Eh, the 3Dmark11 results actually lineup with other sites results.
So unless they're smart enough to fake the rest but the ones with fixed scores I'd say they're fairly legit.
I'd still like to know their test setup though.
So unless they're smart enough to fake the rest but the ones with fixed scores I'd say they're fairly legit.
I'd still like to know their test setup though.
No doubt the 7970 can perform very well in certain titles. What we can agree on is also the unpredictable performance across different games which tends to drag its overall performance index down. Seems like AMD's new architecture does not like Cry Engine or Frostbite 2 too much and having a serious issue with the ego engine in Dirt games. Lets hope these are just down to premature drivers.
Meghan54
Lifer
- Oct 18, 2009
- 11,684
- 5,228
- 136
Wow, so now buying a particular brand of video card means you joined that company's "team?"
I wasn't aware of that fact----that one had to think about joining the "team" when you buy an nvidia or amd video card. I was under the obviously mistaken impression that one could simply buy a video card based on price, performance, power use, etc.
Now I see I have to join a team. Sad.
chimaxi83
Diamond Member
- May 18, 2003
- 5,457
- 63
- 101
Lonbjerg
Diamond Member
- Dec 6, 2009
- 4,419
- 0
- 0
Look at my posting history...3D-******-marks is useless...No matter who wins in that "game" :whiste:
No profanity please.
Moderator jvroig
Last edited by a moderator:
chimaxi83
Diamond Member
- May 18, 2003
- 5,457
- 63
- 101
I tend to agree, synthetics are just a relative gauge of theoretical performance. It's a relevant benchmark nonetheless, it's a fairly easy way to compare cards, and almost every review site out there uses it. This site, [H], and I believe bit-tech don't use it though, I like that because it's geared towards real world gaming (and on this site, GPGPU compute) performance, as it should be.
3dmark scores make it easier for the misinformed to make buying decisions though
Last edited by a moderator:
if you think its useless why the hell you come here ?
please stay away from this thread.
GaiaHunter
Diamond Member
- Jul 13, 2008
- 3,732
- 432
- 126
The problem of 3d mark is twofold.
First it faces all the problems that any "static" benchmark faces - being always the same scene it means AMD and NVIDIA can optimize for it to the extreme. While optimizing is actually good, AMD and NVIDIA don't optimize every game to the same extent they optimize these benchmarks (it would be expensive to do so and much harder due to the fluidity of the games).
Second, being a point benchmark (even though it is based on a measurable metric), it means the score reflect the expectations of what the developers of 3dmark believe is going to be important in future games - they might true but they also might be completely wrong.
I wouldn't put much faith on 3dmark when comparing different architectures, although 3dmark might be decent when comparing cards of the same architecture.
First it faces all the problems that any "static" benchmark faces - being always the same scene it means AMD and NVIDIA can optimize for it to the extreme. While optimizing is actually good, AMD and NVIDIA don't optimize every game to the same extent they optimize these benchmarks (it would be expensive to do so and much harder due to the fluidity of the games).
Second, being a point benchmark (even though it is based on a measurable metric), it means the score reflect the expectations of what the developers of 3dmark believe is going to be important in future games - they might true but they also might be completely wrong.
I wouldn't put much faith on 3dmark when comparing different architectures, although 3dmark might be decent when comparing cards of the same architecture.
So far as Kepler is concerned it is comming out too late. If it came out the same time as the 7970 and slayed it then that would be alright. But take into consideration that for me personally my gfx card lasts up 2 years. If it releases 6 months after the 7970 consider the fact that I would have used up 1/4 of my new gfx cards life.
Lonbjerg
Diamond Member
- Dec 6, 2009
- 4,419
- 0
- 0
Because some people might actually be fooled into thinking that a 3Dmark score is a valid metric..but it's not.
Lonbjerg
Diamond Member
- Dec 6, 2009
- 4,419
- 0
- 0
Really?
Remember XT2900 vs 8800GTX...3Dmark is worthless:
http://hardware.gotfrag.com/portal/story/38332/?spage=5
You live in the past? Additionally, 3DMark06 was quite CPU bound. 3DMark Vantage and 11 are not.
Lonbjerg
Diamond Member
- Dec 6, 2009
- 4,419
- 0
- 0
This much is obvious. 3DMark11 gives reasonable results in line with expectations. Give or take 10% here and there, that doesn't change the overall pecking order.
exar333
Diamond Member
- Feb 7, 2004
- 8,518
- 8
- 91
A non-flagship, single GPU for $400-460 is nothing to sing praises about. I don't miss the days of paying ~$500 for a high mid-range card.
Lonbjerg, you have thoroughly communicated your position on the topic of the relevance of 3DMark11 benchmarks.
10% of the posts in this thread alone are from you and are on that very off-topic side-discussion.
This thread isn't about the relevance of 3DMark11, it is about the results of that benchmark when testing 7950, 580, and 7970 vid cards.
Persist any further in derailing this thread into a "3dmark sucks" thread is going to get you a vacation for threadcrapping and derail.
Knock it off.
Administrator Idontcare
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 + WCL Discussion Threads
- Started by Tigerick
- Replies: 24K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter