DX11 Market Steam Hardware Survey - AMD maintains 80% of the DX11 Market

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

GodisanAtheist

Diamond Member
Nov 16, 2006
7,138
7,632
136
ATI's 5 series were fantastic cards and did a wonderful job bringing high performance to DX9 and DX10 games 6 months before NV had anything to say about it, and 9 months before NV had an affordable alternative.

Hell I own[ed, see sig] 2 5770's and I loved them, fantastic cards.

That being said, you have to be blind or infused with a certain special kind of fanboy rage to even attempt to defend ATI's tessellation performance. ATI's fixed function tessellator was probably a big part of the reason the 5 series launched on time (it was old hat by then) but now its really demonstrated to be the weakest point in their DX11 arch. There is really no other way to slice it. Does it provide playable framerates? Yeah sure. Does it provide the same framerate the competitor does? NO. And that's all that matters. Quit whining about it and accept it.

It looks like the 6900 series spec sheets were saying something about twin tesselators, and seeing how a single tessellation engine provides roughly half the competition's frames, two should (roughly) bring both ATI & Nvidia to parity, no?

Then everyone wins.

+1 GTX460, sexy little card it is.
 

Borealis7

Platinum Member
Oct 19, 2006
2,901
205
106
its not a MARKET trend, its STEAM trend. how many % of people in the world who buy video cards use steam? and allow reporting of this information? i bet its single digits.

(obviously i don't use steam, but i assume it asks you before adding your info to the database.)
 

toyota

Lifer
Apr 15, 2001
12,957
1
0
its not a MARKET trend, its STEAM trend. how many % of people in the world who buy video cards use steam? and allow reporting of this information? i bet its single digits.

(obviously i don't use steam, but i assume it asks you before adding your info to the database.)
but how could a pc gamer not own at least one title thats on Steam?
 

toyota

Lifer
Apr 15, 2001
12,957
1
0
very very few titles are steam exclusives. if you don't like valve games then you can buy anything else on steam elsewhere.
I just cant believe an avid pc gamer would not have any games like HL 2, TF 2, Portal, L4D or others. and some games such as Metro 2033, FEAR 2, Mafia 2 and Just Cause 2 require Steam even if they are not made by Valve.
 

Skurge

Diamond Member
Aug 17, 2009
5,195
1
71
I just cant believe an avid pc gamer would not have any games like HL 2, TF 2, Portal, L4D or others. and some games such as Metro 2033, FEAR 2, Mafia 2 and Just Cause 2 require Steam even if they are not made by Valve.

You do have a point, not to mention stuff like CODMW2 and BLOPs.
 

Lonbjerg

Diamond Member
Dec 6, 2009
4,419
0
0
its not a MARKET trend, its STEAM trend. how many % of people in the world who buy video cards use steam? and allow reporting of this information? i bet its single digits.

(obviously i don't use steam, but i assume it asks you before adding your info to the database.)

Every single survey out there would love to have a base of +100.000 participants.

You know how many people are asked in a eg. Gallup poll?
~1.000 people.

Read up on polls, sample sizes and accuracy:

To be sure, there is some gain in sampling accuracy which comes from increasing sample sizes. Common sense - and sampling theory - tell us that a sample of 1,000 people probably is going to be more accurate than a sample of 20. Surprisingly, however, once the survey sample gets to a size of 500, 600, 700 or more, there are fewer and fewer accuracy gains which come from increasing the sample size. Gallup and other major organizations use sample sizes of between 1,000 and 1,500 because they provide a solid balance of accuracy against the increased economic cost of larger and larger samples. If Gallup were to - quite expensively - use a sample of 4,000 randomly selected adults each time it did its poll, the increase in accuracy over and beyond a well-done sample of 1,000 would be minimal, and generally speaking, would not justify the increase in cost.
 

Lonyo

Lifer
Aug 10, 2002
21,938
6
81
Every single survey out there would love to have a base of +100.000 participants.

You know how many people are asked in a eg. Gallup poll?
~1.000 people.

