League isn't even close to a demanding game.

It is when you do 4 player split screen. My GPU cores are pegged at 99%, and my CPU is anywhere between 80-100% usage. Displaying 4 different 1080p images at the same time takes quite a bit to do.
 

Blaylock

"That Backfired" Senior Member
Joined
Jun 5, 2013
Location
Go Blue!
If it's 4 player split screen like you mentioned earlier it's only a single 1080p image. If it was hooked to 4 different monitors then it would be 4x1080p.
 
OP
D

Dlaw

Member
Joined
Apr 26, 2012
Location
New York, USA
If it's 4 player split screen like you mentioned earlier it's only a single 1080p image. If it was hooked to 4 different monitors then it would be 4x1080p.

No, you're both misunderstanding here. I play at UHD, on both my main monitor and living room TV (see my signature). When I play with my family, we play split screen (4 player). Therefore, every player has a corner of the tv, with a resolution of 1920x1080 (1080p). When my computer has to render 4 different 1080p images, I see the usages I stated above.

Edit: I just reread that, and realized it may have come off a little rude. I wasn't trying to make it seem like that, just was trying to type it out as coherently as possible while keeping it easy enough to type on my phone.
 
Last edited:

EarthDog

Gulper Nozzle Co-Owner
Joined
Dec 15, 2008
Location
Buckeyes!
But it's still 4k...and rendering the same amount pixels.

Is there a big difference between the two as far as gpu/cpu use? FPS?
 
OP
D

Dlaw

Member
Joined
Apr 26, 2012
Location
New York, USA
The same amount of pixels as 4 1080p screens ;P

Yes, here's a shot of HWMonitor idle:
idle.png

One after playing 1 player (exhibition, with 7 bots, ~1 minute):
1player.png

And one after playing 4 player split screen (exhibition, 4 bots, ~1 minute):
4player.png

Usages are considerably higher in 4 player mode. CPU usage was a little lower than last time I tested this, maybe because there was less scoring this time? I don't know, but as you can see, both GPUs are +90%. I could have sworn it used more memory, but that was a day that I ran a lot of benchmarks, so I may have forgot to restart HWMonitor before the test.

As far as FPS, the game takes a big hit when playing 4 player with everything (including AA) maxed, which is what I did for this test. I can run it with up to 3 people like this without any FPS issues at all (stays pegged at 60), but as soon as I add a fourth, I see drops in the 20s, with average FPS staying in the 30 range. This is with the GPUs at stock, by the way.

Also note, that I see different usages based on what arena I'm on, and also that these usages are fairly constant (as read by IXTU's and GPU-Z's graphs) during the match.
 

EarthDog

Gulper Nozzle Co-Owner
Joined
Dec 15, 2008
Location
Buckeyes!
The same amount of pixels as 4 1080p screens ;P
Which is 4k uhd, no? 3840x2160. Same amount of pixels.

Not sure why there are no fps dips when you go three, but see huge drops with 4. It has to render 3 unique, wonder what make 4 different?
 
OP
D

Dlaw

Member
Joined
Apr 26, 2012
Location
New York, USA
I do have V-Sync enabled, as I don't need more than 60fps. I wonder if it's proportionate, in that I lose a percentage of fps with every player added. Maybe I'll check it out with vsync off some time.
 

wingman99

Member
Joined
Dec 10, 2003
V-Sync Just limits the max FPS to your monitor refresh rate, so you won't have more FPS with it disabled with every player added.
 
Last edited:

EarthDog

Gulper Nozzle Co-Owner
Joined
Dec 15, 2008
Location
Buckeyes!
I do have V-Sync enabled, as I don't need more than 60fps. I wonder if it's proportionate, in that I lose a percentage of fps with every player added. Maybe I'll check it out with vsync off some time.
I'd like to see those results, particularly with fps. :)

If it's proportionate, you have to getting really high fps on single or dual. Pegged at 60 down to the 20s is a huge drop.
V-Sync Just limits the max FPS to your monitor refresh rate, so you won't have more FPS with it disabled with every player added.
the point was to see how the fps dropped with adding more players. ;)
 
Last edited:
OP
D

Dlaw

Member
Joined
Apr 26, 2012
Location
New York, USA
I'd like to see those results, particularly with fps. :)

If it's proportionate, you have to getting really high fps on single or dual. Pegged at 60 down to the 20s is a huge drop.
the point was to see how the fps dropped with adding more players. ;)

These are my thoughts as well. I'll see about checking it out tomorrow.
 
OP
D

Dlaw

Member
Joined
Apr 26, 2012
Location
New York, USA
I'm running these tests right now, here're the results:

1 player, FXAA set to high, settings maxed on Wasteland for 1 minute:

High-140
Low-80
~Average-115 (what I saw the most)

2 players, same settings:

High-110
Low-80
~Average-95

3 players, same settings:

High-95
Low-74
~Average-84

4 players, same settings:

This is really interesting:

High-51
Low-22(!)
~Average-33-35

If someone else could do some testing on this game and confirm this? This almost looks like some kind of optimization issue at 4 players, doesn't it?
 

EarthDog

Gulper Nozzle Co-Owner
Joined
Dec 15, 2008
Location
Buckeyes!
Yeah, that is a BIG drop there for whatever reason... wish I could help, but I don't own the game.
 
OP
D

Dlaw

Member
Joined
Apr 26, 2012
Location
New York, USA
Yeah, that is a BIG drop there for whatever reason... wish I could help, but I don't own the game.

The weirdest thing about it is how consistent 1-3 players is in framerates, and then the massive drop at 4 players. These are not the results I was expecting at all.
 

EarthDog

Gulper Nozzle Co-Owner
Joined
Dec 15, 2008
Location
Buckeyes!
Yeah, I would call that 'something wrong' with 4. Perhaps reach out to the makers? Or see what the google machine says are my only suggestions.
 

ATMINSIDE

Sim Racing Aficionado Co-Owner
Joined
Jun 28, 2012
Maybe you're running out of CPU for the 4-player portion... try messing with the overclock (raise it) and see if you pick up the minimum framerate.