Battlefield 4 Video Card Performance and IQ Review @ [H]

Priller said:
Which is a difference of 2-300 rpm and that could make a difference.
Click to expand...

The card was performing at 947MHz during all of the run-throughs, therefore, it did not make a difference to performance.
 
Great review, thank you! While the 290 looks to be an amazing buy, what I really find incredible isn't the average FPS of the 290x, it's the high minimum - 44 FPS, which is fantastic. Can't wait to the see the Crossfire/SLI review; running them at 2560x1600 should enable some awesome eye-candy...
 
Interested to see how xfire 280's look compared to the higher tier cards.

My pair of 7970s should be about evenly matched with some 280's in xfire
 
[L]imey;1040390634 said:
Interested to see how xfire 280's look compared to the higher tier cards.

My pair of 7970s should be about evenly matched with some 280's in xfire
Click to expand...

I believe the common comparison to the 280x is the 7970 GHz edition.
 
David_Schroth said:
The 13.8 driver that we used for testing includes the 47% fan limit for the R9 290 and I manually verified that the limit was set there prior to testing. The change to 9.2 changes the driver of the fan from being a percent based throttle to an RPM based throttle.
Click to expand...

Correct which also changes the RPM at which the fan runs.

Brent_Justice said:
It does not, the fan with 290 was fixed in 13.11 Beta 8 driver.
Click to expand...

I guess you should just look at the link before you think Im wrong.

http://www.pcper.com/reviews/Graphi...rmance-Fan-Speeds-R9-290X-and-R9-290/Catalyst

David_Schroth said:
The card was performing at 947MHz during all of the run-throughs, therefore, it did not make a difference to performance.
Click to expand...

That's good to hear as some dont get this result.

Do you use an open air test bench or in a case with good air flow?
 
[L]imey;1040390634 said:
Interested to see how xfire 280's look compared to the higher tier cards.

My pair of 7970s should be about evenly matched with some 280's in xfire
Click to expand...
Obviously I'm interested in seeing [H]'s data, but I can tell you that 7970 xfire has been pretty strong. All Ultra preset at 1440P w/ 2x MSAA is in the 60-80FPS range (wide range due to maps/mp variability). I found 4X MSAA to be very playable as well but dropped the average 10-15 FPS (45-70 fps or so). This is all based on my experience at stock speeds @ 1440P and an educated guess from when I decided to look up at the framerate counter built into the game (perfoverlay.drawfps 1).

WinSomeLoseNone said:
I believe the common comparison to the 280x is the 7970 GHz edition.
Click to expand...

I think his point was that if [H] tests the 280x xfire, it would be reasonable to expect the same level of performance from 7970 CF, which [H] would be less likely to test now that its essentially EOL from a branding standpoint.
 
Priller said:
That's good to hear as some dont get this result.

Do you use an open air test bench or in a case with good air flow?
Click to expand...

We test all GPUs on an open bench as we have always done. This is not going to change.

That said, I have testing two 290X CrossFire in my personal system now and will be giving a short write-up on it.
 
As for the Ultra/Low bug on Terrain Decoration, David is not able to retest that right now to see if this impacted our testing, but Brent has dug into some of it this morning. Right now we are going to assume that this bug did affect us in our testing, but it looks to not be much of an impact due to the map we tested on. The map we used was a city map with little or no vegetation. Brent did some quick and dirty tests and found framerate differences of 2FPS. That said, all the cards were tested with the exact same bug in the Global settings so our results are still valid for cross-card comparisons.

We have a few BF4 articles laid out going forward, but given the state of BF4 still having very much of a Beta feeling, I think we are going to hold off for the Mantle API before we do any more testing. Hopefully most of the bugs will have been cleared by then and make worthy testing a bit easier to pull off.
 
Activate: AMD said:
Obviously I'm interested in seeing [H]'s data, but I can tell you that 7970 xfire has been pretty strong. All Ultra preset at 1440P w/ 2x MSAA is in the 60-80FPS range (wide range due to maps/mp variability). I found 4X MSAA to be very playable as well but dropped the average 10-15 FPS (45-70 fps or so). This is all based on my experience at stock speeds @ 1440P and an educated guess from when I decided to look up at the framerate counter built into the game (perfoverlay.drawfps 1).



I think his point was that if [H] tests the 280x xfire, it would be reasonable to expect the same level of performance from 7970 CF, which [H] would be less likely to test now that its essentially EOL from a branding standpoint.
Click to expand...

You understood my point perfectly, thanks Activate: AMD.

I'm on a qnix @ 96hz, so my goal with any game is to have the settings as high as possible while keeping minimum framerate near refresh rate.

I'd probably need to drop all AA off the game to achieve my FPS goals, but I've found AA less necessary since moving up to 2560x1440.

Thanks for the insight.
 
I guess I am not the only one who is seeing the inconsistent framerate issues on nVidia card.

The game feels quite stutter in actions. It's got me quite annoyed while running on 2560*1440 resolution.
 
[L]imey;1040391009 said:
You understood my point perfectly, thanks Activate: AMD.

I'm on a qnix @ 96hz, so my goal with any game is to have the settings as high as possible while keeping minimum framerate near refresh rate.

I'd probably need to drop all AA off the game to achieve my FPS goals, but I've found AA less necessary since moving up to 2560x1440.

Thanks for the insight.
Click to expand...
I have a QNix sitting on my front porch right now, so assuming I can overclock it at all I'll be right in the same boat. Without AA I think it'd be doable to get close the the refresh rate, though it might still be tough to maintain 96+ FPS constantly. Hopefully Mantle will give enough of a performance boost to make a difference on that front.
 
Off topic, but out of curiosity what are your 7970's xfire temps with your current set up? idle and bf4 load?
 
I personally consider these results pointless until the Mantle patch. As of now they really hold no meaning or value.
 
bf4-settings.gif


bf4-fps.gif


Also reviewers have found that latency and smoothness are the best on a 780 GTX Ti. I am getting very confused at this point :/. What should we make out of the results? Does Nvidia need to work on the drivers or BF4 is very buggy and might exhibit different characteristics on different maps for different video cards.

bf4-99th.gif
 
madgun said:
bf4-fps.gif
Click to expand...

Here to try and bash on AMD in yet another thread huh.

That is a mighty fine graph you have there, want me to make my own to refute it? Oh, and how about linking to the actual article? of course if you did that then we would know techreport tests using single player which as we all know is a different animal then multiplayer.
 
darkstarcow said:
Here to try and bash on AMD in yet another thread huh.

That is a mighty fine graph you have there, want me to make my own to refute it? Oh, and how about linking to the actual article? of course if you did that then we would know techreport tests using single player which as we all know is a different animal then multiplayer.
Click to expand...

I am not sure where are you getting your delusions from but where am I bashing AMD? I am only asking for a reason for the varied results and have presented actual data.

Moreover, I have already linked to the article on techreport.
Stop being so extra sensitive. It is not as if AMDs paying you or something :rolleyes:
 
Madgun,

Multiplayer and sigle player performance differences are very different. As well, keep in mind that H wasn't able to test the 780ti in this review. They only tested the stock 780 and stock Titan. I'm sure the 780ti would win, especially after NV releases an updated driver.

This reminds me of a similar situation of the 680 vs 7970. The 7970 was a few frames shy of the 680 in single player, but the 7970 was a god awful stuttery mess at the time in multiplayer. Whereas Kepler wasn't. Drivers cause big performance differences, and so does single player vs multiplayer. Techreport tested single player which is quite different IMO.

Besides which, even if AMD wins BF4 that's not going to make me ditch my nvidia cards for a 290. Nah, no thanks. Even though I really love BF4.
 
Thank you for the review. Please note:

Blacklash said:
Some people are claiming the terrain decoration setting is not working properly; selecting low sets an ultra level and an ultra level sets low.

http://battlelog.battlefield.com/bf4/forum/threadview/2955064768159361567/
Click to expand...
These claims are true.

So true, that right now people runing Windows 7 with Terrain Decoration set to Low produce amazing terrain decoration in comparison to people running Windows 8 with TD set to Low:

Windows 7 - TERRAIN DECORATION ON LOW
N6F8Fpo.jpg


Windows 8.1 - TERRAIN DECORATION ON LOW
DD3JdOx.jpg


It has been confirmed by one of the Moderators over at the thread linked above (in quotes) and passed on to the developers.

I just hope they fix it soon, since I'm on Windows 8.1 and pretty much screwed in terms of improving what you see above.

WINDOWS 8.1 - TERRAIN DECORATION ON MEDIUM
http://i.imgur.com/zsRfEnR.jpg

WINDOWS 8.1 - TERRAIN DECORATION ON HIGH
http://i.imgur.com/ChPioA1.jpg

WINDOWS 8.1 - TERRAIN DECORATION ON ULTRA
http://i.imgur.com/wKVdxEb.jpg
 
madgun said:
Also reviewers have found that latency and smoothness are the best on a 780 GTX Ti. I am getting very confused at this point :/. What should we make out of the results? Does Nvidia need to work on the drivers or BF4 is very buggy and might exhibit different characteristics on different maps for different video cards.
Click to expand...

Honestly, it may be a little while before we can really call it. It's still a new game with tons of stuff that needs ironing out - game code and device drivers alike I'm sure.
 
One thing I have noticed on 8.1 with GTX670's is that textures seem to take a minute to load, but I'm rarely taking the time to look closely at them. However, I thought it was the Post-AA that was messing them up!
 
i just verified the terrain decoration thing on my rig. Windows 7 64bit, R9 290X, on the Rogue Transmission map.

Medium setting = Ultra in game
Low setting = High in game
High and Ultra settings = Low in game (both settings looked the same to me)

things like this are probably annoying as shit for you reviewers.
 
