6950 2GB Sapphire Toxic OC Review [H]

[GoodRevrnd]

If I adjust the power slider in CCC then those clocks seem to overwrite the Trixx clocks. This is a problem since CCC only goes up to 950mhz. Is there a work around?

 

[happy medium]

If I adjust the power slider in CCC then those clocks seem to overwrite the Trixx clocks. This is a problem since CCC only goes up to 950mhz. Is there a work around?

Only use one overclocking program at a time. I disabled overclocking in CCC r and just used MSI afterburner with enable" unofficial overclocking" enabled in cfg. file..

 

[apoppin]

I cant stand the dribble that is a 30 page long testing of games at 1200x800 or whatever with a 580 or a 6970 which alienbabeltech does for example. We really need to know how a Radeon 6990 does at that resolution in Farcry or better yet, Hawx (lol..).

As usual, you have no clue and give misinformation about ABT.
:thumbsdown:

Here is my game benching list. All games are benched at 1680x1050,1920x1080 or 1920x1200, 2560x1600 and 5760x1080.
i have never benched Far Cry for ABT.

• Batman: Arkham Asylum
• Serious Sam, Second Encounter HD (2010)
• Wolfenstein
• Left 4 Dead
• Mafia II
• Crysis
• World in Conflict
• Far Cry 2
• Just Cause 2
• Resident Evil 5
• Alien vs. Predator
• Battleforge
• STALKER, Call of Pripyat
• F1 2010
• Metro 2033
• Lost Planet 2
• H.A.W.X. 2
• Civilization 5
• Total War: Shogun II
• DiRT 3
• Crysis 2

 

[happy medium]

Here is my game benching list. All games are benched at 1680x1050,1920x1080 or 1920x1200, 2560x1600 and 5760x1080.
i have never benched Far Cry for ABT.


Batman: Arkham Asylum
Serious Sam, Second Encounter HD (2010)
Wolfenstein
Left 4 Dead
Mafia II
Crysis
World in Conflict
Far Cry 2
Just Cause 2
Resident Evil 5
Alien vs. Predator
Battleforge
STALKER, Call of Pripyat
F1 2010
Metro 2033
Lost Planet 2
H.A.W.X. 2
Civilization 5
Total War: Shogun II
DiRT 3
Crysis 2

Thats gotta be the best freaking game list I've seen on the internet, along with the most informative in depth reviews I've seen.

I much rather see that list of games at 3 popular resolutions, than 4 games at 2500x1600 with 16xfsaa at 31 fps minimums and 1 clunky old apples for apples comparison.

Why would you just use 2500x1600 as your main set of bechmarks anyway, like they do at (H)? Mabe to please the 2% of the population that uses it?

 

[GoodRevrnd]

Only use one overclocking program at a time. I disabled overclocking in CCC r and just used MSI afterburner with enable" unofficial overclocking" enabled in cfg. file..

So does the bug only occur when OCing with CCC or what? Does manually setting voltage in Trixx override the power bug? It sounds like Powertune is something that kicks on as a safeguard only with CCC overclocking but I'm not familiar with this feature.

 

[HendrixFan]

Why would you just use 2500x1600 as your main set of bechmarks anyway, like they do at (H)? Mabe to please the 2% of the population that uses it?

If they include other resolutions, how is that the "main" set? Honest question.

My guess for focusing on that resolution is that it is the maximum for gaming unless you are running SLI or crossfire with multiple monitors. It is the most accurate way to test the "limit" of any given card.

 

[happy medium]

If they include other resolutions, how is that the "main" set? Honest question.

My guess for focusing on that resolution is that it is the maximum for gaming unless you are running SLI or crossfire with multiple monitors. It is the most accurate way to test the "limit" of any given card.

What I mean is for 2500x1600 you see a gigantic set of benchmarks , a full 2 paragraghs of information with another 2 graphs of gameplat benchmarks.
This takes up 3/4 of the page.

