The GT300/Fermi Thread

[Sam_oslo]

$700 a supercomputer does not make.

It's application specific and the main reason you can only find one nvidia supercomputer in the supercomputer TOP-100. the "GPU supercomputer" is still a non-argument.

The new articulate of GF100 is aiming to dominate and change this landscape in 2010.
GPGPU is a very new technology, and has been in child-stage until now. It is really surprising to have 1 supercomputer in TOP-100 already based on the old technology. Stay tuned, everything indicating that you are going to get surprised soon.

[Olivon]

nVidia GF100 Fermi silicon cost analysis - BSN*

[saaya]

nVidia GF100 Fermi silicon cost analysis - BSN*

cypress=100$? that would mean its @ 33% yields? 0_o
and first he says gt300 is @ 40% yields right now, then he says quadro and tesla will subsidize geforce until nvidia can reach 40% yields with gt300... huh? :P

interesting read, but a bit weird... and i dont trust the numbers :P

[Olivon]

interesting read, but a bit weird... and i dont trust the numbers :P

Right, difficult to know ... B1 rev. will be the retail version ?

[neliz]

The new articulate of GF100 is aiming to dominate and change this landscape in 2010.

Hitler aimed to rule the world.. just saying...

GPGPU is a very new technology, and has been in child-stage until now. It is really surprising to have 1 supercomputer in TOP-100 already based on the old technology.

That's just as much as "ATI" has.. and their system is considered beyond poor

Stay tuned, everything indicating that you are going to get surprised soon.

You cannot surprise someone that knew more about Fermi back in July then you do now. You sound awfully optimistic.. too optimistic.

hmmm neliz probably meant card MP then..

No I didn't.. the investors "mass production" refers to production RAMP, they are still quite a few weeks off from real mass production.

[Sam_oslo]

..
You cannot surprise someone that knew more about Fermi back in July then you do now.

...

You didn't know nothing back then, and you still don't know anything. Nobody knows for sure yet, because the retail product (or any of details) is not out yet.

A couple of other kids have been speculating with this kind of childish arguments since July.

[saaya]

No I didn't.. the investors "mass production" refers to production RAMP, they are still quite a few weeks off from real mass production.

ahhh so he WAS twisting words again
i thought there wasnt much you could interpret when saying MP has started... but i guess you can always claim your doing MP in a special way...
so MP starts in your company means you set up a row of meetings for the next weeks at which you will discuss how and when to start MP

You didn't know nothing back then, and you still don't know anything. Nobody knows for sure yet, because the retail product (or any of details) is not out yet.

A couple of other kids have been speculating with this kind of childish arguments since July.

take it easy, no need to get rancid

[neliz]

You didn't know nothing back then, and you still don't know anything. Nobody knows for sure yet, because the retail product (or any of details) is not out yet.

Hahahaha. So, in your little boxed up world something like Fermi just pops out of thin air and is there? It doesn't require years of work and nvidia doesn't update their partners at all? they just say "oh, here's our new product, go and have fun with it!"

So according to you there are no cards out at developers and partners and nv suddenly doesn't inform anyone.

I wonder how I leaked the Fermi name a couple of days before Fuad posted his first article about it prior to GTC. (though I must admit some people posted it before but were not quite sure. Theo's 24x24? saying that for months where "respected" websites and NV PR spinners have been saying it's smaller since last May. I hope they're not paying you to spread the Fermi word, because you're doing a pretty bad job.

The truth will come out. GF100 might look good now.. but let me tell you, AMD never planned on positioning HD5870 against Fermi.

A3 went into production back in Early December. What I'm hearing right now is that A3 will probably not be the final revision.
Much like GT200b1 was shown in May 08. Demo'ed with b2 in September, reviewed and sold in minor (tesla) quantities in b2 up to November and finally became available (many months after) in b3.

Sam_Oslo's reply here \/

[Sam_oslo]

Hahahaha. So, in your little boxed up world something like Fermi just pops out of thin air and is there? It doesn't require years of work and nvidia doesn't update their partners at all? they just say "oh, here's our new product, go and have fun with it!"

So according to you there are no cards out at developers and partners and nv suddenly doesn't inform anyone.

I wonder how I leaked the Fermi name a couple of days before Fuad posted his first article about it prior to GTC. (though I must admit some people posted it before but were not quite sure. Theo's 24x24? saying that for months where "respected" websites and NV PR spinners have been saying it's smaller since last May. I hope they're not paying you to spread the Fermi word, because you're doing a pretty bad job.

The truth will come out. GF100 might look good now.. but let me tell you, AMD never planned on positioning HD5870 against Fermi.

A3 went into production back in Early December. What I'm hearing right now is that A3 will probably not be the final revision.
Much like GT200b1 was shown in May 08. Demo'ed with b2 in September, reviewed and sold in minor (tesla) quantities in b2 up to November and finally became available (many months after) in b3.

Sam_Oslo's reply here \/

Are you indicating to have inside info?, or have contact with some partner?

Even if you have heard something about the the name and such, it doesn't mean that you should take off totally and get so sure that you are acting.

Nothing is that sure yet, and nobody can say anything for sure before benchmarking the retail GPU, but those early benchmarks are showing something else than your speculations based on some info that you have heard from somebody.

[zalbard]

Are you indicating to have inside info?, or have contact with some partner?

Maybe neliz = Jen-Hsun Huang? How can you be sure on ze interwebz?!

[h0bbes]

Even though i prefer ATI (9800 pro was my last desktop card:P) i really want Fermi to be successfull since i will buy a new system sometime soon and those prices have to go down + it might push ATI harder. Compared to what MOST people are saying/hypothesizing/guessing/hoping here, neliz seems to know his stuff better. Anyway as people have already said countless times, we can't be sure about anything until we get some more independent reviews and probably by that time i will stop caring!

[Tim]

Are you indicating to have inside info?, or have contact with some partner?

Even if you have heard something about the the name and such, it doesn't mean that you should take off totally and get so sure that you are acting.

Nothing is that sure yet, and nobody can say anything for sure before benchmarking the retail GPU, but those early benchmarks are showing something else than your speculations based on some info that you have heard from somebody.

You managed to dodge and meh everything neliz posted there, well done!

[Sam_oslo]

You managed to dodge and meh everything neliz posted there, well done!

Thx, but it was doomed to happen, anyways. Nobody can act as sure as him without proper documentation.

EDIT:
It is always fun to predicts the trends, and make some educated guess and assumption, but acting too sure about a unreleased product is not the proper way to act.

[saaya]

So according to you there are no cards out at developers and partners and nv suddenly doesn't inform anyone.

actually they (AIBs) still dont have any cards at all i heard

You need to focus man
AI enabled = botmatch ( like somebody is playing the game and does all the ai calculations )
AI disabled = flyby

heh, was way late last night

Maybe neliz = Jen-Hsun Huang? How can you be sure on ze interwebz?!

Thx, but it was doomed to happen, anyways. Nobody can act as sure as him without proper documentation.

EDIT:
It is always fun to predicts the trends, and make some educated guess and assumption, but acting too sure about a unreleased product is not the proper way to act.

let me put it this way.. transistor per transistor nvidia has higher efficiency/performance than ati whether you agree or not like it or not.. you guys can turn blue it aint gonna change

depends at what perf segment you look at... for some reason cypress doesnt seem to scale very well, especially if you look at the 5970...

I only care because none of them offer what I want. (That being a decent >1GB card).

My only hope is Ati "canned" the "six" 2GB HD5870 to make something new and shiny, like say perhaps a 2GB (across the range) 5890 which is clocked @ 1Ghz and has faster memory and a few tweaks here and there...

And that Fermi has 1.5GB (at least) with some versions/editions offering 3GB!
John

beats me why no single AIB has launched a 2GB 5870

Just curious, couldn't you just buy one of those 5870 "six" (when they come out) and overclock it a bit?

1. its not even released, let alone widely available
2. i dont think he gives a damn about 6 monitor support, so why should he pay extra for that? on the contrary, you end up with mini dp connectors so you need an adapter for every display you hook up to it...

20%...

fully functional? thats 250$ a piece... thats just about good enough to not lose money on a 500$ card isnt it?

You may be able to verify some of his info, but he doesn't even allow others to be optimistic, while he himself acts to know everything in such a details that ha can't get surprised . All this without showing any benchmarks or documentation about his claims. He says::

I personally require benchmarks and documentation for this kind of bold claims, otherwise i don't take it serious. You can consider it as good info if you want.

so when you go out on a date you ask her to prove that shes really single and actually female before talking to her, right?

geez, just take it for what it is... take it or leave it...

I'm just warning people here that think that GF100 will be [strike]God's[/strike]nVidia's gift to man. If it was, it would've been last year.. not anymore. We're talking about an expensive, marginal (10-30%) upgrade over HD5870

cheers

if price wouldnt be an issue, i think most people would def go with a fermi instead of a 5970... even if its slower... i hope ati has a faster single card up their sleeves...

[FischOderAal]

if price wouldnt be an issue, i think most people would def go with a fermi instead of a 5970... even if its slower... i hope ati has a faster single card up their sleeves...

Of course and I know why: CF and SLI plain SUCKS! There are quite a few people that buy the fastest gfx with one GPU on the market, no matter the costs. You're much less dependant on drivers and bugfixes compared to dual-gpu-solutions.

[zerazax]

You didn't know nothing back then, and you still don't know anything. Nobody knows for sure yet, because the retail product (or any of details) is not out yet.

A couple of other kids have been speculating with this kind of childish arguments since July.

No actually he was spot on with everything about RV870 back then, and broke plenty before others on Fermi