BrentJ
Member
What are you playing BF4 for, the single player missions, or the multiplayer. I'd argue, most of the BF4 purchases are to play the multiplayer aspect of the game, not single player.
We test what is relevant to gamers.
What are you playing BF4 for, the single player missions, or the multiplayer. I'd argue, most of the BF4 purchases are to play the multiplayer aspect of the game, not single player.
We test what is relevant to gamers.
BrentJ
Member
We evaluated the retail version of the game, in gold release form. This is the same version every other gamer out there will be playing of the game. It is 100% real-world and relevant to gamers.
BrentJ
Member
Because that's what gamers are playing.
It'd be like buying a ferrari and not testing it on the race track because that can be inconsistent, and instead testing the windshield wiper speed, of which no one cares about.
RussianSensation
Elite Member
- Sep 5, 2003
- 19,458
- 765
- 126
Other professional sites that tested BF4 also show that AMD has an advantage in this title. As Groover says, this is not unexpected if you have been following BF4's benchmarks.
TechSpot shows the same thing where a 280X/7970GE trades blows with a 780.
I would expect NV to make up 10-15% with future drivers but AMD will offset that with a 10-20% increase due to Mantle. For BF4, 2x 290s are going to be unbeatable against a 780Ti.
TechSpot shows the same thing where a 280X/7970GE trades blows with a 780.
I would expect NV to make up 10-15% with future drivers but AMD will offset that with a 10-20% increase due to Mantle. For BF4, 2x 290s are going to be unbeatable against a 780Ti.
Saddly you can't produce constant data in the matrix so your tests have no real world value other than during x test with y parameters v product produced z results.
It's not comparable data, which is your purpose or job so to speak.
BF4 SP and MP engine is exactly the same, there is no reason to add the MP variable when testing GPU performance because it is completely unscientific and at it's core an uncontrolled test.
I would really like to see what kinda FPS could be achieved with 270x at 1920x1200 with AA disabled, and everything else on ultra.
RussianSensation
Elite Member
- Sep 5, 2003
- 19,458
- 765
- 126
Disgree. The engine is the same but the workload is different. MP is far more demanding on both the GPU and CPU. While you may not have 95 percentile comparable results with other sites when testing MP, if you test MP map 3-5 times within the course of 3 hours and record FPS in that one continuous gaming session, your averages should be fairly close during that 1 period of testing.
Also you are missing one important point - if AMD GPUs are showing a more consistent frame rate with less variability compared to most NV GPUs in the same price brackets, that means we have a pattern. This is also corroborated by other sites btw.
This was also true for GTX670/GTX680 vs. HD7950/7970 in BF3 until Catalyst 12.11s but I don't ever recall you making claims that BF3's multiplayer should not be tested. Hmm...
A video speaks a thousand words sometimes - You can fire up multiplayer and clearly see which GPU is faster in the same area of the game. Just swap out the GPUs in the same system and restart the game!
http://www.youtube.com/watch?v=qR3ewLMbywY
Now repeat this in 1 hour, 2 hours and then 3 hours and you should be forming a strong statistical average based on recorded data. WOW, magic.
Last edited:
http://www.youtube.com/watch?v=Wh0BZZU6c1k
FYI the Terrain Detail Decoration setting is bugged. Medium = Ultra, Ultra = Low, Low = High, High = Medium
FYI the Terrain Detail Decoration setting is bugged. Medium = Ultra, Ultra = Low, Low = High, High = Medium
chimaxi83
Diamond Member
- May 18, 2003
- 5,457
- 63
- 101
Laughable. We all know who's OK with [H] testing methodology when Nvidia has the performance lead. When Nvidia "feels" better to them, they're quoted all over this forum. Now? AMD bias!
What happened Balla? Are you ok? You changed your stance quickly, something happened?
http://forums.anandtech.com/showpost.php?p=35676195&postcount=43
According to hw.fr the 1GHz R290X is 9% faster at 1440 than the 980MHz GTX 780. R290X just needs to go higher, there is no other way to put it. 1250 on water for a setup that will cost much more than non reference 500-570 dollar GTX 780s that can hit 1300-1400MHz all day isn't going to cut it :|
Because BF4 Single Player is what everyone cares about, right?
Ouch.
Better get that Mantle going because they clearly aren't doing something right, like supporting DX11 features?
Interesting! Maybe I can gain a few FPS by setting this to Ultra (Low), I have it on Low (High) right now...
So AMD has the lead in BF4. This isn't surprising since they had development exclusivity on it - i'm sure things can and will change once newer sets of performance game ready drivers are released. Or maybe not. It's like AMD having the lead, in say, Sleeping Dogs. Who gives a crap.
