These are done on Windows 8 with the latest Drivers.
1080p Ultra
1600p Ultra
Note: stock 7970 and 680 are used for these benchmarks, so beware of the clock profile/Max OC differences:
stock 7970 runs at their 925Mhz stock core clock on benchmarks, they can reach a 1200Mhz Overclock on Average, which is a 30% Overclock
stock 680 runs at 1058 Boost Clock+a 50-110 Kepler Boost= 1100Mhz+ max boost clock on Benchmarks, due to Voltage lock, they generally can't reach a 1350Mhz Overclock, even that is only a 22% Overclock
So while these benches show that 7970 are about equal with 680(which on average $70+ more expensive), at max OC 7970 will pull ahead by at least 8%.
I'm so irritated with their conclusions on the last page... They say
"Stepping up to 2560x1600 or a triple-display configuration necessitates a Radeon HD 7970 or GeForce GTX 670, at least."
Which is just laughable, considering that 7970 actually outperforms the GTX 680 according to their chart (0.1fps difference for average and 2 fps higher minimum fps than GTX 680 - small, I know, but if they want to draw such conclusions, they need to analyze graphs properly). They say that a 7970 is necessary while completely ignoring the 7950 range for this review, which is cheaper and will most likely keep up with the GTX 670 just fine.
SOOOO glad I switched from Nvidia after seeing these results! Too bad the game isn't good enough for me to want to purchase. I'm sticking with BF3 ty very much EA
Figures a 7970 owner would say this.So it's going to be 8% ahead of my 1400mhz+ 680 Lightning? Just because their tied in the graph doesn't mean you have to throw out averages to attempt to justify a 'theoretical victory'.
Figures a 7970 owner would say this.So it's going to be 8% ahead of my 1400mhz+ 680 Lightning? Just because their tied in the graph doesn't mean you have to throw out averages to attempt to justify a 'theoretical victory'.
Figures a 7970 owner would say this.So it's going to be 8% ahead of my 1400mhz+ 680 Lightning? Just because their tied in the graph doesn't mean you have to throw out averages to attempt to justify a 'theoretical victory'.
Figures a 7970 owner would say this.So it's going to be 8% ahead of my 1400mhz+ 680 Lightning? Just because their tied in the graph doesn't mean you have to throw out averages to attempt to justify a 'theoretical victory'.
Figures a 7970 owner would say this.So it's going to be 8% ahead of my 1400mhz+ 680 Lightning? Just because their tied in the graph doesn't mean you have to throw out averages to attempt to justify a 'theoretical victory'.
The victory is not theoretical, for a 7970 running at stock clock/0% OC to tie a GTX running with a 9% factory overclock(1100+Mhz from 1006Mhz stock) is already a substantial victory.
The graph shows that 7970 ties with 680 which is both $60+ more expensive and used a 9% factory OC/Kepler Boost, just point out that on average 7970 Overclocks much further than 680 percentage wise and destroys it in Price/Performance, meaning in actual OC use 7970> 680 despite the graph shows they tie in average FPS(and 7970 wins in Min FPS)
If you are going to pull extreme examples like 1400mhz+ Lightnings( only 27% OC from stock 680's 1100Mhz, so it is still 3% behind an average 7970's 30% OC), do keep in mind 1250Mhz(35%) 7970s would be 8% ahead of that, and 1250Mhz on 7970 isn't as rare as a 1400Mhz 680(only a few Lightning or Classy with EVbolt can reach that).
Figures a 7970 owner would say this.So it's going to be 8% ahead of my 1400mhz+ 680 Lightning? Just because their tied in the graph doesn't mean you have to throw out averages to attempt to justify a 'theoretical victory'.
No need to get all bent out of shape. Your 680's are indeed amazing but its not like you can even get 680 Lightnings like yours anymore. The 7970's are cheaper and mostly faster these days which makes them better cards overall. Now if MSI would kindly start putting out 680 Lightnings like the first batch I'd love to get two!
Yes, which is why I was so frustrated with this review / comparison. It's very difficult to extract useful comparison information, because a lot of these reviews are either not done at realistic clock speeds, or with current drivers, or comparisons are just odd.
Figures a 7970 owner would say this.So it's going to be 8% ahead of my 1400mhz+ 680 Lightning? Just because their tied in the graph doesn't mean you have to throw out averages to attempt to justify a 'theoretical victory'.
The victory is not theoretical, for a 7970 running at stock clock/0% OC to tie a GTX running with a 9% factory overclock(1100+Mhz from 1006Mhz stock) is already a substantial victory.
The graph shows that 7970 ties with 680 which is both $60+ more expensive and used a 9% factory OC/Kepler Boost, just point out that on average 7970 Overclocks much further than 680 percentage wise and destroys it in Price/Performance, meaning in actual OC use 7970> 680 despite the graph shows they tie in average FPS(and 7970 wins in Min FPS)
If you are going to pull extreme examples like 1400mhz+ Lightnings( only 27% OC from stock 680's 1100Mhz, so it is still 3% behind an average 7970's 30% OC), do keep in mind 1250Mhz(35%) 7970s would be 8% ahead of that, and 1250Mhz on 7970 isn't as rare as a 1400Mhz 680(only a few Lightning or Classy with EVbolt can reach that).