Read up on polls, sample sizes and accuracy:

Read up on sampling methods.

http://en.wikipedia.org/wiki/Sampling_(statistics)#Sampling_methods

This might help.


If I go to Texas and as people about their political preferences, that's not going to tell me anything about US political attitudes. Even if I ask every single Texan.
Equally, asking Steam users about their hardware isn't going to tell me anything about the overall computer market.
 

Lonbjerg

Diamond Member
Dec 6, 2009
4,419
0
0
Read up on sampling methods.

http://en.wikipedia.org/wiki/Sampling_(statistics)#Sampling_methods

This might help.


If I go to Texas and as people about their political preferences, that's not going to tell me anything about US political attitudes. Even if I ask every single Texan.
Equally, asking Steam users about their hardware isn't going to tell me anything about the overall computer market.

Steam is made up of gamers.
You find me other people buying a GTX580 or a 6970 than gamers?
Perhaps you sould head you own words?
 

Lonyo

Lifer
Aug 10, 2002
21,938
6
81
Steam is made up of gamers.
You find me other people buying a GTX580 or a 6970 than gamers?
Perhaps you sould head you own words?

Yes, it's a selection of gamers, therefore it's not the whole market.
The poster you quoted said that Steam trends didn't show market trends. It doesn't, because the market is more than just gamers.
Are you that dense?

Someone: "Steam shows Steam trends not market trends"
You: "Learn statistics man"
Me: "Steam isn't representative"
You: "Learn statistics because Steam shows gamers".

Gamers still aren't the entire market, no matter how many times you tell people to learn about statistics.
 

Lonbjerg

Diamond Member
Dec 6, 2009
4,419
0
0
Yes, it's a selection of gamers, therefore it's not the whole market.
The poster you quoted said that Steam trends didn't show market trends. It doesn't, because the market is more than just gamers.
Are you that dense?

Someone: "Steam shows Steam trends not market trends"
You: "Learn statistics man"
Me: "Steam isn't representative"
You: "Learn statistics because Steam shows gamers".

Gamers still aren't the entire market, no matter how many times you tell people to learn about statistics.

So you couldn't think of any none-gamers that is buying a GTX580...gotcha.
 

Lonyo

Lifer
Aug 10, 2002
21,938
6
81
I just cant believe an avid pc gamer would not have any games like HL 2, TF 2, Portal, L4D or others. and some games such as Metro 2033, FEAR 2, Mafia 2 and Just Cause 2 require Steam even if they are not made by Valve.

I own a few Steam games.
I haven't added my hardware to the Steam survey since maybe February or something.
Just because you have it doesn't mean you add your stuff.
I also haven't used Steam more than about 3 times in the last 6 or so months.
 

Lonbjerg

Diamond Member
Dec 6, 2009
4,419
0
0
Dude, what are you even trying to say?

Read my post, it states quite cleary what I am typing.
You find any none-gamers buying GTX580 in larger quantities, then you have a point..otherwise you don't.
 

Lonbjerg

Diamond Member
Dec 6, 2009
4,419
0
0
I own a few Steam games.
I haven't added my hardware to the Steam survey since maybe February or something.
Just because you have it doesn't mean you add your stuff.
I also haven't used Steam more than about 3 times in the last 6 or so months.

That is what one would would call "statistically insignificant"...so what is your point?
 

Lonyo

Lifer
Aug 10, 2002
21,938
6
81
Read my post, it states quite cleary what I am typing.
You find any none-gamers buying GTX580 in larger quantities, then you have a point..otherwise you don't.

Your post is unrelated to mine. I never said anything about GTX580s.
I said that Steam isn't representative of the overall graphics card market. It still isn't.
Just because it provides a sample of the gamer market still doesn't make it representative of the gaming market.
One reason for this is that not everyone has Steam games, or reports their hardware for the survey.
Add to that the fact that many NV games come bundled with Mafia 2, which as noted above requires Steam, and you are more likely, based on that, to have more NV users having Steam accounts. That means it's even less representative at the higher end because the sample is quite likely to be skewed.

