- Status
very good review got a lot of info out of it. The one thing that strikes me about these new cards is that if these being just mid range cards what the heck are the new high end flag ship ones going to be like. If these can almost keep up to a HD5870 then the new flag ship cards should be totally out of this world for performance & those are the ones that I personally will be looking at.
I also found it totally BS that Nvidia because they had no answer to AMD's new cards told a company like EVGA to produce & market that overclocked 460 as a mainstream product which we all know it is not. last I check there were only a handful released & almost all online outlets ran out of stock before the reviews even were released on the internet. If they wanted to rain on AMD's parade maybe they should have released a true new card called it a geforce 461 GTX & then not made themselves look so desperate by pulling this PR stunt & lets face it that is all Nvidia did was pull a big PR stunt. I am by no means a AMD/ATI fanboy I own both companies cards at the moment & like them both just fine. I just find it sad that Nvidia had to resort to these tactics to try to make AMD/ATI look bad. Now when someone asks me what I have inside my towers I might say something like oh I got dual 5970's in one tower & my other I will say not important its going to be replaced very soon. lol
I also found it totally BS that Nvidia because they had no answer to AMD's new cards told a company like EVGA to produce & market that overclocked 460 as a mainstream product which we all know it is not. last I check there were only a handful released & almost all online outlets ran out of stock before the reviews even were released on the internet. If they wanted to rain on AMD's parade maybe they should have released a true new card called it a geforce 461 GTX & then not made themselves look so desperate by pulling this PR stunt & lets face it that is all Nvidia did was pull a big PR stunt. I am by no means a AMD/ATI fanboy I own both companies cards at the moment & like them both just fine. I just find it sad that Nvidia had to resort to these tactics to try to make AMD/ATI look bad. Now when someone asks me what I have inside my towers I might say something like oh I got dual 5970's in one tower & my other I will say not important its going to be replaced very soon. lol
SininStyle
Distinguished
- Dec 15, 2009
- 51
- 0
- 18,630
I think the biggest problem Fanboys suffer is complete loss in logic for performance for the dollar. This problem only costs them money and FPS and doesnt seem to effect the people with more then 2 brain cells to rub together.
I cant seem to grasp how anyone can consider the 460 obsolete when its beats the 6850 when both cards are overclocked and just falls behind the 6870 when both are OCed just a few FPS. The 6870 is $240 before shipping the 460 is $190 and usually a rebate available.
Do the math. $50+ more for 3fps on a maxed out card?
Eh to each their own. its pretty obvious the fanboy are out in groves over these 68xx cards and i fail to see them being better then a 460. COMPARABLE sure. this is not a slam dunk or a owning. Anyone who says otherwise is a fanboy seriously. The 6870 is a waste of money buy a 6850 and OC it for close to the same results. now you got competition on the 460 1gb thats just about equal. Flip a coin, buy your favorite, either case you get great performance for cheap money do such a price war. The price war exists because they are so close.
I cant seem to grasp how anyone can consider the 460 obsolete when its beats the 6850 when both cards are overclocked and just falls behind the 6870 when both are OCed just a few FPS. The 6870 is $240 before shipping the 460 is $190 and usually a rebate available.
Do the math. $50+ more for 3fps on a maxed out card?
Eh to each their own. its pretty obvious the fanboy are out in groves over these 68xx cards and i fail to see them being better then a 460. COMPARABLE sure. this is not a slam dunk or a owning. Anyone who says otherwise is a fanboy seriously. The 6870 is a waste of money buy a 6850 and OC it for close to the same results. now you got competition on the 460 1gb thats just about equal. Flip a coin, buy your favorite, either case you get great performance for cheap money do such a price war. The price war exists because they are so close.
- Mar 7, 2003
- 12,012
- 0
- 40,790
I guess when faced with conflicting info, the best way to decide which is correct is to simply believe the site that tells you what you want to hear.
