[SA] News of Nvidia’s Pascal tapeout and silicon is important

[Silverforce11]

Fun to watch people attack the messenger, instead of actually discussing the message.

According to some, CEO show casing fake products is no issue or no need for concern. They can simply do no wrong.

 

[ShintaiDK]

Fun to watch people attack the messenger, instead of actually discussing the message.

The message is homebrew rubbish.

 

[NTMBK]

EDIT: Actually, forget about it.

 

[dark zero]

Pascal could be another fermi. Fermi was their big compute arch wasn't it? before kepler when they started cutting things. They've been talking about going big on compute again, and to me they would lose their energy efficiency advantage and possibly run into problems that AMD met head-on years ago with GCN.

HBM issues is just more on top of building a good compute GPU.

Could be a disaster. Could be good, could be just ok.

Actually both need to be successful in order to avoid a big problem in the market and the additional risk of Intel attacking the GPU market

 

[96Firebird]

Fun to watch people attack the messenger, instead of actually discussing the message.

My how things change...

 

[maddie]

Faith based thinking. It's not what is said, but who says it that's important.

State of the world.

 

[S.H.O.D.A.N.]

The message is homebrew rubbish.

I have yet to see anyone dispute any of Demerjian's claims regarding the Zauba shipping manifest and what the gear listed there is meant to do.

 

[Mondozei]

That article is one long fundraising pitch for Charlie's site. It's very easy to look smart when you wait until the last moment when all the facts are in and then claim you have superior knowledge or smarts. Well, you don't. You just waited until everyone had their turn and have the luxury of the viewings things in the rearview mirror. It's easy to look smart then.

There's no reason to think Charlie has any more technical knowledge than anyone else, he simply has the common sense to wait instead of pushing for rumormongering on incomplete information for more clickbait.

That makes him more responsible, but it says nothing about his technical ability. Ultimately his endless and resentful appeals to subscribe to his site (narrowly disguised as tech site journalism "critique") is what is most laughable.

The reality is that nobody quite knows the state of Pascal but the point that people shouldn't jump to conclusions is something we all accept, yet the reason why most of the tech press jump to conclusions on the faintest of evidence is because they know most readers out there will swallow the bait. Let's not pretend most people in this forum are different. Just look at the endless Pascal threads based on incomplete information. That's fine. That's what forums are for, speculation and discusssion.

 

[NTMBK]

My how things change...

Not one of my finer posts, I admit.

 

[railven]

Actually both need to be successful in order to avoid a big problem in the market and the additional risk of Intel attacking the GPU market

I say bring it on. We need a third wheel in this two player game!

 

[Kenmitch]

The message is homebrew rubbish.

Why?

Ignore who said it and analyze the message and pick it apart then.

NVidia will struggle out of the gate as they have in the past when going to a new node with a new gpu technology....How long and what sacrifices they'll make are the variables.

The past is always used to predict AMD's future while it's shrugged away when convenient.

 

[stahlhart]

Fun to watch people attack the messenger, instead of actually discussing the message.

Tell me about it.
-- stahlhart

 

[xthetenth]

Fun to watch people attack the messenger, instead of actually discussing the message.

It's standard Charlie reading procedure. You know what he's going to conclude, you know what the people who disagree with him are going to conclude. What matters and can be informative is how much each side stays on facts.

Even if we remove the net conclusion Charlie came to since we all know his bias, there's still some unsettling facts presented there. He's dead on about the tools, there's no disputing that. The faked proto at CES is also a blunder, though not quite wood screw levels this time. It doesn't mean that we should assume AMD is months ahead. That said, I'd be willing to wager that AMD will be first out the gate in the mid and high range sections of the market if only because they've demonstrated functioning hardware already. Will it be better faster, more efficient or more successful? We won't know until both sides have played their hands.

This is a good point, and there's one thing worth pointing out. Those tools are for that particular chip, not for every pascal part. That does not necessarily mean that other chips aren't already being set up. There is room for goofy scenarios like big pascal and a small pipe cleaner being first priority with that being a lagging mid sized chip, but I personally doubt big pascal will reach yields that allow consumer pricing for months, and I think it will come out a half generation late because of it.

Overall though the really contentious point that's missing from the article is any backing to the assertion that a pascal chip's test gear arriving means that no pascal chip is being tested yet. That will be interesting.

Why didnt AMD allow the journalist to take photos of the die?

That is a good question. However, with journalists having seen a chip, and more importantly with a demonstration that they have something that can hit a certain performance/power level, I think Polaris is a much likelier bet.

I think that's one and a half legit points to use against the article and overall pro-AMD interpretation of current events. I personally doubt them, but at least they might lead to interesting discussion.

 

[Gikaseixas]

The past is always used to predict AMD's future while it's shrugged away when convenient.

So much truth in a single statement

Anyways, i really think that Nvidia will launch their next generation GPU's after AMD but that will not be a problem because many will just wait for NV and decide what offers are on the table. I will wait unless Polaris offers 100% improvement at launch

 

[railven]

Why?

Ignore who said it and analyze the message and pick it apart then.

NVidia will struggle out of the gate as they have in the past when going to a new node with a new gpu technology....How long and what sacrifices they'll make are the variables.

The past is always used to predict AMD's future while it's shrugged away when convenient.

If NV has a history of struggling to get a on a new node, then I say let them struggle. If that struggle leads to the products they put out.

 

[PPB]

If NV has a history of struggling to get a on a new node, then I say let them struggle. If that struggle leads to the products they put out.

Yeah, it has nothing to do with people's mindshare. The most sold Maxwell v2 GPU is the one with the xbar/IMC crippled and 3.5GB fast/0.5GB slow memory.

Such good products to wait for.

 

[96Firebird]

Mine is doing great! :thumbsup:

 

[xthetenth]

Mine was fantastic. I managed to make a profit flipping it for a similar card that's steadily picking up performance relative to it.

Seriously though, it is actually a good card. It just isn't reassuring from a longevity perspective and a lot of people passed over a better way to turn dollars into framerate.

 

[PPB]

Mine is doing great! :thumbsup:

Saving this post for late 2016 Gameworks Games/DX12 games

 

[96Firebird]

Can't wait!

 

[bystander36]

According to some, CEO show casing fake products is no issue or no need for concern. They can simply do no wrong.

If they are just showing a product for the sake of marketing of hype, who the cares. You can do that job with a fake product. It's only a problem if he's trying to show what it can do, in which case he either has a real one in the test machine (no problem there), or he is faking results (that's a problem).

 

[Azix]

I say bring it on. We need a third wheel in this two player game!

rather it not be intel. They can buy their way to the top as they did with CPUs even if they made inferior GPUs

 

[railven]

rather it not be intel. They can buy their way to the top as they did with CPUs even if they made inferior GPUs

If that were true we'd be using Larrabee's.

I'd rather Intel buy AMD's ATI patents and put NV in the ground.

 

[moonbogg]

I'd like an intel GPU. That would be badass. Imagine a cool looking heat sink with a skull on it or something. That would be neato.

 

[Zanovar]

I say bring it on. We need a third wheel in this two player game!

So much this.