Steam is not representative of the market. It's probably not representative of gamers either, especially at the high end where the sample size is small, and there are factors which will likely impact the chance of someone being registered on Steam.

It might be a useful tool for getting an idea of the way that gaming hardware is going, when taken as a whole, but it can't be used to say "this is what the market is, this is how well things are selling, this product is successful, these are the marketshares of each company".
 

MangoX

Senior member
Feb 13, 2001
594
108
106

I was overreacting a bit. I have neither bias towards any GPU company (my last 4 cards were 5870, GTX295, 8800GTS, X800XL btw) but those very recent AMD IQ accusations by NV left a bad taste in my mouth. But really I will still buy Nvidia, as long as their products are good.
 

Lonbjerg

Diamond Member
Dec 6, 2009
4,419
0
0
Your post is unrelated to mine. I never said anything about GTX580s.
I said that Steam isn't representative of the overall graphics card market. It still isn't.
Just because it provides a sample of the gamer market still doesn't make it representative of the gaming market.
One reason for this is that not everyone has Steam games, or reports their hardware for the survey.
Add to that the fact that many NV games come bundled with Mafia 2, which as noted above requires Steam, and you are more likely, based on that, to have more NV users having Steam accounts. That means it's even less representative at the higher end because the sample is quite likely to be skewed.

Steam is not representative of the market. It's probably not representative of gamers either, especially at the high end where the sample size is small, and there are factors which will likely impact the chance of someone being registered on Steam.

It might be a useful tool for getting an idea of the way that gaming hardware is going, when taken as a whole, but it can't be used to say "this is what the market is, this is how well things are selling, this product is successful, these are the marketshares of each company".

We are not talking about the entire graphics market.
If we where the numbers would be something like 50% Intel (IGP's).

But we are talking about gaming GPU's
Bought by gamers.
DX11 GPU's are irrelanvt to all others than gamers.
And the large sample size makes it darn accurate.
A point you still avoid to answer..says it all dosn't it?
 

Grooveriding

Diamond Member
Dec 25, 2008
9,110
1,260
126
Dude, what are you even trying to say?

He doesn't seem to know. :eek:

Ever since his posts in a physx thread claiming he was running Mafia 2 on high with high physx on a GTX 285 at 1920x1200 with smooth framerates and then another thread were he was praising the almighty tessellator of Fermi and how we would all be going forward with such technology but he doesn't even own a DX11 card himself.

After those two blunders I started writing off his posts/opinions for what they clearly are.
 

SlowSpyder

Lifer
Jan 12, 2005
17,305
1,001
126
ATI's 5 series were fantastic cards and did a wonderful job bringing high performance to DX9 and DX10 games 6 months before NV had anything to say about it, and 9 months before NV had an affordable alternative.

Hell I own[ed, see sig] 2 5770's and I loved them, fantastic cards.

That being said, you have to be blind or infused with a certain special kind of fanboy rage to even attempt to defend ATI's tessellation performance. ATI's fixed function tessellator was probably a big part of the reason the 5 series launched on time (it was old hat by then) but now its really demonstrated to be the weakest point in their DX11 arch. There is really no other way to slice it. Does it provide playable framerates? Yeah sure. Does it provide the same framerate the competitor does? NO. And that's all that matters. Quit whining about it and accept it.

It looks like the 6900 series spec sheets were saying something about twin tesselators, and seeing how a single tessellation engine provides roughly half the competition's frames, two should (roughly) bring both ATI & Nvidia to parity, no?

Then everyone wins.

+1 GTX460, sexy little card it is.


I am neither blind nor am I infused with a certain kind of fanboy rage while I make defending statements about AMD's tessellator. There is no denying that it is weaker than Nvidia's part in tessellation. My arguement is that AMD's is still 'good enough' for right now and probably good enough until this generation of cards, including current GeForce parts, won't be what we want to game on anyway. There will be other limiting factors in future games as well.