As for temps, we're one of the only sites that actually measures ambient temperature every time we test to make sure we're reporting the true delta, not the absolute temps that do not take the ambient into account.
As for our 6850 being slower, you did hear that some reviewers got 1120 sp-enabled 6850's and didn't bother to check... right?
But yeah, it's probably easier for your ego to assume we're the ones who dropped the ball.
- Mar 7, 2003
- 12,012
- 0
- 40,790
We tested all cards at reference clocks, with the exception of adding 32 MHz to the 460 as described in great detail. It's hard to find 460s at reference clocks on the market so we averaged out the clocks on all the cards we could find being sold at the lowest price.
To us, this is fairer than blindly following reference clocks when, in reality, those aren't the clocks most people will get if they buy that card. Frankly, the 32 or so mhz increase wouldn't be detected in benchmark results but we did it because given the circumstance it seems the fair thing to do.
We also overclocked the 6850 and 6870 to see how far they can go, and we reported those results, too.
You should really read the article before criticizing it so heavily. Otherwise it looks like you're just out to get us because we aren't saying what you wish we were.
IzzyCraft
Distinguished
- Nov 20, 2008
- 1,438
- 0
- 19,290
[citation][nom]cleeve[/nom]We tested all cards at reference clocks, with the exception of adding 32 MHz to the 460 as described in great detail. It's hard to find 460s at reference clocks on the market so we averaged out the clocks on all the cards we could find being sold at the lowest price.To us, this is fairer than blindly following reference clocks when, in reality, those aren't the clocks most people will get if they buy that card. Frankly, the 32 or so mhz increase wouldn't be detected in benchmark results but we did it because given the circumstance it seems the fair thing to do.We also overclocked the 6850 and 6870 to see how far they can go, and we reported those results, too.You should really read the article before criticizing it so heavily. Otherwise it looks like you're just out to get us because we aren't saying what you wish we were.[/citation]
Not arbitrary at all. People keep complaining about the 460 and overclocking, but nvidia has been very lenient to their patenters allowing them to do a ton of things with them. I find it just as annoying as people always recommending one card saying that drivers aren't mature for them.
Not arbitrary at all. People keep complaining about the 460 and overclocking, but nvidia has been very lenient to their patenters allowing them to do a ton of things with them. I find it just as annoying as people always recommending one card saying that drivers aren't mature for them.
For me, Barts sounds like an excuse to vendors - ditching bullets from failed delivery of 32nm TSMC's process.
Very disappointed.
Very disappointed.
"wherefore art thou" actually means 'why are you' not 'where are you'. Not being a correctional snob, just that I made this mistake a few times myself and found out I was wrong, much to my suprise, so just passing on the info.
Crysis *SHOULD* be tested whether you are tired of it or not, it's one of very few PC centric titles that still looks absolutely amazing and isn't a 'shoddy port' who's performance woes are due to bad coding, also many PC players may have passed over it due to slower PCs back then and now it doesn't get a mention even though it's easily one of the most impressive technical showcases on PC still (And contary to 'fashionable belief' it actually plays a good game too thanks to it's 'create your own epic' freedom and super physics).
Crysis' engine is stunning and has so much to offer (playing around in the editor will show how well programmed it is feature wise and power wise), all new GPUS should use it as the pinnacle of benchmarking on PC regardless of if there are slower/heavier titles (usually ports like GTAIV, use that for CPU benches) out there that can show frame rate differences more readily (perhaps). We want to see how a beautifully programmed engine can scale, not how a poorly rushed port can be 'polished' after the fact by throwing overly powerful (for the game) hardware at it!
Above all, the people who follow GPU announcements are the upper niche of PC gamers and we appreciate Crysis more than any butchered port from consoles, know your demographic, please your demographic. I'm still not tired of Crysis and I can't wait to upgrade my system many times to play through it again and again in higher and higher res at faster and faster frames as the years go by. And I have a feeling Crysis two will in no way be a better benchmark as it's already known to be far more linear and with many 'clever workarounds' to keep performance up, in fact the hardware is probably less important for Crysis 2 in general given all the evidence (and the cross porting concessions) we know about so far.
