By using this site, you agree to our Privacy Policy and our Terms of Use. Close

Forums - Gaming - Head2Head: Splinter Cell: Conviction vs. Metal Gear Solid 4 Performance

Source: http://www.lensoftruth.com/?p=26459

“Which Stealth Action Exclusive Assassinates the Competition?”

Head2Head – Splinter Cell: Conviction vs. Metal Gear Solid 4

Length: 00:12:51

Which Stealth Action Exclusive Assassinates the Competition? Watch the video to find out.


“So, you’re saying I’m the dominant one” This week we’d like to dedicate this Head2Head to our fans who’ve requested this comparison for some time now. We’ve listened to your requests and we delivered. Today, we infiltrate our way through Splinter Cell: Conviction and Metal Gear Solid 4: Guns of the Patriots to see which exclusive is considered the epitome of stealth action. Smoke ‘em if you got ‘em, because your in for one wild ride.



Graphics: Graphically, both games looked absolutely amazing throughout but with having a two year gap in release dates, one would think Metal Gear Solid 4 would be lacking, well that simply wasn’t the case. First up, we know that Splinter Cell: Conviction is using a heavily modified Unreal Engine 3 while Metal Gear Solid 4 utilized a proprietary engine, so graphical expectations for both were high.  Although Splinter Cell: Conviction came out two years later, Metal Gear Solid 4’s native resolution was higher. Here are the facts. Metal Gear Solid 4’s native resolution was 1024×768 (2xAA, temporal), while Splinter Cell Conviction’s native resolution fell a little short at 1024X576 (2xAA).



While Metal Gear Solid 4’s native resolution was slightly superior, Splinter Cell: Conviction leveled the playing field with its environments. There’s no doubt,  environments in both games were highly interactive beyond anything we’ve experienced in a stealth game thus far.  However, we felt that the environments in Splinter Cell: Conviction were slightly more destructible and polished throughout. For example, texture quality in Splinter Cell: Conviction (including normal maps and specular maps) sported a higher resolution throughout, but that’s to be expected when running through the Unreal Engine 3 architectures.

Furthermore, Splinter Cell: Conviction encouraged the player to shoot out lights, climb out windows, and destroy literally every object in a room to remain concealed. Don’t get us wrong here, Metal Gear Solid 4 had a slew of ways to remain concealed, but Splinter Cell: Conviction took this aspect to the next level. However, we do acknowledge the two year gap between development cycles, so improvements in this area were expected.  What we didn’t expect will be revealed next.



Performance: On to the main point of this article – the performance differences. As seen in our analysis video it’s clear that there is a definitive winner in the performance category. While both games managed to hover around 30 frames a second most of the time, Metal Gear Solid 4 had absolutely zero screen tearing.  Although Splinter Cell: Conviction did have screen tearing, it mainly occurred in cut-scenes keeping the gameplay smooth. Below are our average frame rates and screen tearing percentages our analyzer captured.


Metal Gear Solid 4 Frame Analysis Splinter Cell Conviction Frame Analysis
Clip 1:
Length of clip: 8282 frames
Average FPS of clip: 28.44
Percent of torn frames: 0.0
Clip 1:
Length of clip: 8282 frames
Average FPS of clip: 30.53
Percent of torn frames: 2.83
Clip 2:
Length of clip: 9934 frames
Average FPS of clip: 29.86
Percent of torn frames: 0.0
Clip 2:
Length of clip: 9943 frames
Average FPS of clip: 29.86
Percent of torn frames: 0.86
Clip 3:
Length of clip: 9196 frames
Average FPS of clip: 26.91
Percent of torn frames: 0.0
Clip 3:
Length of clip: 9196 frames
Average FPS of clip: 29.03
Percent of torn frames: 4.46
Clip 4:
Length of clip: 9219 frames
Average FPS of clip: 28.94
Percent of torn frames: 0.0
Clip 4:
Length of clip: 9219 frames
Average FPS of clip: 29.76
Percent of torn frames: 0.0
Clip 5:
Length of clip: 8643 frames
Average FPS of clip: 29.38
Percent of torn frames: 0.0
Clip 5:
Length of clip: 8643 frames
Average FPS of clip: 30.19
Percent of torn frames: 0.96
Global percent of torn frames: .0.0
Global average FPS: 28.72
Global percent of torn frames: 4.60
Global average FPS: 29.60


