[steam] Hardware survey now provides more info

[Lonbjerg]

If you base your posts on camshots of obsolete roadmaps, I can understand why at times you post in conflict with reality:

Now that has to be the funniest response/spin I've seen in a long time. You can't be serious. This was from Nvidia themselves. So, it doesn't matter as long as they update the roadmap whenever they fall behind.

So T10 launched 2007 and Fermi 2009?

Instead of your explanation, could it be that people using these outdated camshots really don't know what the shoot represents?

Do you tell Anand that he got the year wrong here and here then?:
T10 2008 (you say it should be 2007)
http://www.anandtech.com/show/2549/21
Fermi 2010 (You say it should be 2009)
http://www.anandtech.com/show/2977/...d-gtx-470-6-months-late-was-it-worth-the-wait

Unless you think NVIDIA voilated some temporal "law"?

 

[Lonbjerg]

:thumbsdown:

So T10 launched at 2007 and Fermi at 2009?

It's not a hard question...unless you really, really don't want to reply it :sneaky:

Wich roadmap has the correct dates?

 

[BD2003]

Lol tough crowd in here.

 

[Grooveriding]

snip

Planned launch of Nvidia 28nm and 22/20nm GPUs moved back to 2012 and 2014

Despite Nvidia CEO Huang Jen-hsun previously saying that the company is set to announce its new 28nm GPU architecture at the end of 2011 and 22/20nm in 2013, sources from graphics card makers have pointed out that Nvidia has already adjusted its roadmap and delayed 28nm Kepler and 22/20nm Maxwell to 2012 and 2014.

Everything you've posted amounts to 'no, no it isn't true' and throwing your hands over your ears. If you have some facts, and your need to believe against reality is not a fact, please offer them up.

 

[Lonbjerg]

So according to Groove & Co...T10 launched at 2007 and Fermi at 2009, NVIDIA dosn't know what it's doing...and it got nothing to with TMSC, AMD delaying HD 7950...it's all NVIDIA fumbeling...thanks for the ride

 

[Will Robinson]

Did they say that?...quotes please:sneaky:
LOL at the "tough crowd" comment(BD2003)...WELCOME TO HELL:twisted::biggrin:

 

[tviceman]

This thread has been completely derailed.

 

[Arzachel]

So according to Groove & Co...T10 launched at 2007 and Fermi at 2009, NVIDIA dosn't know what it's doing...and it got nothing to with TMSC, AMD delaying HD 7950...it's all NVIDIA fumbeling...thanks for the ride

Umm, you might have misunderstood the point here.

Just like you said, T10 launched on 2008, Fermi did on 2010. Nvidia used that roadmap after they launched Fermi, when you can clearly see that it's false. They only fixed that after the "Kepler is late" statement.

 

[Rifter]

I dont really see whats so hard to understand. Since Fermi it has been well known Nvidia is following a 2 year release cycle. Fermi was in 2010, Kepler will be in 2012, why the hell is this so hard to understand for some people?

 

[blackened23]

Who use steam.

Is this a joke? Everyone that plays PC games uses steam almost, unless you're stuck in 2005 and playing terribly old games.

 

[SirPauly]

I dont really see whats so hard to understand. Since Fermi it has been well known Nvidia is following a 2 year release cycle. Fermi was in 2010, Kepler will be in 2012, why the hell is this so hard to understand for some people?

I don't think it is a big deal but the thought process may be because the Fermi delay wouldn't effect Kepler.

 

[Lonbjerg]

I dont really see whats so hard to understand. Since Fermi it has been well known Nvidia is following a 2 year release cycle. Fermi was in 2010, Kepler will be in 2012, why the hell is this so hard to understand for some people?

Because it soiunds better to say "NVIDIA is late!!!...AGAIN!!!`....OMFG!!!!"

Than to read up on the boring facts and stop using a blurry camshoot that dosn't align with reality and just write:

"NVIDIA is planning a Q2 2012 launch for Kepler"

The old roadmap had wrong dates(read years) on ALL archs...now it's correct...and some people seem to think that it is bad...or something.

 

[Grooveriding]

I dont really see whats so hard to understand. Since Fermi it has been well known Nvidia is following a 2 year release cycle. Fermi was in 2010, Kepler will be in 2012, why the hell is this so hard to understand for some people?