Crysis *SHOULD* be tested whether you are tired of it or not, it's one of very few PC centric titles that still looks absolutely amazing and isn't a 'shoddy port' who's performance woes are due to bad coding, also many PC players may have passed over it due to slower PCs back then and now it doesn't get a mention even though it's easily one of the most impressive technical showcases on PC still (And contary to 'fashionable belief' it actually plays a good game too thanks to it's 'create your own epic' freedom and super physics).
Crysis' engine is stunning and has so much to offer (playing around in the editor will show how well programmed it is feature wise and power wise), all new GPUS should use it as the pinnacle of benchmarking on PC regardless of if there are slower/heavier titles (usually ports like GTAIV, use that for CPU benches) out there that can show frame rate differences more readily (perhaps). We want to see how a beautifully programmed engine can scale, not how a poorly rushed port can be 'polished' after the fact by throwing overly powerful (for the game) hardware at it!
Above all, the people who follow GPU announcements are the upper niche of PC gamers and we appreciate Crysis more than any butchered port from consoles, know your demographic, please your demographic. I'm still not tired of Crysis and I can't wait to upgrade my system many times to play through it again and again in higher and higher res at faster and faster frames as the years go by. And I have a feeling Crysis two will in no way be a better benchmark as it's already known to be far more linear and with many 'clever workarounds' to keep performance up, in fact the hardware is probably less important for Crysis 2 in general given all the evidence (and the cross porting concessions) we know about so far.
It's a shame you didn't put overclocked gtx 470 benchmarks in to show just how badly it deficates all over the 6870 when you compare the both of them overclocked. Especially since the 470 can be had for a mere 10 dollars more.
TheGreatGrapeApe
Champion
- Feb 18, 2003
- 20,043
- 0
- 50,960
Yeah I probably did waste my time in complaining about it, but I'm not going to go down that road explaining that it's not my job here, and I won't bother with the attempt to pull me into 'justifying' me being 'right or wrong' for my criticism of the focus on rather than the numbering itself, but I don't see how that changes the reality that more space was spent focusing on the naming than on any previous (and far worse) naming issues.
My justification question would be about not having a reference baseline for the HQV results especially since a segment was dedicated to it, would it have required more time than the 'naming section' to cut and past the AMD slide you obviously referenced but did not provide that others did?
How does a subjective test provide insight if you don't know how other products rate, are other perfect, terrible, what? Without something to compared it to, neither the 204 nor the 198 #s mean much on their own.
SininStyle
Distinguished
- Dec 15, 2009
- 51
- 0
- 18,630
Its amazing how quickly the tide has turned. Use to be endless Nvidia fanboyism every where now its AMD. How can you be so blind to what AMD is doing? The 6870 is pretty much maxed out on its clock where the 6850 and 460 both have SO much headroom to OC. Yet this point is lost on the 460 for some reason. Using benchmarks with the $240 6870 maxed out compared to the $180 460 without an OC is stupid and dont understand why it continues to be done.
The 6850 and 460 OC so well they can compete with the 6870 which has no room to improve. Why would anyone spend $240 on it when the others can do the same thing for $50 less. stop using tunnel vision and use some logic. Your whining about a factory overclocked card being compared to another one...the 6870. AMD in short maxed the card out to compete so Nvidia has answered by doing the same.
stupid marketing crap on both sides use your brains. They think the sheep are so dumb they will only see the numbers on a bench and not figure it out.