And at the BOTTOM of the page......
2 graphs and a few sentences about the much more popular resolution.

http://www.hardocp.com/article/2011/08/29/sapphire_toxic_hd6950_video_card_review/5

 

[dennilfloss]

That thing is really nice but about one inch longer than my 4870. Wouldn't fit in my case.:|

 

[Grooveriding]

If they include other resolutions, how is that the "main" set? Honest question.

My guess for focusing on that resolution is that it is the maximum for gaming unless you are running SLI or crossfire with multiple monitors. It is the most accurate way to test the "limit" of any given card.

I see using 2560x1600 as no different than testing things like 580 SLI or a 6990. Enthusiasts want to see what the top end performance is like whether or not most of them feel the need to use it themselves.

 

[3DVagabond]

Why would you just use 2500x1600 as your main set of bechmarks anyway, like they do at (H)? Mabe to please the 2% of the population that uses it?

It's called "Maximum Playable Settings". That's one of the main features that differentiates [H] reviews from others. Also, they actually play the games rather than run canned benches, and they don't bother running tests on old games, or, games that the cards can crank through with max settings at +100fps.

That is more info that you wouldn't have, if everyone just reviewed the cards the same way.

 

[happy medium]

It's called "Maximum Playable Settings". That's one of the main features that differentiates [H] reviews from others. Also, they actually play the games rather than run canned benches, and they don't bother running tests on old games, or, games that the cards can crank through with max settings at +100fps.

That is more info that you wouldn't have, if everyone just reviewed the cards the same way.

SO your saying that they do cater to the top 1/2% off enthusiast?
That was my point. I guess if your target framerate for Dirt 3 is minimum 30 fps and you run at 2500x1600, its a good site to go to.

 

[Elfear]

SO your saying that they do cater to the top 1/2% off enthusiast?
That was my point. I guess if your target framerate for Dirt 3 is minimum 30 fps and you run at 2500x1600, its a good site to go to.

Lol. What percentage of people do you think buy a $500 GTX 580? Why would I want to read a review of the top GPUs at 1080p when a $200 GPU can get the job done? Both expensive graphics cards and high resolutions are used by only a small percentage of people. They kind of go together. :hmm:

Besides, as others have pointed out it's just the way [H] does reviews. They find the minimum playable settings for each card and compare. You can either use high resolutions or 16xAA. Personally, I could care less about super high levels of AA since the visual return is negligible after a certain point while resolution scales pretty well with visual quality.

 

[happy medium]

What percentage of people do you think buy a $500 GTX 580?

Alot less than those who buy a 269$AR gtx570 and overclock and get the same fps @ the same resolution. WHich brings me to my other question since knowone seems to answer any of them, but just scoots around them.

WHy no 269$ gtx570 overclocked benchmarks in the apples to apples part of the review?
Seems more like a red apples vs green apples to me.

The only answer I recieved over at (H) was time restraints but they had plenty of time to fix all the issues trying to get the 6950 to unlock and overclock.

My guess is if you overclocked a gtx570 card at the same price point, the fanboys would not have nothing to pound there chest about, especially at 1900x1080.

 

[3DVagabond]

SO your saying that they do cater to the top 1/2% off enthusiast?
That was my point. I guess if your target framerate for Dirt 3 is minimum 30 fps and you run at 2500x1600, its a good site to go to.

No, I'm not saying that. That is what you are attempting to spin. I find it difficult to believe that you lack the faculties to understand [H]'s review technique.

I understand that you prefer 30 games running 6 resolutions of the same canned frames of gameplay and synthetic benches to base your online discussions on. When looking to buy a card, that is good info to have to give you a general feel of where a product lies in the market landscape. It's not all of the info you need to make an informed decision, though. It's too easily manipulated. Maximum playable settings while really playing real games is very pertinent info to have. We don't need another site to give us the same old canned results. That is what I'm saying.

 

[happy medium]

ABT does not use all canned benchmarks and niether does Anandtech according to the reviewers I've spoke with .Where do you get this information from? and how many newer games that are being reviewed actually have a built in canned benchmark? Not many.