The argument has nothing to do with the 2 year release cycle. We're well aware of it, as seen here in this slide from GTC 2009 where Fermi was promised for later that year and where the infamous fake fermi with woodscrews was held up:

Nvidia planned fermi for 2009, kepler for 2011, maxwell for 2013. Yes, a two year release cycle. Fermi was delayed into mid-2010 and nvidia said the delay would not cause delays to their future products - so Kepler should of been 2011 and maxwell 2013 as stated originally. But:

Planned launch of Nvidia 28nm and 22/20nm GPUs moved back to 2012 and 2014

Despite Nvidia CEO Huang Jen-hsun previously saying that the company is set to announce its new 28nm GPU architecture at the end of 2011 and 22/20nm in 2013, sources from graphics card makers have pointed out that Nvidia has already adjusted its roadmap and delayed 28nm Kepler and 22/20nm Maxwell to 2012 and 2014.

So the roadmap was adjusted for the kepler and maxwell delays. Fermi showing as 2010 is nothing more than the fact that the new roadmap is a result of Fermi being delayed and coming after the delay took place.


Nvidia has even said themselves there was a delay, if there was no delay, kepler would of been last year and maxwell next year; a 2 year release cycle. I agree, it's not very hard to understand. Lonbjerg was just posting falsehoods and needed to be corrected as he didnt seem to understand.

 

[SirPauly]

It isn't the first time or last time that road-maps may change. The nature of the beast, one may imagine.

 

[Lonbjerg]

Oh, I see we went from T10 in 2007 and Fermi in 2009 to woodscrews...funny (not really)

But still no substance...is NVIDIA to blame for the delay of 7950 too?

*smiles and looks at TMCS 40 and 28 nm processes*

 

[Arzachel]

Oh, I see we went from T10 in 2007 and Fermi in 2009 to woodscrews...funny (not really)

But still no substance...is NVIDIA to blame for the delay of 7950 too?

*smiles and looks at TMCS 40 and 28 nm processes*

You know that you can't really shift all the blame on TSMC, when AMD use the same fabs, right?

 

[MrK6]

You know that you can't really shift all the blame on TSMC, when AMD use the same fabs, right?

Logic is straight poison to fanboys, do you want to clean up the body?

 

[SHAQ]

What do 80% of these posts have to do with Steam's hardware survey?

 

[Lonbjerg]

You know that you can't really shift all the blame on TSMC, when AMD use the same fabs, right?

That is why they are currently flooding the market with GPU's? :sneaky:
That is not why they delayed the 7950? :sneaky:

The logic is fuzzy here...eh?

On Steam and it's more detail.
All for it.
Steam is (like it or not) a better tool than pure guesses.
The more datapoints the better :thumbsup:

 

[Spikesoldier]

so did they fix the CF reporting now? or still bugged?

 

[SirPauly]

On Steam and it's more detail.
All for it.
Steam is (like it or not) a better tool than pure guesses.
The more datapoints the better :thumbsup:

I agree with this. The more data points, the better. Sure doesn't hurt discussions and think it is fun and welcomed.

 

[exar333]

Keep in mind that NV is only launching low-power kepler initially as well. Fermi launched with the 480 GTX, not a die-shrunk 285.

Also, just because '28nm works for one vendor' doesn't mean it works the same for others. Problems may or may not be shared; same with successes. To think fabbing one chip for AMD is the same as fabbing a different chip for NV is ignorant, at best.

 

[SirPauly]

I have no idea what nVidia is going to launch -- read rumors and speculation though.

 

[tviceman]

Keep in mind that NV is only launching low-power kepler initially as well.

Rumors are just rumors. Nvidia has made no comment on what GPU's they are launching first or when exactly they are launching. Everyone here who is saying 6 months from the hd7970's launch or low end Kepler's first are all just stating other rumors which have no verifiable sources.

This thread is still completely off topic.

 

[SirPauly]

I reread this and enjoyed it:

On the specific DX11 GPU side, the 5770 still reigns supreme as the top DX11 graphics card, coming in at a whopping 14%. The gtx460 (all models) trails a distant second with 10.7%, and the gtx560 (all models), with strong momentum, has surpassed the hd5800 series cards to come in at third with 8.81%..

Enjoyed the words. It was like it was a competitive race, like horse racing!