gamegpuBattlefield 4 Benchmarks

[Final8ty]

View Poll Results: Do you use Post AA (FXAA) in BF4?
I enable Post AA (any level) (or leave it enabled as the default setting) 35 26.12%
I disable Post AA 99 73.88%

http://hardforum.com/showthread.php?t=1788367

 

[myfriend]

This, pretty much. When I have that type of disposable income, I plan to upgrade to the top card every one or two years.

Back to something a bit closer to the topic, I just don't feel that this game scales as well as it should. An ideal game has reasonable settings for everyone. Someone who spends $400+ on a graphics card is not going to be playing at 1080p. I might be all wrong though. Who knows?

hey, so according to these benchmarks you can play on setting ultra with msaa at 62 average fps so no need to spend extra cash; So get this game and enjoy it

 

[Contra9]

This, pretty much. When I have that type of disposable income, I plan to upgrade to the top card every one or two years.

Back to something a bit closer to the topic, I just don't feel that this game scales as well as it should. An ideal game has reasonable settings for everyone. Someone who spends $400+ on a graphics card is not going to be playing at 1080p. I might be all wrong though. Who knows?

Whoa, hold on. I have a comparable system to yours but is older still. I have a 2500k @stock, 8GB 1600 ram and a 7950 boost at stock clocks (925 constant) and I get an easy average of 60fps at 1080p ultra with no msaa.

I just installed Fraps last night to check it. There were some dips to high 40's but it wasn't for long. It is also smooth as silk and I stomp people. I spend most of the time on foot in the middle of fights so I would feel it if my system wasn't enough.

The truth about pc gaming these days is the hardware is way stronger than the games require up to 1080p. Graphics cards and cpus have stagnated badly the last few years, thus the ridiculously high prices. It used to be a new $400 card would offer a total new experience, now a $600 card gives 20% more fps at the same settings. Consoles are mostly to blame, but resolutions haven't gotten higher in a widespread way. I pc game a lot and generally stay 2 years behind picking up bargains. I will continue doing it until something drastic changes.

Just enjoy the games at whatever settings you feel comfortable with and avoid hardware forums.

 

[lavaheadache]

I know, but all of the faster "new" cards are over $400. But yeah, I know, any "working adult" should have no problem buying graphics cards every few months.

Having to spend $400+ for 1080p is insane to people who don't take money for granted.

not sure why you just popped an attitude with me but if that's your thing, cool.

I'm pretty sure I was making the point that just because your system is new to you doesn't mean that it is new tech and able to push the latest and greatest like you want it too. Heck, Even the latest and greatest leaves desire for more.

 

[lavaheadache]

The fact that you're acting as if everyone has that kind of disposable income proves my point. I don't see the point in upgrading more than once a year other than bragging rights.

wow, I see that you are blowing his spot up too. Niiiiice, good way to make friends around here. Nobody is attacking you or implying that you suck cause you have a 7950.

 

[Makaveli]

Screw them all. Lets do it ourselves. Get a few guys with different CPUs, lower details and go to same spot on map at same time, same server and report the numbers.

I don't have battlefield 4 yet but I would be game to do some testing and would purchase it.

 

[Silverforce11]

So Techspot, Sweclockers and now Guru 3d all consistent with each other in results.

The odd one out, MP getting heaps faster than single player is pure wrong. BF MP especially 64 players has always been more punishing.

It looks like R290X is already schooling Titan, WITHOUT Mantle... makes sense when the rep called it a "ridicule" with Mantle active.

 

[moonbogg]

I don't have battlefield 4 yet but I would be game to do some testing and would purchase it.

Alright cool. Let me know and we'll take some numbers down. I get frustrated with how review sites do things. They play it safe by testing single player. They also test CPUs with massive GPU bottlenecks, so all the numbers are exactly the same with different CPUs. Totally pointless. I get why they do it, but its still worthless.

 

[finbarqs]

are there any built in benchmarks for BF4? I have 4770K @ 4.2GHz, and I'm probably gonna push my gtx 780 just to hit > 60 FPS @ 1680x1050p

 

[Grooveriding]

are there any built in benchmarks for BF4? I have 4770K @ 4.2GHz, and I'm probably gonna push my gtx 780 just to hit > 60 FPS @ 1680x1050p

There is one but it doesn't work currently. My guess it suddenly starts working December... when Mantle arrives for Battlefield 4 :sneaky: AMD is already putting a beating on nvidia in BF4. Should be interesting to see what Mantle does as well.

 

[Insomniator]

Wow glad to see the 7950 doing so well!

 

[96Firebird]

Any other MP results? Seems most of the benches are from SP...

 

[skipsneeky2]

Alright cool. Let me know and we'll take some numbers down. I get frustrated with how review sites do things. They play it safe by testing single player. They also test CPUs with massive GPU bottlenecks, so all the numbers are exactly the same with different CPUs. Totally pointless. I get why they do it, but its still worthless.

Will be benching my 770 and my cpu in sig tomorrow at 1366x768,will give you some numbers as well as i need to adjust some settings once i install the game and play.Aim usually for a 60+ minimum experience so if i can get it i will let you know.

Wifes laptop screen is busted so shes using my u2412m for the moment,rocking a 1366x768 monitor for the mean time till the lap top is fixed so basically all my games are currently cpu bottlenecked atm.:awe:

 

[Will Robinson]

Next time take 800x600.



Butthurt.

Actually that chart is at 1920x1200 which...amazingly is the same resolution as my monitor.