The 6850 and 460 OC so well they can compete with the 6870 which has no room to improve. Why would anyone spend $240 on it when the others can do the same thing for $50 less. stop using tunnel vision and use some logic. Your whining about a factory overclocked card being compared to another one...the 6870. AMD in short maxed the card out to compete so Nvidia has answered by doing the same.
stupid marketing crap on both sides use your brains. They think the sheep are so dumb they will only see the numbers on a bench and not figure it out.
hardcore_gamer
Distinguished
- Mar 27, 2010
- 540
- 0
- 18,980
This is rediculous. Just because it is a new chip with new technologies doesn't mean that it should be labelled with a flagship naming convention when it really isn't the flagship card. They don't even have the nVidia 480 in the test linup.
For the last three product releases (since the 3870) the x870 has been the flaship single GPU card. They should have called these cards 6350 and 6370 and reserved the 6870 for the real top of the line card that hands down crushes the competition.
I am not a gamer. I don't have any stake in either company. I like to watch the competition. But this move by AMD is seriously flawed. It can only create confusion, disappointment and disillusion.
For the last three product releases (since the 3870) the x870 has been the flaship single GPU card. They should have called these cards 6350 and 6370 and reserved the 6870 for the real top of the line card that hands down crushes the competition.
I am not a gamer. I don't have any stake in either company. I like to watch the competition. But this move by AMD is seriously flawed. It can only create confusion, disappointment and disillusion.
- Mar 7, 2003
- 12,012
- 0
- 40,790
[citation][nom]TheGreatGrapeApe[/nom]Without something to compared it to, neither the 204 nor the 198 #s mean much on their own.[/citation]
I think it's pretty clear that if the maximum is 210 points, a ~200 point score would suggest the 6800 series is pretty close to perfect. That in itself is useful information, and it definitely means something.
Would I liked to have added more results, say from a 470? Of course I would, but like the article states I didn't have time. I don't think you're getting this part. As it was, the article was posted after the NDA lifted. I'm not sure why you're having trouble assimilating that.
I do plan to follow up in the near future with an exhaustive HQV 2.0 test of a bunch of new cards. But as for this review, I didn't have time. The test I did manage to squeeze in was probably partially the reason why we published a little late, and I certainly wasn't going to add another couple hours on top of that with another fresh install so I could test the 470 properly.
Seriously, aside from further posturing or BS, dude... you have absolutely no idea how balls-to-the-wall this deadline was. I am not exaggerating in the slightest. I got off the plane with my sample cards, tested full-bore for days with almost no sleep, barely had time to write anything at all. I wish to god I had just an extra 2 days to have done this properly. Graphics card companies are somewhat overprotective of information and give us very little time. Chris has a well-deserved co-credit for hours of editing and sprucing up the content, not for co-testing. This isn't an excuse, this is the reality--you have to pick what you can do in the time allotted, I probably tested more games than I could handle but we thought that was important. I probably sacrificed a proper HQV line-up in favor of the morphological AA testing, I do think that's more of a landmark for the 6800 series. I have planned to do a large-scale HQV article in the near future anyway, so it will get covered.
Most sites didn't test any HQV or morphological AA, it's not because they're lazy, it's probably because they weren't willing to kill themselves for the sake of a single article.
I think it's pretty clear that if the maximum is 210 points, a ~200 point score would suggest the 6800 series is pretty close to perfect. That in itself is useful information, and it definitely means something.
Would I liked to have added more results, say from a 470? Of course I would, but like the article states I didn't have time. I don't think you're getting this part. As it was, the article was posted after the NDA lifted. I'm not sure why you're having trouble assimilating that.
I do plan to follow up in the near future with an exhaustive HQV 2.0 test of a bunch of new cards. But as for this review, I didn't have time. The test I did manage to squeeze in was probably partially the reason why we published a little late, and I certainly wasn't going to add another couple hours on top of that with another fresh install so I could test the 470 properly.
