Battlefield 3 Laptop Performance Review: Road Warrior?
Multi-player Performance, Conclusion
Alienware m17x: Multi-player Performance
Single player performance on the m17x is solid, but the campaign is hardly the reason why most players are attracted to Battlefield 3. That honor goes to the multi-player component. So how does our Alienware handle that?
To find out, I jumped in to some matches on the map Caspian Border on a 64/64 player server. Obviously, I could not standardize the benchmarks here, so I simply took benchmark samples from standard gameplay.
The benchmark results from mutli-player are roughly equivalent to those from the campaign, but both low and medium detail took a hit. Low, in fact, lost almost 10 FPS compared to the campaign.
On the other hand, ultra and high barely saw any decrease. This suggests to me that there is some additional overhead in the multi-player component that’s not related to rendering graphics that slows the game slightly, which is hardly a surprise. Keeping track of 64 players can’t be easy.
In my gaming experience, I found high to be the best blend of graphics quality and performance. I never once found myself feeling as if the gameplay was being held back due to a low framerate. Some gamers may disagree with this, but unfortunately, they don’t gain much from downgrading to medium detail. Only low offers noticeably better performance, but the difference in visual quality is massive. I’m not willing to make that trade in a game as beautiful as this.
A Quick Comparison
So, we now know how Battlefield 3 performs on the m17x, but there is no frame of reference. Is this good, or bad? To help with context, I ran Just Cause 2, which up until now was the most demanding game routinely used by PC Perspective to test laptops.
Normally, I test Just Cause 2 using a built-in benchmark. Since Battlefield 3 was not tested this way, however, I felt it would be better to use actual gameplay. Details were set at high as possible with the exception of SSAO and Point Light Specular, which were turned of. This is roughly equivalent to high in Battlefield 3, and the best approximation I can manage, as most of Just Cause 2’s settings have no medium or ultra/extreme option.
As it turns out, Battlefield 3 at high is much more demanding than Just Cause 2 at similar settings. Visually, however, I'd sat Battlefield retains an edge. Just Cause 2 clips out objects like vehicles at long ranges, while Battlefield 3 does not. Lighting is also obviously better than Just Cause 2, which has a relatively bright and cartoonish appearence.
Overall, Battlefield 3 performed well on the Alienware m17x. If you currently own a gaming laptop built within the last two years, I suspect you will have no trouble playing Battlefield 3 at low or medium detail. Those with recently purchased gaming laptops may be able to handle high. Ultra is available, but is unlikely to be enjoyable on any modern laptop besides those with dual graphics.
Our comparison to Just Cause 2 suggests that Battlefield 3 will be a very difficult game for most laptops to play. Laptops with integrated graphics or entry-level discrete graphics are rarely able to play Just Cause 2 smoothly even at low detail settings and 1366x768 resolution.
If you’re on the fence about buying Battlefield 3 because you are concerned about your laptop’s hardware, I suggest that you wait. If you do not own a gaming laptop, you should be prepared for the possiblity that smooth framerates will be unavailable at any setting besides low. Even then, older machines with discrete graphics may have difficulity.
We will be using the game as part of our laptop benchmark suite from now on, and we’ll be able to form a more accurate picture over time.
Update: Low and Mid-Range Performance
Since I originally wrote this article, I’ve had the chance to test Battlefield 3 are two systems that are much more representative of low and mid-range graphics. One was the ASUS U36S, and the other is the ASUS K53T.
The ASUS U36S, equipped with a Core i5 processor and Nvidia GT 520M graphics, is representative of low-end discrete graphics. The ATI Radeon HD 6470M is close in performance, and the newer Nvidia 610M is just a little bit quicker. After updating the drivers on the Nvidia GT 520M, we managed to achieve an average framerate of 17.4 FPS at medium detail, while low detail chugged along at a still-not-quite playable 23.7 FPS. Both settings felt as choppy as the averages would suggest, and bursts of particle effects (common in this game) seemed to cause slowdown from time to time.
What this tells us is the Battlefield 3 is too demanding for low-end integrated and discrete graphics solutions. Most systems will be able to start the game and play it, but your experience will not be enjoyable, and the frame rate is low enough to reduce your ability to play the game well.
Next up is the ASUS A53T, which is equipped with Radeon HD 6720G2. This is a combination of a Radeon HD 6650M discrete graphics component with the AMD A6-3400M APU in CrossFire. A Radeon HD 6650M is not a bad discrete GPU by itself, so this combo is representative of mid-range discrete graphics for laptops. Parts like the Radeon HD 6570M, Nvidia GeForce GT 540M and Nvidia GeForce GT 630M are in the same ballpark.
Battlefield 3 was much smoother on this system, achieving 35 FPS at medium detail settings. Although there were some noticeable moments where long frametimes caused jerkiness, such issues were not so common as to spoil gameplay. Overall, playing BF3 on medium was an enjoyable experience, and it didn’t look bad, either.