NDA for Cayman is lifted 22nd?!

Page 5 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Teizo

Golden Member
Oct 28, 2010
1,271
31
91
Of course you would if you're biased in the same respect (i.e. towards NVIDIA). If you can't see anything wrong with Anandtech going against their "editorial policy" simply because a company asked them to, then you have no place in this discussion.
Was Nvidia aware of Anandtech's policy? Or, did they just lower the price of their card and ask them to add it to the review since it was in a similiar price bracket?

It has been stated over and over again by Anandtech that Nvidia did not force them to include the card in their review and they chose to include the card in their review at their own discretion since they decided that though it went against their policy, Nvidia's request did have merit.

The notion that Nvidia somehow played dirty is silly. This is a business, and Nvidia made a business decision to stay competitive. AMD forced their hand with two good cards, but it is foolish to think Nvidia would just sit there and take a beating lying down. They chose to compete, and did so aggressively, and it was a move that obviously has paid off for them, especially now that both the 6850 and 6870 have seen price increases which has detracted from their incredible initial value.
 
Last edited:

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
If I felt that strongly about Anandtech and there bias , I cetainly would not post in there forums.

Anandtech actually invited it's readers to voice their opinion on the subject. Shows that they don't want a bunch of sheeple as subscribers. They were very forthcoming. I doubt there was any malice on their part. I think it also shows that they weren't entirely comfortable with the decision either. No one has condemned Anandtech for this one review. If it became a habit, that would be a different story, for me anyway.

There are a few people around here that, if they can't rebut the post they attack the poster. This is what you are doing when you post like you did above. It's very poor style.
 

RussianSensation

Elite Member
Sep 5, 2003
19,458
765
126
Did you just say something about fanboy excuses...?

HINT: I called 35% nonsense and I named the usual _real_ margin. ;)

So instead of providing any support, you just called my claim "non-sense".

#1) Resolution Perspective - Checking for consistency of performance advantage across many resolutions with and without AA

A brief summary across most resolutions for 11 games. I averaged every single resolution with and without AA - GTX580 is 36% faster than the HD5870.

If you don't like that website, there is another popular one which summarized all resolutions for 16 games.
GTX580 = 100 vs. HD5870 = 77.

100/70 gives 30% faster performance on average.

#2) Canned Benches vs. Actual Gameplay Perspective - Some argue that built-in game benchmarks may not be representative of actual gameplay. Fair enough.

Every single benchmark below is a Manually recorded run with videos provided for each game section tested (usually 10 min of gameplay). Results provided at 1920x1080 VHQ - 4AA or 8AA

GTX480 vs. 5870

1. Test Drive Unlimited 2 = + 39%
2. Two Worlds 2 = + 48%
3. COD : Black Ops = + 25%
4. ArcaniA: Gothic 4 = + 9%
5. Medal of Honor = equal at 62 fps (but GTX480 has a 24% advantage in Minimum framerates)
6. Front Mission Evolved = + 12%
7. Lara Croft and the Guardian of Light = + 21%
8. Darksiders = + 40%
10. F1 2010 = - 8%
11. Final Fantasy 14 = + 28%
12. Kane and Linch 2 = - 10%
13. Lost Planet 2 = + 95%
14. Mafia 2 (without PhysX) = + 45%
15. Starcraft 2 = + 14%
16. ArmA 2 = + 19%
17. Ghost Sniper Warrior = + 8%

GTX480 is 22% faster on average than a HD5870 in 17 games.

Although I am not a fan of some of these games, I thought referring to a greater variety of less mainstream titles can provide for an ever wider perspective. Interesting how HD5870 performs even worse once we start looking at non-mainstream games and when testing is done manually. I guess it makes sense why HardOCP included F1 2010 into their review - it's one of the few games remaining where 5870 has any chance at all against the GTX580.

#3) Review Sample Size Perspective - If we accept the possibility that some reviews can be outliers, by averaging as many reviews as possible, we can minimize the chance that any biased review will heavily influence the overall conclusion.

