Star Wars: The Force Unleashed II
|Xbox 360||PlayStation 3|
|Surround Support||Dolby Digital||Dolby Digital, 5.1LPCM, DTS|
We got quite excited about this one. When LucasArts released the Xbox 360 and PlayStation 3 demos for the game, it was clear that the tech team had done a phenomenal job in creating a state-of-the-art rendering engine: deferred lighting, custom anti-aliasing, advanced motion blur... It may not have reached the giddy heights of the technically brilliant God of War III engine, but there's no doubt that The Force Unleashed II demo looked great, played well and that all augured well for the full retail game.
Unfortunately, the fact that the shipping release was so short, so repetitive and yet still felt as though some levels were artificially prolonged suggested that while the core engineering team knew their onions, the game designers had come up short. Still, there remains little doubt that this is a seriously good-looking game and certain levels beyond the demo stage really emphasised the strengths of the engine. As a game, we weren't massively impressed, but as a tech demo, it still managed to hold our attention.
As expected, the final release proves to be an especially close conversion.
In-keeping with the demo, we see no real differences between the two games during the run of play, but it is clear that the minor visual refinements are seen on PlayStation 3. The SPU-driven motion blur operates with more samples than the GPU-based Xbox 360 implementation, while the custom directional anti-aliasing (DLAA) is ever so slightly more refined on the Sony platform.
In cut-scenes, we see that the Xbox 360 game has additional texture detail on the characters' faces (called "tension maps" by the developers) but this is nothing more than a bug and if it hasn't been fixed in a patch by now, expect it to updated when the first PS3 DLC drops - this analysis was carried out several weeks ago and alas our copy has been returned so we cannot check the recent 1.01 patch.
In truth, as The Force Unleashed II plays out there is very little difference between the look of the two games - only on fast-shifting pans does the difference in the motion blur really look that much better on PS3, while any DLAA differences only seem to be noticeable upon forensic analysis of static screenshots.
The DLAA system the Lucas engineers employ is generally excellent, and proves that a state-of-the-art custom anti-aliasing system can be deployed in a cross-platform game, but in the higher contrast areas it does seem to be the case that the same kind of problems MLAA has with sub-pixel edges proves to be challenging for this new technique too.
Regular Digital Foundry readers may recall that Lucas engineer Dmitry Andreev responded to our demo analysis, explaining some of the differences but apparently taking issues with our assessment of performance: that here the Xbox 360 version commands the advantage. Andreev suggested that the only way to effectively measure performance is in totally like-for-like conditions, and he reckoned that we'd see both versions out-performing each other at different points.
So to kick off with, that's exactly what we did.
What we see is a remarkably level 30 frames per second with only the odd drop in performance. We do see the occasional series of frames go over-budget, resulting in screen-tear, and while it is the PS3 that more often tears, Andreev is right: there are areas where the PS3 performs better. More importantly, in almost all cases, there are very few instances where the tearing descends from the top of the screen into an area the player is likely to notice.
However, while a comparison of cut-scenes is a valid exercise in comparing engine performance, the snippets of gameplay we have included in the test above really don't do justice in analysing the overall experience across the run of play. Here, the best test remains exactly what we did for the demo - picking out sections of gameplay from the same areas and comparing them all in order to get a feel for general performance. Exact like-for-like comparisons of specific rendering scenarios aren't possible of course, but as an overall feel for how the game plays and in what instances frames are dropped, the testing is perfectly valid.
Again, we do see very similar scenes where PlayStation 3 outperforms 360, but there's little doubt that as play continues, the PS3 tears much more often and additionally we see some "hot spots" where frame-rate takes a significant hit while remaining fairly consistent on the Xbox 360.
Thankfully these hot spots are few and far between and while the additional tearing on PS3 is sometimes noticeable, it does not really affect your enjoyment of the game. What we have here is a situation similar to James Bond 007: Blood Stone where we have two very similar games each with the own strengths and weaknesses: PS3 enjoys a marginally higher level of image quality at the expense of some performance. However, while there really isn't much in it, the additional tearing and dropped frames on the PS3 game are more noticeable than the lower quality motion blur, and so on balance, the Xbox 360 rendition of The Force Unleashed II is probably the one to get if you happen to own both consoles.