Kepler news and discussion

Page 56 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.
I have noticed a bit of a change in recent months regarding HardOCP's coverage. It seems they pretty much are disgusted with the Crossfire driver situation. They rightly have high praise for the Kepler cards and some of the driver level additions, most notably FXAA and Adaptive VSync.

Prior to Kepler, you could guarantee that if a game was AMD favoring, it was a lock to make it into their benchmark suite. For example, Dragon Age 2, which has never run properly on even the best Nvidia cards was quickly adopted for close to a year. When I read the following conclusion in their Dragon Age 2 Performance Review, I was not at all surprised that they then immediately started to include the game in all of their very limited benchmark suites:

Quote:
"Obviously, performance is seriously degraded on NVIDIA video cards, when compared to AMD video cards. A quick Google search showed us that the problem is widespread, with very little outward activity on it from NVIDIA. This performance issue has been a problem since the Dragon Age 2 demo. Before the retail game was released, NVIDIA released a series of beta drivers, citing improved compatibility with Dragon Age 2. In our investigation, we found that "improved compatibility" does not necessarily mean "improved performance." NVIDIA and BioWare have some work to do to get this game running faster on GeForce video cards."
http://www.hardocp.com/article/2011/03/28/dragon_age_2_gameplay_performance_review/9

They only stopped using DA2 after the 7900 series proved to be underperforming in the title. This review of the 7950 was the last time they used DA2:
http://www.hardocp.com/article/2012/01/30/amd_radeon_hd_7950_video_card_review/7

They also are official co-sponsors of the "AMD and HardOCP Game Experience", so they definitely have corporate sponsorship from AMD as well.
317553_289931801034838_215093775185308_1148088_1371512047_n1.jpg


I look at HardOCP's change of heart this way, they clearly have been trying to give AMD a leg up if they can, but they just can't do it anymore with the long list of problems they have written about. They cite complaints about AMD and their drivers in their forums. I really think they are looking out for their readers and doing their best to try to shake up AMD and force them to address these problems. In the end, I think they are honest about their results and I think their test suite has gotten more neutral in recent months, or at least reflective of popular game titles. To me, they are the ultimate AMD fan who just can't give them a pass on the BS anymore.
 
@matto

very true. if someone says hardocp are favoring nvidia maybe that person never did go to their site before or they rarely go to hardocp. in my impression hardocp are much closer to amd than nvidia. getting involved with amd gaming experience was one of the proof they were more red than green. it seems that amd gaming experience has become yearly event for them :).
and if i remember correctly there were once a guy from amd driver team directly responding to amd user problem through hard forum.

maybe they are not perfect but if you want to know the limit of a particular card then [H] might pick your interest since they like to push the cards to their limit with extreme setting.

still remember 6970x3 vs 580x3? 😛
 

I do remember that, and I recall that they benched them and "discovered" that 6970's in tri-Crossfire was better than tri-SLI 580's. Then people in their forums criticized that their CPU wasn't clocked very high and that the 580's were likely being bottlenecked. To their credit, HardOCP rebenched both setups with a higher CPU overclock and...

Quote:
"In our redux today, using our new 4.8GHz CPU, GTX 580 3-Way SLI has come out of its hibernation and is peeking around its fort ready to lay aim on the first prey it sees. That prey happens to be AMD Tri-Fire. In all of our testing today, the GTX 580 3-Way SLI setup performed faster than AMD Radeon HD 6990+6970 Tri-Fire, except for Dragon Age II. ...But let's plainly say it for a third time, our critics were right on some points and we were wrong on some. We are glad we revisited this article."
http://hardocp.com/article/2011/05/03/nvidia_3way_sli_amd_trifire_redux/6
 

It depends on how hardocp OCed the 680(1228).I've heard by pulling 7ghz from the 680's memory,one adversely compromises the oc ability of the cuda cores.People attained 1293mhz stable from the 680 by upping the memory to 6800mhz.
 
lol, well that statement is kinda winded....

First everyone said that is AMD Bull Dozer had more cores and not higher clock speeds making it a lower performance family. Now people are complaining about the new clock mesh saying "Hot clocks don't make a difference"
 
Actually, the clocks are lower.
It actually scales better with clocks.
That being said, they dont have the squeeze out every last watt nVidia brings at stock, tho it wouldnt stop most from ocing anyways.
The turbo is nice, and on top cards, probably not as useful save for benching at stock, but midrange, where ocing isnt as used by most gamers, itll shine more, and is what Im interested in seeing its implementation in those lessor cards
 
Status
Not open for further replies.