AMD Bulldozer FX-8150 vs Intel Core i5 2500K 8-Core Gaming Benchmarks Linus Tech Tips

Linus Tech Tips ·Linus Tech Tips ·2012-05-07 · 975 words · ~4 min read
Floatplane YouTube

Transcript

JSON SRT VTT 72
0:08 guys this is the MSI 990 fxa
0:12 gd80 and it is an am3 plus motherboard featuring full support for SLI Crossfire
0:17 X and AMD bulldozer 8 core CPUs this is
0:20 an fx 8150 installed on a Crosshair 5 formula now the Crosshair 5 formula is
0:25 the board that was used by 99% of the reviewers out there for their bulldozer
0:30 launch coverage so what I'm aiming to do here is to find out whether the
0:35 Crosshair 5 formula will perform any different from the MSI equivalent
0:40 high-end gaming skew with a single
0:44 6970 8 gigs of DDR3 memory and an 8150
0:50 FX 8 core CPU from AMD so I have done my
0:55 first set of run throughs now I can't
0:58 get the Witcher 2 running properly right now it's something to do with the game
1:01 updating and my save files won't load and so I'm going to have to examine that
1:06 but for now I want to bring you the results that I can so what I'm going to
1:09 be doing is I'm going to be running the Crosshair 5 the 990 fxa gd80 and then
1:14 for completeness sake I'm going to throw a 2500k into the
1:20 mix and uh that will be on a z68 board right
1:26 there so that's the z68 AG gd80 just to see how the 8 150 Compares against its
1:32 closest competitor price-wise on the Intel side so I've done my run through
1:36 with the 990 fxa gd80 board now can see
1:39 what it's called down there just so you guys know that I'm telling you what it
1:43 is and I've only encountered one very curious result so on the top are my
1:48 numbers with the 990 fxa and on the bottom are my numbers with the Crosshair
1:51 F so you can see these gaming numbers are very close Metro 2033 is within uh
1:56 margin of error Battlefield Bad Company 2 within margin of error crisis 2 within
2:01 margin of error and lost yeah Lost
2:04 Planet 2 is identical and finally Dirt 3
2:08 is well within margin of error but cinebench for some reason performed
2:12 dramatically better on the gd80 so before I put my Sandy bridge where is it
2:16 2500k on the bench I'm going to go and I'm going to throw the Crosshair 5 back
2:20 on there just to give it another shot at that cinebench Benchmark let's see if
2:24 maybe it was just an anomaly looks like it was purely an anomaly guys 5.84 is is
2:30 well within margin of error of 5.82 so
2:34 I'm going to go ahead and say that these boards perform pretty much identically
2:40 with the 8150 at stock speeds now I'm
2:44 going to go ahead and get on with my Sandy Bridge testing well my Sandy
2:48 Bridge numbers are complete so this is with the same test platform except for a
2:51 different CPU and different motherboard obviously and I think the overall
2:56 conclusion is that a whole lot is being made about the bulldozer performance or
3:02 lack thereof against its Sandy Bridge counterparts but looking at the test
3:05 settings that are being used in a lot of places I don't really feel they're all that realistic so for all of the games
3:10 that I'm running I'm running at 1080p which is the full resolution of this
3:14 monitor which most gamers are transitioning towards if they're not
3:17 already using it and then I'm also running at high details with
3:22 anti-aliasing and an isotropic filtering on in most cases and you can see here
3:27 that in most of the games most of the applications I'm running my average and
3:31 my minimum frames per second are very very playable at those settings
3:37 so yeah this is one of the things where
3:40 I'm kind of looking at it going well okay yes the 2500k does perform better
3:46 you can see there are games where it really pulls away such as crisis 2 where
3:50 we're looking at outside the margin of error in terms of the performance
3:54 difference somewhere in the 8% range Lost Planet 2 it does perform a little
3:58 bit better remember that one's a Ken bench mark so that should be fairly
4:01 representative Dirt 3 it does pull ahead this is a CPU limited game and it's
4:06 about 5% better um but it doesn't perform as well in something heavily
4:10 threaded like cinam Ben for example I mean yes it's not that far behind
4:14 considering the fact that the 2500k is a cheaper processor than the 8150 but you
4:19 also have to bear in mind that the platform is more expensive if you want to build a gaming rig like let's say
4:24 dual 6970 gaming rig with uh high-end
4:28 gaming CPU but RAM etc etc you're looking at on the low end a
4:34 970 chipset for AMD or a 990x which are
4:38 both still Crossfire capable whereas with Intel if you want a crossfire or
4:42 SLI ready board you have to use a p67 or
4:45 a z68 boards which are significantly more expensive so you do have to factor
4:48 that in so thank you for checking out my little video where we did establish that
4:52 there's really no difference in performance between one 990 FX board and
4:58 another with the bulldozer 8150 FX chip
5:01 and then that the performance Delta between the 2500k and the 8150 is while
5:08 significant yes it is significant in some cases uh may not be a deal breaker
5:13 especially if you're running at sort of real world resolutions in Real World
5:17 Games don't forget to subscribe to lus check tips for unboxings reviews and
5:21 other computer videos