 


Conclusion: We understand why a performance analysis of two entirely different engines may be pointless to some, but we’d like think we’re just adding some factual data to two already great games. We also know that both of these games have even more technical achievements that we may have not touched on, but we feel that if you own either console you need to experience these two for yourself. Lastly, while Metal Gear Solid 4 and Splinter Cell: Conviction are both spectacular games, what we really want to know is which one of these two bad asses would win in fight? Use the poll below to vote.



A banner stolen from some site xD

Release Final Fantasy Versus XIII nowwwwwwwwww!!! lol :P

Around the Network

so an two year old game against an game released last year, and Metal Gear Solid 4 would of started before PS3 even launched and would of had to be using early sony prototypes to build there engine on there would of been some problems. 



Of Course That's Just My Opinion, I Could Be Wrong

Splinter Cell is running on an highly modified Unreal Engine 2.5



Snake wins in a with Sam Fischer. He has all the skills Fischer has, but has ridiculous future technology, and is a super soldier. Sam Fischer.....is just some dude lol.

When Fischer takes out a walking nuclear tank more than once, give me a ring



Snake would win obviously.



Around the Network

They just insulted Snake with this comparison.

At least this time around both the games are running sub HD.



That comparison is even worse than the Reach vs. Resistance 2 one....

and yeah Conviction uses Unreal Engine 2.5 just like Bioshock.



Kind of obvious really.  A multi-platform engine (2.5 I think not 3) vs a custom engine build specifically for the platform.  I'd expect MGS4 to run more smoothly.

I wonder what's next for them to compare?  Uncharted 2 vs Gears 2?  Both 3rd person, etc.

I think if they're going to do this they should at least align it to something more specific - such as comparing pros/cons of custom engines vs off the shelf or something like that.

Otherwise they seem like oddities really.



Try to be reasonable... its easier than you think...

Reasonable said:

Kind of obvious really.  A multi-platform engine (2.5 I think not 3) vs a custom engine build specifically for the platform.  I'd expect MGS4 to run more smoothly.

I wonder what's next for them to compare?  Uncharted 2 vs Gears 2?  Both 3rd person, etc.

I think if they're going to do this they should at least align it to something more specific - such as comparing pros/cons of custom engines vs off the shelf or something like that.

Otherwise they seem like oddities really.

Splinter Cell is running a custom engine though.  They took Unreal 2 modified the hell out of it and made Splinter Cell.



darkknightkryta said:
Reasonable said:

Kind of obvious really.  A multi-platform engine (2.5 I think not 3) vs a custom engine build specifically for the platform.  I'd expect MGS4 to run more smoothly.

I wonder what's next for them to compare?  Uncharted 2 vs Gears 2?  Both 3rd person, etc.

I think if they're going to do this they should at least align it to something more specific - such as comparing pros/cons of custom engines vs off the shelf or something like that.

Otherwise they seem like oddities really.

Splinter Cell is running a custom engine though.  They took Unreal 2 modified the hell out of it and made Splinter Cell.

99% of the time a modded Unreal engine delivers worse performance.  Always have and probably always will.  Epic always produce better performing games because they understand every byte of the engine.  Other developers mod on top of the base and invariably they don't do as good a job.

There's some interesting reads on this going right back to early Unreal engine days where Deus Ex devs wondered if the money saving in using an off the shelf engine actually made sense due to the extra work they had to put in modding it and the challenges of keeping unfamiliar code stable.

Personally, as the consumer, I prefer custom engines - they almost always outperform a devloper working on someone elses engine.  But I understand commercial drives make using something like Unreal Engine attractive.



Try to be reasonable... its easier than you think...