Clearly you are upset....don't worry its just a game and these are early benchmark results anyway.
I'm sure you'll enjoy the benchmarks a lot more when the Mantle update is applied in December.

 

[Carfax83]

So Techspot, Sweclockers and now Guru 3d all consistent with each other in results.

The odd one out, MP getting heaps faster than single player is pure wrong. BF MP especially 64 players has always been more punishing.

It looks like R290X is already schooling Titan, WITHOUT Mantle... makes sense when the rep called it a "ridicule" with Mantle active.

Just to point out, the fact that AMD has the edge this early in the stage doesn't mean anything, because a patch and or driver update could easily swing things in NVidia's favor; especially when the game is obviously so unpolished and unrefined, and in need of multiple patches.

 

[f1sherman]

So Techspot, Sweclockers and now Guru 3d all consistent with each other in results.

The odd one out, MP getting heaps faster than single player is pure wrong. BF MP especially 64 players has always been more punishing.

It looks like R290X is already schooling Titan, WITHOUT Mantle... makes sense when the rep called it a "ridicule" with Mantle active.

Indeed, Nvidia's driver performance is classy as allways, but AMD is also looking great in single player.

Will be even better with Mantle no doubt :ninja:

 

[Silverforce11]

Indeed, Nvidia's driver performance is classy as allways, but AMD is also looking great in single player.

Will be even better with Mantle no doubt :ninja:

Sweclockers did MULTI-PLAYER (Siege of Shanghai) benches, conquest mode.

 

[24601]

I don't know about you, but I'm getting (more than) 20 fps less in 64 player siege of shanghai conquest now compared to beta, and yes, I'm running the latest drivers.

I had to drop down to Ultra with no MSAA and Low shaders to keep my minimums above 60 fps Q.Q (Dips to 55fps or lower sometimes on some maps though)

Also, all blur off in config file and single buffering with 0 frame render ahead, vsync off, 90 FoV

This is all with a single 7970 @ 951 core 1450 ram, 3570k@4.5ghz, 32gB ram

With all blur off in config file you really get to see how horribad this game looks compared to Crysis 3.

Also, if you ain't playing 60+ player Battlefield, you're playing Call of Duty. Get your consolitis CoD self outta here :O

Mantle is SORELY needed for 64 player Battlefield (For AMD/ATi cards). I hope it actually works when it comes out, and that it actually comes out on time.

BTW, its quite funny, either the game or the driver turns off tessellation and normal mapping for a model once it reaches about 45 degrees turned. You can see this if you turn your character model in the weapon selection menu. IDK if this is a Battlefield 4 thing that happens with Nvidia cards as well, or if it's AMD/ATi's "optimization" for tessellation.

 

[selni]

Just to point out, the fact that AMD has the edge this early in the stage doesn't mean anything, because a patch and or driver update could easily swing things in NVidia's favor; especially when the game is obviously so unpolished and unrefined, and in need of multiple patches.

This, the game is all but unplayable currently - no surprise given that it's EA and that the BF3 release was bad too, but it's clearly unfinished/unstable (mantle coming in December instead of release is another pretty obvious sign of that). I wouldn't get too excited about benchmarks yet, actually playing MP for more than 15 minutes without crashing seems rare so far.

 

[Grooveriding]

I see so many complaints about crashing but I do not get any crashes except for on Shanghai when the building comes down which is a known bug.

Also if you are trying to play 64 player games on an i5, there is your problem. Time to get an i7 unless you want poor performance. This game is unforgiving to CPUs without HT or 6 cores +.

 

[5150Joker]

So Techspot, Sweclockers and now Guru 3d all consistent with each other in results.

The odd one out, MP getting heaps faster than single player is pure wrong. BF MP especially 64 players has always been more punishing.

It looks like R290X is already schooling Titan, WITHOUT Mantle... makes sense when the rep called it a "ridicule" with Mantle active.

Why are you comparing R9 290X with Titan? According to you Titan isn't even a gaming card. :whiste:

I see so many complaints about crashing but I do not get any crashes except for on Shanghai when the building comes down which is a known bug.

Also if you are trying to play 64 player games on an i5, there is your problem. Time to get an i7 unless you want poor performance. This game is unforgiving to CPUs without HT or 6 cores +.

I was getting a ton of crashes at first and realized it may be my overclock. Once I raised the CPU voltage, 99% of the crashes stopped but I still do get one here and there. I suspect it may be the same with others who have frequent crashes in BF4.

 

[24601]

I see so many complaints about crashing but I do not get any crashes except for on Shanghai when the building comes down which is a known bug.

Also if you are trying to play 64 player games on an i5, there is your problem. Time to get an i7 unless you want poor performance. This game is unforgiving to CPUs without HT or 6 cores +.

People are trying to say 8350 is providing equivalent performance to i5s, and that just isn't true except in non-cpu bottlenecked situations and on Nvidia GPUs.

 

[Grooveriding]

I think that is it as well in some cases. The game is so intensive it is exposing unstable system overclocks.

 

[3DVagabond]

Butthurt.

So a card being 2x faster in MP vs. SP makes sense to you? Please explain for the rest of us.

 

[Silverforce11]

Why are you comparing R9 290X with Titan? According to you Titan isn't even a gaming card. :whiste:

Its just for you, since you love Titan as a gaming card since its "better" than the 780.. point is, its walking over Titan already, it stomps all over the 780.

NV needs the 780ti to be Titan +20% if they hope to compete with custom cooled R290X that doesn't throttle.. I really hope its awesome, because for $699 if its any less its meh.