140 Benchmarks across 14 Review Websites:

GTX580 = 38.8% faster than HD5870 overall
GTX580 = 49% faster than HD5870 in DX11 overall
GTX580 = 39.4% faster than HD5870 in DX11 without Hawx 2 and Unigine (2 benches considered "NV biased")
GTX580 = 18.8% faster than GTX480 overall

If you feel these 14 review websites are all biased, feel free to point out the "proper" review website.
 
Last edited:

Arkadrel

Diamond Member
Oct 19, 2010
3,681
2
0
From PCWorld:

192574-gtx480chart-hawx_original.jpg



1920x1200= 92 fps vs 76 fps. (21% differnce)


192574-gtx480chart-dirt2_original.jpg


1920x1200= 79 fps vs 66 fps. (19% differnce)



192574-gtx480chart-stalker_original.jpg


1920x1200= 41 fps vs 42 fps. ( -2.4% differnce)



192574-gtx480chart-wic_original.jpg



1920x1200= 70 fps vs 57 fps. (22% differnce)






Its probably fair to say that in games, the 480 is about ~20% faster than a 5870.

@RussianSensation, that russian site your linking, the links dont work, so.... its just your # ect without anything to back them up. You should have used numbers from a site that worked, its more crediable/believable that way.

Your spread sheet is abit misleading though... because its run at 2560x1600 (with high AA settings) in all of them, at which point the extra RAM on the 480 becomes something that gives it a boost compaired to the 1GB cards. There are probably instances where the 1GB is ram limited by that resolution, could you do the spread sheet with 1920x1200? or 1680x1050 resolutions?

You might be giveing a slightly false impression of the cards and the differnces between them in performances, when 1 card might be ram limited and another not to get the result differnces.
 
Last edited:

Rifter

Lifer
Oct 9, 1999
11,522
751
126
@RussianSensation, that russian site your linking, the links dont work, so.... its just your # ect without anything to back them up. You should have used numbers from a site that worked, its more crediable/believable that way.

The site works, its slower than hell and a bit buggy with the scripts but it works. I checked up to game # 7 and all info he posted is backed up in links provided. Even have gameplay footage videos(which taker FOREVER to load, good thing im bored at work).
 

Arkadrel

Diamond Member
Oct 19, 2010
3,681
2
0
@Rifterut,

o_O I just get a error message when I try those links.


GTX580 = 38.8% faster than HD5870 overall
GTX580 = 39.4% faster than HD5870 in DX11 without Hawx 2 and Unigine (2 benches considered "NV biased")
GTX580 = 18.8% faster than GTX480 overall

^ above isnt hard to believe though, it makes sense.

The 6970 will be disapointing if its not atleast 40% faster on avg than the 5870's then so it can match the 580.
 
Last edited:

Rifter

Lifer
Oct 9, 1999
11,522
751
126
i get a error message about the scripts too, but if you hit ok/yes instead of cancel it does load the page fine, just slow.
 

Rifter

Lifer
Oct 9, 1999
11,522
751
126
should also mention im running IE 8 at work, will try it on firefox when i get home to see if that works
 

Ares1214

Senior member
Sep 12, 2010
268
0
0
TPU tested a load of games, and got the 580 as being an average of 25% faster in all games they tested. So cant we just use 25% as how much faster it is?

perfrel_1920.gif


A lot of people think the 480 is SO much faster than the 5870, but its really only 10%.
 

maddie

Diamond Member
Jul 18, 2010
4,878
4,951
136
Wow, how quickly people forget issues or blur the lines (on purpose or not). Having an overclocked card in the Anandtech 6850/6870 review was not the problem. If Anandtech had taken any GTX 460 sample and said "hey guys, we cranked the clocks on the 460 and added it to the charts for the comparison since it's a direct competitor and we know GTX 460's clock well, yada yada yada" I think they would have been thumbed up/high-fived for an enlightening review. The problem was that Anandtech specifically went against its editorial policy simply because NVIDIA asked them too. That really makes me, and I'm sure many others, question their journalistic integrity. Maybe it was just a bad call and they've learned their lesson, we all make them; however, that still doesn't stop me from asking what else they might or have caved in on as well.