ARe you saying that Nvidia ONLY has the ability to make a canned banchmarks faster on there cards and AMD dont? Where is this unfair ?

You wanna talk manipulation? How about because a 6950 unlocked/overclocked at the same price point as a gtx570, is faster @ 2500x1600, the 6950 card is faster. Thats essentially what (H) is implying/saying in this review.

At no point in the review did they say "but dont forget gtx570's overclock at least 11% with no voltage increase and or unlock (which is not garanteed) and would beat this card in 9 out of 10 benchmarks at normal resolutions..

I love how some of you guys actually believe that when (H) does a benchmark showing a 6950 card beating a gtx570 and 20 others dont that somehow (H) is right and the other 20 reviews sites are wrong.

OK stop the presses, (H) is the only place to get real benchmarks that count.

 

[Grooveriding]

Why include a 570 ? 570 is irrelevant at 2560x1600, not enough VRAM.

 

[Madcatatlas]

Alot less than those who buy a 269$AR gtx570 and overclock and get the same fps @ the same resolution. WHich brings me to my other question since knowone seems to answer any of them, but just scoots around them.

WHy no 269$ gtx570 overclocked benchmarks in the apples to apples part of the review?
Seems more like a red apples vs green apples to me.

The only answer I recieved over at (H) was time restraints but they had plenty of time to fix all the issues trying to get the 6950 to unlock and overclock.

My guess is if you overclocked a gtx570 card at the same price point, the fanboys would not have nothing to pound there chest about, especially at 1900x1080.

What is all this "fanboy" talk you do? I can see your colors strutting nice and green, but who are these fanboys you so repeated try to provoke with your posts?

If you cant see the GOOD in having different methods for testing hardware, you should be ashamed and go learn some basic reflection skills.

 

[happy medium]

If you cant see the GOOD in having different methods for testing hardware

What different methods?
SO are you saying that Apoppin and Ryan lied to me?
All sites have different methods, does that make them have the right method, of coarse not.
I'm saying that (H) methods cater to a elite group of enthusiast that use hardware that only 1% of many can afford, and that method sucks. ANd they also seem to think because one card wins at 2500x1600 16xfssa that it will somehow win at lower more commonly used resolutions which most readers play at,which is completly false.

They tell half the story, and usually half the story that don't really pertain to 99% of the viewers out there.

I'll end my debate there.

 

[3DVagabond]

ABT does not use all canned benchmarks and niether does Anandtech according to the reviewers I've spoke with .Where do you get this information from? and how many newer games that are being reviewed actually have a built in canned benchmark? Not many.

Never mentioned ABT or AT. I was only attempting to answer a question of why [H] does what they do and the value of it. I don't come in to try and crap on a site or a review style. Again, I don't believe for a minute that you aren't smart enough to know what I said.

ARe you saying that Nvidia ONLY has the ability to make a canned banchmarks faster on there cards and AMD dont? Where is this unfair ?

Nope. Never said any of that, either. All I said is I want more info than that because that info is of limited use. Are you trying to spin what I say just to discredit me? Surely you can read what I posted and not see anything you are crediting me with in my posts.

You wanna talk manipulation? How about because a 6950 unlocked/overclocked at the same price point as a gtx570, is faster @ 2500x1600, the 6950 card is faster. Thats essentially what (H) is implying/saying in this review.

At no point in the review did they say "but dont forget gtx570's overclock at least 11% with no voltage increase and or unlock (which is not garanteed) and would beat this card in 9 out of 10 benchmarks at normal resolutions..

Clearly shows the performance of a stock 6950 as well. How many reviews to you see where the comparison cards are O/C'd? Typically it's just the review example they do it with. This is not an [H] exclusive.

I love how some of you guys actually believe that when (H) does a benchmark showing a 6950 card beating a gtx570 and 20 others dont that somehow (H) is right and the other 20 reviews sites are wrong.

Never said that either. :spin-spin-spin:

OK stop the presses, (H) is the only place to get real benchmarks that count.

More stuff I never said. You're really on a roll misrepresenting my post. Talking about canned benchmarks I did say this, "When looking to buy a card, that is good info to have to give you a general feel of where a product lies in the market landscape. It's not all of the info you need to make an informed decision, though." See. I read both styles of reviews. I'm just glad that the info is available to help make a better comparison.

minimum character generator.

 

[GoodRevrnd]

So does the bug only occur when OCing with CCC or what? Does manually setting voltage in Trixx override the power bug? It sounds like Powertune is something that kicks on as a safeguard only with CCC overclocking but I'm not familiar with this feature.

*cough*

 

[happy medium]

*cough*

I'm sorry, I dont use Trixx so I cant help you any further.
I just know having 2 programs trying to control your overclock at the same time never worked for me.

 

[Will Robinson]

So let me see if I have this correct...?
You don't have a 6950,are not planning on buying one(as your post history clearly indicates) and yet you now have gone to the effort of thread crapping across multiple ,reputable websites in order to discredit their testing methods and somehow elevate your preferred choice of card(GTX570)to be the pwnage card of the year?
http://hardforum.com/showthread.php?t=1632724&page=4
Ew K...
"This gaming performance the SAPPHIRE TOXIC HD6950 produced stomped its competition in the relevant price range. On top of the great overclock that this card comes with, we had plenty of head room to push it even farther. With the overclock on the faster 1536 Shader BIOS the card was able to produce equal or better gaming performance than the much more expensive HD 6970 and GeForce GTX 580, at a fraction of the price". [HardOcp].........ouch

ps.Quoting "Kitguru" as a superior web review to the [H] results is amusing.

 

[badb0y]

I do remember that, but then I more recently remember their somewhat irrational hatred of crytek just because a game was hard to install. It's a two edged sword - that arrogance that allowed them to stand up to nvidia or ati or whoever was great when used for the right reasons, but these days it has just become an arrogance to say or do whatever they like. That's a shame.

[H] reviewers have become fanboys with a bigger audience, and if you don't like it well you know where the door is. This doesn't sit well with trying to produce professional review of anything. In this case the conclusion should have either been based on all the cards been o/c or none of them. As they picked o/c they should have also made more of the fact there are no guarantee's - you might get lucky and it'll unlock and o/c to 6970 levels, you might not.

Brent - I take it you don't really want my opinion because if I'd posted that on [H] you'd have banned me or at least told me where to go. Not even sure why you are here as you don't even like anandtech as far as I know - I never see you linking to anandtech articles on your site. If you want to talk further feel free to PM me.

Crysis 2 was disappointing when it released. By the time they released the DX 11 patch I already lost interest and moved on.

Crytek sold us out for a bunch of pussy 360 gamers

Pretty much sums up my feelings of the game.

 

[Spikesoldier]

How is a unlocked overclocked 6950 as good as a stock gtx570 at the same price?
gtx570 for 269$ AR..........
http://www.newegg.com/Product/Produc...82E16814134125

Wouldn't you have to overclock the gtx570 also?

I mean the gtx560ti cost 65$ less too and will overclock to close to the same performance.

Am I missing something?
I mean this is sort of like comparing a gtx460 @ 850 core to a 6870 at the same price point, is it not?
I wonder why Ryan at Anandtech got chewed out but for Kyle its ok?

its easy to argue price/performance when you are selecting the cheapest card in the 570 family pitted against a top sku and cooling for the 6950 family.

op could easily get the cheapest 6950 vs cheapest 570 and now the difference between what the 570 offers and the 6950 has is met with a larger difference in price.

 

[happy medium]

op could easily get the cheapest 6950 vs cheapest 570 and now the difference between what the 570 offers and the 6950 has is met with a larger difference in price.
__________________

The difference is 20$ more for the gtx570 and no unlocking for the cheapest 250$ 6950 so it would be even slower. , Prices from Newegg.

I remember when the 6950 2gb was 219$ AR , it was a good deal then.
The 320$ gtx570 came down about 50$ since then and the 6950 went up 30$.