Seriously, aside from further posturing or BS, dude... you have absolutely no idea how balls-to-the-wall this deadline was. I am not exaggerating in the slightest. I got off the plane with my sample cards, tested full-bore for days with almost no sleep, barely had time to write anything at all. I wish to god I had just an extra 2 days to have done this properly. Graphics card companies are somewhat overprotective of information and give us very little time. Chris has a well-deserved co-credit for hours of editing and sprucing up the content, not for co-testing. This isn't an excuse, this is the reality--you have to pick what you can do in the time allotted, I probably tested more games than I could handle but we thought that was important. I probably sacrificed a proper HQV line-up in favor of the morphological AA testing, I do think that's more of a landmark for the 6800 series. I have planned to do a large-scale HQV article in the near future anyway, so it will get covered.
Most sites didn't test any HQV or morphological AA, it's not because they're lazy, it's probably because they weren't willing to kill themselves for the sake of a single article.
dragonsqrrl
Distinguished
- Nov 19, 2009
- 1,280
- 0
- 19,290
Some leaked HD6970 benchmarks have surfaced, and at least in synthetic benchmarks it seems to perform about 10% better then the stock GTX480. It's looking pretty impressive, especially the Heaven results.
The card is scoring X12000 in 3D Mark Vantage, with a P score of P24499. Unigine Heaven results (1920 x 1200, 4xAA, 16xAF) giving a result of 36.6 fps.
* a HD5870 scores P19337 in 3D Mark Vantage, and 17.3 fps in Heaven
* a GTX480 scores P21106 in 3D Mark Vantage, and 29.5 fps in Heaven
http://guru3d.com/news/amd-radeon-hd-6970-results-surface/
The card is scoring X12000 in 3D Mark Vantage, with a P score of P24499. Unigine Heaven results (1920 x 1200, 4xAA, 16xAF) giving a result of 36.6 fps.
* a HD5870 scores P19337 in 3D Mark Vantage, and 17.3 fps in Heaven
* a GTX480 scores P21106 in 3D Mark Vantage, and 29.5 fps in Heaven
http://guru3d.com/news/amd-radeon-hd-6970-results-surface/
[citation][nom]demonhorde665[/nom]thugh they are getting confusing with teh naming , i'm intriqued by teh 6850 i noticed in every benchmark it was behidn teh 5850 and ahead of the 5830 yet cost much clsoer to waht eh 5770 costed when it first came out, bassicaly ti is getting near 5850 performance for alot less money than a 5850 , sounds liek it wil be the one i would grap but first i'd like to see benches next to a 5770 since i already got a 5770[/citation]
Whoa, liek, teh, waht... are you a lolcat?
Whoa, liek, teh, waht... are you a lolcat?
shineon2010
Distinguished
- Jan 4, 2010
- 274
- 0
- 18,790
builderbobftw
Splendid
- Feb 6, 2010
- 3,601
- 0
- 22,810
doogansquest
Distinguished
- Jul 29, 2010
- 86
- 0
- 18,630
I own a GTX470, and I love it. But AMD is just brilliant with the graphics market right now. They've owned it for the last 12-16 months, and this is a dagger to completely deflate NVidia's GTX400 series sails. Now comparable cards for cheaper. Just wait until the 6xxx drivers come out!
WarraWarra
Distinguished
- Aug 19, 2007
- 252
- 0
- 18,790
Why is it that the more expensive cards suffer so much at 2560x1600 and what should be a weak card 6850 percentage wise looses less fps at high versus big diff at low res as compared to expensive cards ?
Is it just that NV has actual weaker cards when cheaper as opposed to same 6870 with locked cores / locked features labeled as 6850 and causing this results ?
Is it just that NV has actual weaker cards when cheaper as opposed to same 6870 with locked cores / locked features labeled as 6850 and causing this results ?
- Status
TRENDING THREADS
-
-
News First credible report of RTX 5090 FE with melted connector appears — third-party cable likely cause- Started by Admin
- Replies: 53
-
Discussion What's your favourite video game you've been playing?- Started by amdfangirl
- Replies: 4K
Latest posts
-
Question Thunderbolt: My experience, why I ditched it, and what I did.- Latest: dev_cyberpunk
-
-
Facebook
Twitter
Instagram