Is that right? How do you figure? And please don't link me to one of T2k's posts.



As we mentioned on the front page of this article, AMD and NVIDIA don’t officially have competing products at the same price points. The 6870 and 6850 are more expensive than the GTX 460 1GB and 768MB respectively, and above the 6870 is the GTX 470. However NVIDIA is particularly keen to have a competitor to the 6870 that isn’t a GTX 470, and so they’re pushing a 2nd option: a factory overclocked GTX 460 1GB.


As a matter of editorial policy we do not include overclocked cards on general reviews. As a product, reference cards will continue to be produced for quite a while, with good products continuing on for years. Overclocked cards on the other hand come and go depending on market conditions, and even worse no two overclocked cards are alike. If we did normally include overclocked cards, our charts would be full of cards that are only different by 5MHz.

However with the 6800 launch NVIDIA is pushing the overclocked GTX 460 option far harder than we’ve seen them push overclocked cards in the past –we had an EVGA GTX 460 1GB FTW on our doorstep before we were even back from Los Angeles. Given how well the GTX 460 overclocks and how many heavily overclocked cards there are on the market, we believe there is at least some merit to NVIDIA’s arguments, so in this case we went ahead and included the EVGA card in our review. As a reference point it's clocked at 850Mhz and 4GHz memory versus 675MHz core and 3.6MHz memory for a stock GTX 460, giving it a massive 26% core overclock and a much more moderate 11% memory overclock.


Happy medium:

Would an apology for being wrong be so difficult to accomplish. I seem to find you ignore an argument as soon as you have lost, pretending it never existed.
 

Rifter

Lifer
Oct 9, 1999
11,522
751
126
What is with all this backlash over the 460 FTW, holy crap guys.

Personally i like overclocked cards in reviews as i will overclock every card i buy, why because its free(discounting elec costs, but elec is cheap here so TO ME it doesnt matter) performance and i have been doing it for years and have never had one burn out on me while still being in its useable lifespan.

Then there are the people who dont overclock either because they are not that computer literate or just dont wish to or are worried about warranty etc. To them they just want the best card for the money thats plug and play and dont care about overclocking at all. Fair enough but to them a factory OC card(like the FTW) which carries a warranty is the same as a factory clocked card, they buy it, it has warranty and its plug and play.

The only issue here is the whole anandtech went against there policy. Which i admit is weak. However the crybabies who think its some kind of nvidia conspiracy are out of line, ive been around anandtech way to long to believe they would bend to nvidias will especially after they have release statements that have been quoted in this thread explaining themselves.


Bottom line is if the card is priced in the same range and has a warranty in the same time range then the card should be in the review IMO.
 

ArchAngel777

Diamond Member
Dec 24, 2000
5,223
61
91
If you feel these 14 review websites are all biased, feel free to point out the "proper" review website.

I do not think they are biased. By the way, I think it is a bit misleading to average out all resolutions, with AA on and then off. As others noted, I would object to this because 1) The extra video memory would have given it a distinct advantaged when running the insane settings (2560x1600 w/AA). 2) You see anomolies like Lost Planet which raises the overall average. I know that is what an average does, but when you have an extreme high and an extreme low, where they are not the norm nor represent the overall performance of PC gaming then they should be excluded from the average in my opinion.

Not that I really care, because these threads are an eye opener for me. The initial review had the GTX 480 barely faster than the 5870 (less than 10% overall) and now it appears to be a solid 20% faster. This, when combined with the faster GTX 580, makes the 5870 trail by at least 35-40% in my opinion. I have my doubts on whether the 6970 is going to even match what the GTX 580 has done. The best thing I am hoping for is a price war so I can snag one of those GTX 580's.... Unless Cayman really comes through.
 
Last edited:

RaistlinZ

Diamond Member
Oct 15, 2001
7,470
9
91
Yeah, if the 6870 turns out to be 5-10% slower than the 580, I hope they price it aggressively.
 