That's what I'm saying though. It's an extreme example, sure. But what I mean is trying to use averages to claim a 'victory' from these charts is kind of unnecessary. I mean, both are clearly capable of handling the game, so why try to make a winner out of it. And like I said, some examples, maybe considered "extreme", but they're going to give one card or the other the edge depending on which two cards of the thousands of each that're out there are used, but by such a small margin anyways. Plus, price/performance is also subjective because you can measure frames-per-second, but how do you measure the more subjective value of things to someone, like PhysX or driver/Control Panel preference?
Sorry about the rambling. Just trying to say basically that we should just take the graph for what it is: showing that both cards' owners should be content. And not make it about a competition, especially over such an incredibly small margin of difference between the two.
The victory is not theoretical, for a 7970 running at stock clock/0% OC to tie a GTX running with a 9% factory overclock(1100+Mhz from 1006Mhz stock) is already a substantial victory.
The graph shows that 7970 ties with 680 which is both $60+ more expensive and used a 9% factory OC/Kepler Boost, just point out that on average 7970 Overclocks much further than 680 percentage wise and destroys it in Price/Performance, meaning in actual OC use 7970> 680 despite the graph shows they tie in average FPS(and 7970 wins in Min FPS)
If you are going to pull extreme examples like 1400mhz+ Lightnings( only 27% OC from stock 680's 1100Mhz, so it is still 3% behind an average 7970's 30% OC), do keep in mind 1250Mhz(35%) 7970s would be 8% ahead of that, and 1250Mhz on 7970 isn't as rare as a 1400Mhz 680(only a few Lightning or Classy with EVbolt can reach that).
This is worth repeating. A stock 7970 which lost to 680 at release is now on par with, if not beating, a 680 which is using a much higher clock rate. A 7970 at 925Mhz competing with a 680 which is around 1100MHz (or so) is a substantial performance boost fo the 7970.
Crappy console port? Really? Looks like someone is not much of a gamer. I do not where you get that from, but that is irrelevant to the quality of the game which is pretty damn high.
Crappy console port? Really? Looks like someone is not much of a gamer. I do not where you get that from, but that is irrelevant to the quality of the game which is pretty damn high.
From the previews I saw the game looked awesome. Then I heard a bunch of people talking about how terrible multiplayer is and how only the first level or two in single-player looks good and then it gets horrendous. So...not sure what to think. I may just wait for a few patches and see what people are saying at that point.
Crappy console port? Really? Looks like someone is not much of a gamer. I do not where you get that from, but that is irrelevant to the quality of the game which is pretty damn high.
From the previews I saw the game looked awesome. Then I heard a bunch of people talking about how terrible multiplayer is and how only the first level or two in single-player looks good and then it gets horrendous. So...not sure what to think. I may just wait for a few patches and see what people are saying at that point.
Go watch a playthrough, if you want proof whether or not they were truthful. What used to be nextgenwalkthroughs and darksydephil (DSPGaming) has one up.
Crappy console port? Really? Looks like someone is not much of a gamer. I do not where you get that from, but that is irrelevant to the quality of the game which is pretty damn high.
From the previews I saw the game looked awesome. Then I heard a bunch of people talking about how terrible multiplayer is and how only the first level or two in single-player looks good and then it gets horrendous. So...not sure what to think. I may just wait for a few patches and see what people are saying at that point.
Go watch a playthrough, if you want proof whether or not they were truthful. What used to be nextgenwalkthroughs and darksydephil (DSPGaming) has one up.
Im not even sure why this thread started. Posted one slide and then started talking about 680/7970. No real reason for this to be news.....
Quote:
Originally Posted by dph314
From the previews I saw the game looked awesome. Then I heard a bunch of people talking about how terrible multiplayer is and how only the first level or two in single-player looks good and then it gets horrendous. So...not sure what to think. I may just wait for a few patches and see what people are saying at that point.
The MP is actually really good if you just focus on the guns and skill level required to play it. I think a lot people are put off by it because it is quite a bit harder than some FPS mp games. Different play style. You have to play much more tactical and can just run and gun. You also have to use the buddy system and it makes it really fun. Everything but the core gun gameplay does suck however. The menu system is awful and it certianly needs a lot of work.
Figures a 7970 owner would say this.So it's going to be 8% ahead of my 1400mhz+ 680 Lightning? Just because their tied in the graph doesn't mean you have to throw out averages to attempt to justify a 'theoretical victory'.
Your golden 680 is one of very few that will OC that way, and is in no way an accurate representation of an average OC from a 680. You using your 1400MHz+ Lightning would be like throwing out a 1375MHz 7970 and saying "Your 680s don't hold a candle to my 7970", it's completely ridiculous to use "golden" cards as a reference point in comparisons. Average OC to average OC is the way to make an accurate comparison, not cherry picking.
Average OC of a 7970 is 1200/1700, compare that to an average OC of a 680, and they're competitive. Neither one is "miles ahead" of the other card.
2012 = years ago? Okay... I guess the GTX 670 is cannot do 30FPS in Metro 2033.
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Related Threads
?
?
?
?
?
Ask a question
Ask a question
Overclock.net
27.8M posts
541.2K members
Since 2004
A forum community dedicated to overclocking enthusiasts and testing the limits of computing. Come join the discussion about computing, builds, collections, displays, models, styles, scales, specifications, reviews, accessories, classifieds, and more!