Kyle, did you notice more aliasing with AMD cards under same settings? because I did.
 
Porter_ said:
i just verified the terrain decoration thing on my rig. Windows 7 64bit, R9 290X, on the Rogue Transmission map.

Medium setting = Ultra in game
Low setting = High in game
High and Ultra settings = Low in game (both settings looked the same to me)

things like this are probably annoying as shit for you reviewers.
Click to expand...

and for gamers, who are expecting "Ultra" to really be Ultra

these kinds of bugs are unacceptable to me, especially in your release version... isn't that what beta testing was for?
 
AlderaaN said:
Thank you for the review. Please note:


These claims are true.

So true, that right now people runing Windows 7 with Terrain Decoration set to Low produce amazing terrain decoration in comparison to people running Windows 8 with TD set to Low:

Windows 7 - TERRAIN DECORATION ON LOW
http://i.imgur.com/N6F8Fpo.jpg

Windows 8.1 - TERRAIN DECORATION ON LOW
http://i.imgur.com/DD3JdOx.jpg

It has been confirmed by one of the Moderators over at the thread linked above (in quotes) and passed on to the developers.

I just hope they fix it soon, since I'm on Windows 8.1 and pretty much screwed in terms of improving what you see above.

WINDOWS 8.1 - TERRAIN DECORATION ON MEDIUM
http://i.imgur.com/zsRfEnR.jpg

WINDOWS 8.1 - TERRAIN DECORATION ON HIGH
http://i.imgur.com/ChPioA1.jpg

WINDOWS 8.1 - TERRAIN DECORATION ON ULTRA
http://i.imgur.com/wKVdxEb.jpg
Click to expand...

I've not suggested that they are untrue. I've only said that I've not verified them myself. I've yet to purchase BF 4 so I can't test it myself. I'm waiting until things improve with the title before I purchase it. Thank you for the feedback. Hopefully, they will get it together soon.
 
We are suspending further BF4 testing until the Mantle patch in December.

I am not happy with how buggy and unfinished the game is.
 
Brent_Justice said:
and for gamers, who are expecting "Ultra" to really be Ultra

these kinds of bugs are unacceptable to me, especially in your release version... isn't that what beta testing was for?
Click to expand...

well sure, from a gamer's standpoint it's ":rolleyes: that's annoying. but hell i thought the game looked great before, now it looks even better." from a reviewer's standpoint it's ":rolleyes: well that's a lot of time spent on an article using settings i didn't intend to." you guys lucked out by using the Shanghai map, where this really won't affect your results.
 
Brent_Justice said:
We are suspending further BF4 testing until the Mantle patch in December.

I am not happy with how buggy and unfinished the game is.
Click to expand...

I'm sick of AAA titles being released in semi-finished status.

Consumers expect a fully functioning and stable product. Instead we become unwilling beta testers who have to deal with massive performance and stability issues among other smaller issues like these settings.

EA seems to be the worst about this. Remember DA2?
 
PC support is about to hit all time low due to these next-gen consoles.

Thank the pirates for that.
 
Can anyone else with an AMD FX 83*0 comment on their performance? I know my frame rates with my 7950 seem to have a lot of noticeable drops. I have like 4 settings on ultra, rest on high, and no AA and I still see big drops.
 
Brent_Justice said:
and for gamers, who are expecting "Ultra" to really be Ultra

these kinds of bugs are unacceptable to me, especially in your release version... isn't that what beta testing was for?
Click to expand...

It's EA, people will still buy it, they don't care.
 
[L]imey;1040391462 said:
Off topic, but out of curiosity what are your 7970's xfire temps with your current set up? idle and bf4 load?
Click to expand...
I'm not sure if they'd be relevant to you since my GPU's are under water, but mid 40's-low 50's
 
13843169587BDRCzsGHb_5_2_l.gif


I know you give an explanation that there were "more explosions" but you might want to run the test again. I love this site and it doesn't bode well that the information people are hoping to find here is misleading and incongruous.
 
Seventyfive said:
[SNIP]http://www.hardocp.com/images/articles/13843169587BDRCzsGHb_5_2_l.gif[/IMG]

I know you give an explanation that there were "more explosions" but you might want to run the test again. I love this site and it doesn't bode well that the information people are hoping to find here is misleading and incongruous.
Click to expand...

Unfortunately, by the time I realized that the 760 score was higher than the 770, the client side patch from last Friday had already dropped, so going back to rerun the test for better representative data was not going to be very possible due to the wide variety of changes made in the patch.

Quite frankly, this is a very difficult game to benchmark and to get data that is representative of the game play experience. Even when you try to do the same general tasks in a map to try to get a level performance, it is difficult to get the other 63 players to cooperate. Some matches were literally tank shootouts with explosions galore while others there was nary a tank in sight.
 
You must log in or register to reply here.
Back
Top