Lonyo

Lifer
Aug 10, 2002
21,938
6
81
I do not think they are biased. By the way, I think it is a bit misleading to average out all resolutions, with AA on and then off. As others noted, I would object to this because 1) The extra video memory would have given it a distinct advantaged when running the insane settings (2560x1600 w/AA). 2) You see anomolies like Lost Planet which raises the overall average. I know that is what an average does, but when you have an extreme high and an extreme low, where they are not the norm nor represent the overall performance of PC gaming then they should be excluded from the average in my opinion.

Not that I really care, because these threads are an eye opener for me. The initial review had the GTX 480 barely faster than the 5870 (less than 10% overall) and now it appears to be a solid 20% faster. This, when combined with the faster GTX 580, makes the 5870 trail by at least 35-40% in my opinion. I have my doubts on whether the 6970 is going to even match what the GTX 580 has done. The best thing I am hoping for is a price war so I can snag one of those GTX 580's.... Unless Cayman really comes through.

Averages should always be resolution based rather than combining every resolution, otherwise it makes no sense.
Typically NV have had a greater lead (or smaller deficit) at lower resolutions, and the gap decreases as you increase resolution, at least towards 2560x1600, where RAM can sometimes come into play.

Averaging all resolutions doesn't help anyone determine anything, especially if you are trying to inform people. I don't care what the difference between A and B is at 1280x1024, 1680x1050, 1920x1200 and 2560x1600 all averaged together if I'm playing at 1920x1200.
 

ArchAngel777

Diamond Member
Dec 24, 2000
5,223
61
91
Averages should always be resolution based rather than combining every resolution, otherwise it makes no sense.
Typically NV have had a greater lead (or smaller deficit) at lower resolutions, and the gap decreases as you increase resolution, at least towards 2560x1600, where RAM can sometimes come into play.

Averaging all resolutions doesn't help anyone determine anything, especially if you are trying to inform people. I don't care what the difference between A and B is at 1280x1024, 1680x1050, 1920x1200 and 2560x1600 all averaged together if I'm playing at 1920x1200.

Yep, 100% agreed. I play at 1680x1050 and have no plans to upgrade to a larger display anytime soon. I guess that is also why my GTX 280 is still doing the job...
 

tviceman

Diamond Member
Mar 25, 2008
6,734
514
126
www.facebook.com
I still don't understand the horrid controversy around Anand (and other sites) using a factory configured retail card in a review as a comparison to other cards. I agree on the point he made that there are dozens of preconfigured cards operating at different speeds, but having a card or two that is configured to actually provide a substantial difference in benchmarking and gameplay (like the evga gtx460 ftw) is beneficial to readers trying to decide which card at what price is the best purchase for them. He didn't overclock the card himself, and the card in question is still available for purchase right now.

Nvidia has gf114 coming (aka gf104 revised), and they're probably going to release two different versions of the chip and they're both probably going to be faster than the Barts XT and Pro (respectively). AMD would be smart to sanction (if possible) aggressive factory-overclocked versions of the hd6870 to compete with the top gtx560 model, and I would personally like to see (and think it would be beneficial to those looking at upgrading) an "elite" factory clocked hd6870 card in the reviews vs. the gtx560, so long as both parts are priced similarly. However, on the same note, with the huge shader difference between barts and what cayman is supposed to have, I think there is a strong possibility there will be 3 cayman models like there was with cypress. So, more than likely, a 3rd tier cayman card will/would compete with the gtx560.
 

Triggaaar

Member
Sep 9, 2010
138
0
71
GTX580 = 38.8% faster than HD5870 overall

...

The 6970 will be disapointing if its not atleast 40% faster on avg than the 5870's then so it can match the 580.
Although some of the leaked info suggests the 6970 could be that quick, I certainly won't be disappointed if it's 10% slower than the 580 and 20% cheaper.
 