Guess what. Some games favor different architectures. Batman: AO runs better on the 770 than the 280X. Sleeping Dogs runs better on the 280X than the 770. Blizzard games favor nvidia. Square Enix games favor AMD. DIFFERENT GAMES BEHAVE DIFFERENTLY. This must be hard for some to swallow. BF4 runs better on AMD. Who the heck cares?
So instead of arguing, just wait for performance drivers and realize that different games can and will favor different GPU architectures. This may change with newer drivers, or it may not change. Who knows. Why argue about it.
Guess what. Some games favor different architectures. Batman: AO runs better on the 770 than the 280X. Sleeping Dogs runs better on the 280X than the 770. Blizzard games favor nvidia. Square Enix games favor AMD. DIFFERENT GAMES BEHAVE DIFFERENTLY. This must be hard for some to swallow. BF4 runs better on AMD. Who the heck cares?
So instead of arguing, just wait for performance drivers and realize that different games can and will favor different GPU architectures. This may change with newer drivers, or it may not change. Who knows. Why argue about it.
Last edited:
Teizo
Golden Member
- Oct 28, 2010
- 1,271
- 31
- 91
I know why you did it, but I would really prefer review sites to start holding these game developers to higher standards so they stop releasing games still in Beta.
That is, don't benchmark their games until they are actually done patching them so as to not give them the advertisement. The performance numbers are not going to be accurate.
I can appreciate no game will likely be perfect upon launch, and that patches are good, but the way this game was released is like a bad habit on steroids.
Yeah, not sure why people are arguing about this. BF3 ran better on Nvidia though, so maybe AMD wanted to make sure that didn't happen again
Last edited:
People buy battlefield games for multiplayer. That's the typical user experience they need to test.
I pity anyone who bought BF4 for the single player. It was an atrocity and dice should apologize by release a patch that removes it from our hard drives.
I was surprised when I turned on the FPS counter with my 7970 mix of low and ultra settings (mainly to improve vision in game) and no AA and I was getting 140-160fps depending on the map it would drop down to around 100fps.
blastingcap
Diamond Member
- Sep 16, 2010
- 6,654
- 5
- 76
Wait, how can both charts be true? Is 290X faster or slower than TITAN? Are they using same drivers or builds ...?
Lol.
Personally I'm not surprised at the performance given AMD must have worked with DICE closely for this game.
chimaxi83
Diamond Member
- May 18, 2003
- 5,457
- 63
- 101
A clear and direct example of the flip flops these guys do to defend their graphics cards. It's strange. Be more consistent, dudes.
Nothing happened, it looked as if AMD had a driver issue which seemed to be sorted.
MP performance is what matters, comparing graphics cards in MP is apples to oranges however.
When [H] runs a graphics card test they don't load up TR and just pick random areas and do random things. They do the same area, taking the same path, attacking them same things.
Ask them why they do that, and why it's impossible to reproduce it in a MP game.
The difference between the engine in MP vs SP is CPU related, we're comparing graphics cards here.
4.6GHz Ivy with 1600 CL9 RAM could very well be holding back the cards because of the dynamic load being placed on the overall system. Too many non GPU related variables occur in MP, it makes the test pointless.
If you still have questions about why, look at the 760 beating the 770 in apples to apples. Pointless review is pointless.
Last edited:
Gikaseixas
Platinum Member
- Jul 1, 2004
- 2,836
- 218
- 106
AMD > Nvidia happened
In the end i expect Nvidia to close the gap real soon. It's a big title and they must be working non-stop on it.
Also, isn't the text rig the same? Same CPU, Mobo, Memory, etc? That's consistent enough i guess
Last edited:
Not at all, but I know you and several others like to view things in such a manner.
What happened was a review came out that showed AMD might have had a potential performance crippling driver issue in MP that wasn't present in SP results.
MP is what matters, however only is so far as it should be the same as single player on the GPU level.
What that graph represented was that it wasn't on par with the SP results, which indicated a major problem in the driver in MP which is where you want the performance.
This discussion is about benching in a variable environment resulting in pointless results.
There is no question the 290x in uber is going to beat the Titan in BF4, be it in SP or MP given identical runs. However as shown in the same H review a slower card, such as the 760 due to uncontrolled variables can be shown to be faster than a faster card in this case the 770.
H could have just as easily shown Titan being faster if the same anomaly happened and I would be saying the same thing I am now.
5150Joker
Diamond Member
Hard|OCP is only biased if your favorite vendor loses. It's no surprise to me at all that AMD wins with BF4. DICE/EA are big backers of Mantle + AMD so of course it should win unless they screw up really bad. Unfortunately with EA running the show, Frostbuggy 3.0 made a poor showing in BF4 with its constant crashes, piss poor netcode and graphical glitches.
blastingcap
Diamond Member
- Sep 16, 2010
- 6,654
- 5
- 76
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads
- Started by Tigerick
- Replies: 21K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter