MSI GeForce GTX 1060 Gaming X 6 GB

[Ungari]

I think he is not reffering to you, but other people who blindly buy products based on "brand popularity" instead of bang for buck (or in your case, absolute best performance).
It's like in my country the majorityof people buy nvidia just because is nvidia and don't know anything else. Even in the nvidia product costs more and performs the same. They see nvidia brand as superior in quality to amd for "reasons". They have never had an amd graphic card and talk bs of amd.

That is exactly what I meant!

@the54thvoid I was not calling you a monkey but I remarking on the way many people today make purchasing decisions based purely on popularity.

 

[Assimilator]

Much of it has to do with the Monkey See, Monkey Do popularity that Nvidia has enjoyed, as well as the proprietary GameWorks used by many games.
Thermals have been much improved with AMD now, and that used to be a concern to many buyers.

I think you'll find that in most countries, it simply boils down to pricing. For example, here in South Africa we can get custom design GTX 1060s for less than reference design RX 480s. I have no f**king idea why this is, but the fact of the matter is that if I had to pick between those two cards, there's only really 1 logical choice.

It's sad because back around the time of the Radeon 5000 series, the prices here were much closer to what they should be, which is why I went with AMD reference 5850 and then 6950 (unlocked to 6970). But since then the prices have effectively made NVIDIA the only game in town unless you're a die-hard AMD loyalist.

I don't know why AMD cards are so much more expensive here (and from comments in various videocard review threads, in other countries as well) but I feel it's something AMD should investigate. Someone is being greedy, and that is costing AMD sales.

 

[Ungari]

I think you'll find that in most countries, it simply boils down to pricing. For example, here in South Africa we can get custom design GTX 1060s for less than reference design RX 480s. I have no f**king idea why this is, but the fact of the matter is that if I had to pick between those two cards, there's only really 1 logical choice.

It's sad because back around the time of the Radeon 5000 series, the prices here were much closer to what they should be, which is why I went with AMD reference 5850 and then 6950 (unlocked to 6970). But since then the prices have effectively made NVIDIA the only game in town unless you're a die-hard AMD loyalist.

I don't know why AMD cards are so much more expensive here (and from comments in various videocard review threads, in other countries as well) but I feel it's something AMD should investigate. Someone is being greedy, and that is costing AMD sales.

Since Nvidia has been known in the past to spread baksheesh, I suspect that palms are being greased in these countries.

 

[xorbe]

GTX 1060 for 250$ does not exist on the market either. I bet there is better chance of RX480 4GB for 200$ appearing than GTX1060 for 250$.

$249 stock clock EVGA single fan mini card
https://www.amazon.com/gp/product/B01IPVSGEC/?tag=tec06d-20

$249 superclock (after $10 MIR, check EVGA website)
https://www.amazon.com/gp/product/B01IPVSLTC/?tag=tec06d-20

 

[the54thvoid]

That is exactly what I meant!

@the54thvoid I was not calling you a monkey but I remarking on the way many people today make purchasing decisions based purely on popularity.

Well, on balance, I am appalled at how expensive the GTX1080 is so despite its performance boost over my card, I wouldn't touch it with a barge pole. Even on principle.
I keep saying it but I really want AMD to come out with cannons firing on Vega but I have a bad feeling the GTX1080ti will be faster, even in DX12. There is after all only so much compute hardware you can throw into a chip before it melts with power draw.

 

[Assimilator]

Since Nvidia has been known in the past to spread baksheesh, I suspect that palms are being greased in these countries.

[citation needed]

 

[Dukenukemx]

Any reason why your Rise of the Tomb Raider results are different than HardOCPs? Did you update RotTR with the latest Async Compute update? The RX 480 fairs better in their test than in yours.

 

[Ungari]

Any reason why your Rise of the Tomb Raider results are different than HardOCPs? Did you update RotTR with the latest Async Compute update? The RX 480 fairs better in their test than in yours.

This is typical, as many reviewers are beholden to Team Green's Greenbacks.

 

[rtwjunkie]

This is typical, as many reviewers are beholden to Team Green's Greenbacks.

^Source and citations, please? Otherwise it is mere hyperbolic speculation.

 

[Ungari]

^Source and citations, please? Otherwise it is mere hyperbolic speculation.

http://semiaccurate.com/2013/10/05/much-nvidia-pay-origin-pc-drop-amd/

 

[okidna]

Any reason why your Rise of the Tomb Raider results are different than HardOCPs? Did you update RotTR with the latest Async Compute update? The RX 480 fairs better in their test than in yours.

High Settings @ HardOCP (with only textures at Very High) VS Very High Settings @ TechPowerUP ("All games are set to their highest quality setting unless indicated otherwise.")
AND
if you ever played RotTR you will know that FPS numbers are highly variable from area to area (there's area which much more graphically-heavy than other areas) and since both reviews did not state what kind of benchmark (is it the built-in benchmark? what area used for benchmarking? etc.) used to obtain those numbers, in my opinion, we cannot directly compare the results.

 

[Ungari]

RotTR you will know that FPS numbers are highly variable from area to area (there's area which much more graphically-heavy than other areas)

This is where those with an agenda will pick those areas that best show performance from the card.

 

[Dukenukemx]

High Settings @ HardOCP (with only textures at Very High) VS Very High Settings @ TechPowerUP ("All games are set to their highest quality setting unless indicated otherwise.")
AND
if you ever played RotTR you will know that FPS numbers are highly variable from area to area (there's area which much more graphically-heavy than other areas) and since both reviews did not state what kind of benchmark (is it the built-in benchmark? what area used for benchmarking? etc.) used to obtain those numbers, in my opinion, we cannot directly compare the results.

I would think that Very High Settings would be in favor of AMD rather Nvidia, since 1440p is more in favor of AMD than 1080p. But did TechPowerUp update their RotTR with the Async Compute patch? Doom Vulkan was noticeably missing as well as Hitman DX12. I know they said Hitman DX12 isn't stable, but not stable for tests or not stable cause reasons?

 

[okidna]

I would think that Very High Settings would be in favor of AMD rather Nvidia, since 1440p is more in favor of AMD than 1080p.

Again, we're comparing 2 different results from 2 different reviewers without knowing the 2nd most important variable (the first most will be game graphical settings) on this benchmark : what scene/area is used for benchmarking by each reviewer?

But did TechPowerUp update their RotTR with the Async Compute patch? Doom Vulkan was noticeably missing as well as Hitman DX12. I know they said Hitman DX12 isn't stable, but not stable for tests or not stable cause reasons?

You have to ask W1zzard about this.
And as for Hitman, reading a couple of threads in their Steam discussions, it seems to me that the DX12 performance & stability issues isn't 100% fixed.

http://steamcommunity.com/app/236870/discussions/0/359543951706487184/
http://steamcommunity.com/app/236870/discussions/0/366298942109173435/
http://steamcommunity.com/app/236870/discussions/0/359543951707036422/
http://steamcommunity.com/app/236870/discussions/0/359543951710206192/
http://steamcommunity.com/app/236870/discussions/0/359543951705760375/

 

[xorbe]

[H] recently changed their methodology to one that I approve of. They don't blindly use max settings. They use a common set of settings, same for both cards, that allow playable frame rates for both, because in the end, that's what people want. People DON'T want to know that at 4K ultra, that it runs at 15.2 fps, because nobody uses such a setting. One of my main complaints around here ...

 

[Ralfies]

[H] recently changed their methodology to one that I approve of. They don't blindly use max settings. They use a common set of settings, same for both cards, that allow playable frame rates for both, because in the end, that's what people want. People DON'T want to know that at 4K ultra, that it runs at 15.2 fps, because nobody uses such a setting. One of my main complaints around here ...

Using the same settings for all cards makes comparisons easier. W1zzard cannot bench his huge assortment of cards with different settings for each review. I don't think there's one right way to bench, so we're lucky to have many sites with different methodologies.

This isn't directed at you, xorbe, but it seems lately all W1z gets for his hard work is grief from members. We could all stand to be a little more appreciative.

Thanks for the great reviews, @W1zzard!

 

[rruff]

Thanks for the great reviews, @W1zzard!

I sure as hell appreciate his work! Hope he realizes the bitching is a tiny minority. Can't make everyone happy...

 

[Assimilator]

http://semiaccurate.com/2013/10/05/much-nvidia-pay-origin-pc-drop-amd/

Ah yes, good ol' Charlie. Even the name of his site is semi-accurate, a more appropriate moniker would be "ConsistentlyWrong". Everything he publishes is anti-Intel and anti-NVIDIA rumours disguised as news articles, with sources who are always unnamed to "protect" them. Recently he claimed GP106 doesn't even exist, two weeks later and the card has hit retail availability and not a peep from him.

The most amazing thing for me is that there are actually suckers who are willing to pay Charlie good money for his particular brand of bulls**t. Me personally, I'd rather do something more useful with my money, like set it on fire.

tl;dr if you use SemiAccurate as a news source, I'm going to point and laugh at you, and I'll be completely justified.

I know they said Hitman DX12 isn't stable, but not stable for tests or not stable cause reasons?

Hitman DX12 is, to put it politely, a s**t-show. There was a frontpage article on TPU a while back about how the game would immediately crash when launched in DX12 mode when the second episode was released. And as the poster above me has noted, there are still issues.

A lot of people who think DX12 is a magic bullet for performance and multi-GPU are going to be sorely disappointed. It may be faster, but it's a helluva lot more tricky to get right than DX11 ever was.

 

[Ungari]

tl;dr if you use SemiAccurate as a news source, I'm going to point and laugh at you, and I'll be completely justified.

Unfortunately, after Google sanitized Nvidia in the search engine this is all that came up.
There used to be many articles covering this and other scandals involving paid reviews.
The latest story suggesting impropriety was with the leakage of the Nvidia Reviewer's Guide where benchmark guidelines were distributed to sample reviewers that played to Pascal's strengths, and minimized it's weaknesses.

 

[Assimilator]

The latest story suggesting impropriety was with the leakage of the Nvidia Reviewer's Guide where benchmark guidelines were distributed to sample reviewers that played to Pascal's strengths, and minimized it's weaknesses.

Because of course AMD doesn't do the same.

 

[okidna]

The latest story suggesting impropriety was with the leakage of the Nvidia Reviewer's Guide where benchmark guidelines were distributed to sample reviewers that played to Pascal's strengths, and minimized it's weaknesses.

Every reviewer's guide are like that, cherry picked benchmark and tailored settings so it shows the product strength, not weaknesses. Other example :
Fury-X reviewer's guide : http://videocardz.com/56728/amd-radeon-r9-fury-x-reviewers-guide

Both AMD and NVIDIA are profit oriented company, as far as I know there is NO profit oriented company who want their weaknesses exploited especially in their own confidential document (in this case, reviewer's guide). You won't find AMD praising NVIDIA's lower power requirement and you won't find NVIDIA praising AMD's DirectX 12 performance. They're trying to show their strength, to promote and sell their own products not to praise competitor products.

At the end of the day, for us end user, reviewer's guide means jackshit. As long as there's reputable and trusted reviewers (or tech savvy/enthusiast users) every lies or marketing stunt in reviewer's guide can be proven wrong, and also if the real world performance does not satisfy the user, the product simply won't sell.

 

[Ungari]

As long as there's reputable and trusted reviewers (or tech savvy/enthusiast users) every lies or marketing stunt in reviewer's guide can be proven wrong,

That is the problem.
The latest reviews from many popular tech sites, including this one; have skewed their tests in favor of Team Green to the point where one must question their integrity.
This video shows how many sites failed to use DX12 and Vulkan in their testing, and used games that favored Nvidia:

 

[Assimilator]

This video shows how many sites failed to use DX12 and Vulkan in their testing, and used games that favored Nvidia:

Sites aren't using DX12 and Vulkan because there are literally half a dozen games that use those APIs. In other words, DX12 and Vulkan are insignificant at this point in time, and probably will continue to be so for the next year or more. (I note that you fail to mention 3DMark TimeSpy, which coincidentally happens to show that Pascal does just fine in DX12/async. I guess NVIDIA must've bribed FutureMark too, it seems like they're bribing every man and their dog.)

And I guess RotTR DX12 and AotS are now "favouring NVIDIA" because they no longer show performance advantages for AMD cards. I love how AMD fans are happy to tout the benefits of DX12 games when they show AMD cards winning, but then immediately switch sides and claim the same games are biased or irrelevant when NVIDIA wins. Cognitive dissonance anyone?

 

[okidna]

That is the problem.
The latest reviews from many popular tech sites, including this one; have skewed their tests in favor of Team Green to the point where one must question their integrity.
This video shows how many sites failed to use DX12 and Vulkan in their testing, and used games that favored Nvidia:

He made a good recap and good point (many sites missing DX12 games and a Vulkan game) on the video, but one thing that hit me when watching the video is when he said that he "don't really understand many of the game choice in the benchmark suit" when the video shows Tom's HW result.
Putting aside Tom's HW review quality I think I know what game choice that he "don't understand". Yes, the infamous Project CARS, nobody hates and complaint on PCARS more than AMD and its users

 

[Ungari]

Sites aren't using DX12 and Vulkan because there are literally half a dozen games that use those APIs.

This does not mean that these APIs should be ignored from reviews, and because this is the future of gaming consumers need to know these facts in order to make an informed purchase.
Any reviewer that does not test for DX12 and Vulkan is negligent, and it looks like they are skewing the conclusions toward Pascal.
@okidna
Project Cars is an objectionable choice because everyone knows that game engine just does not work at all with AMD, and so it makes Nvidia cards look uber. I wont even get into the allegations of Nvidia paying that developer to hamstring AMD cards.