Dribble

Platinum Member
Aug 9, 2005
2,076
611
136
I still don't understand the horrid controversy around Anand (and other sites) using a factory configured retail card in a review as a comparison to other cards.

The only controversy is in the minds of a few fan boys who are worried about seeing their favourite manufacturer being seen in a negative light. Just shows they have lost the plot somewhat and now care more about some big ugly corporation then they do about discovering what is the best graphics card for their money.

For the rest of us its more information - we can see how the 6800's compare with both a stock and an o/c 460. More information is good, there is no negative side to it.
 

taltamir

Lifer
Mar 21, 2004
13,576
6
76
Although some of the leaked info suggests the 6970 could be that quick, I certainly won't be disappointed if it's 10% slower than the 580 and 20% cheaper.

that is quite true... It isn't just the performance, its also the performance/$ ratio.
 

digitaldurandal

Golden Member
Dec 3, 2009
1,828
0
76
I'd like to chime in on the 480 and 580 discussion going on.

First thing I would like to point out is that the 480 speeds have seen quite a few increases since launch. If any of the reviews were with launch drivers, then I believe at this point they should be ignored.

I do tend to agree that resolution and settings matter quite a bit and numbers do not do a great job alone of showing the differences between playing with one product or another.

If you truly are looking to choose between to competitive products my recommendation would be to look at the resolution you plan to play at. HardOCP does a good job of showing playability, although I always think they do not show enough competitive products in each review.

For some of us the 2560 resolution is indeed the most relative resolution, I use eyefinity and can say with a certainty that if you look at sli 480 surround vs cf5870 eyefinity minimum and average framerates that scalability and high resolutions are in NVs favor by far at 5720x1200.

For some consumers this may not matter, if you game at 16xx resolutions though I doubt you are looking at either of the cards compared. Therefor I think it is relatively prudent to include the 2560 resolution. Many of us that upgrade our cards often have 1920x1200 or higher.

I think that a good effort was made to show evidence of this 35% claim, using multiply sources. Resolution and AA notwithstanding I will repeat my assertion that the numbers alone do not mean much, however if you view minimum frame rates and the ability to run better settings in games you will find that essentially he is correct - the 580 currently smokes the 5870 by a very good margin.

Please take note that I am currently using an 5870 and before that I was using sli 285 gtxs. I consider myself a good judge of features and value and switched because I could sell my GTXs, gain eyefinity and pay for the third monitor with and 5870 from the sale.

If prices drop I will definitely be switching back unless the 69xx series scale well with good minimum frame rates in eyefinity.
 

digitaldurandal

Golden Member
Dec 3, 2009
1,828
0
76
that is quite true... It isn't just the performance, its also the performance/$ ratio.

What about features?
What if the average frame rate is 10% slower but the minimum is 30% slower? What if CF scaling is 20% worse?

It is difficult to place a specific value on a card solely based on the numbers.

I believe that with this generation we will find that if you are going for a single card the 69xx series may be the best value, but that if you plan to use a multigpu setup then NV will win. Of course I am doing my best to wait it out and see if my prediction proves true. Hopefully there will be no delays.
 

SlowSpyder

Lifer
Jan 12, 2005
17,305
1,001
126
So whats the problem? Same price point right? Highly avalable right? Are you just saying they should have called it a gtx460.5?

If I felt that strongly about Anandtech and there bias , I cetainly would not post in there forums.

The problem is you asked for someone to point out where it is AT's own policy that they broke, someone did, then the goal posts moved yet again...

Personally, I don't think I mind AT using the overclocked cards, but I do see why it is an issue with many people. If AT said as a policy they do not turn AMD AI on, yet AMD asked them to run a patch that forced it on a certain game that improved the benchmark in AMD's favor, I can see why people might not like that.

This situation is a little different. The overclocked 460's are available and priced copetitively, so they will be competing in sales. But the fact is that AT broke their own policy at the behest of Nvidia, so that Nvidia would look more competitive against AMD's new cards. I can see both sides of this one.

But you asked for someone to show you where AT said that it was against their own policy to use these overclocked cards. Some posters did. So...