Intel's Raja Koduri Refutes Rumors About Company Cancelling Arc Graphics

[lexluthermiester]

Intel's accelerated computing group head Raja Koduri, who heads the team behind the Arc "Alchemist" graphics, on late-Sunday, refuted rumors about the company shutting down the Arc graphics product line.

Put another way, these rumors are stupidly without merit. Only people who use their head for a seat cushion would believe them.

 

[Daven]

Exactly. Why would anyone in their right mind cancel a multimillion USD project before fruition?

Edit: I love the picture with the loosely connected power plugs!

You’re right. They won’t pull out after spending millions. They will pull out after spending billions! Just look at Itanium. But at least Itanium had a good run for awhile. Not so sure about these GPUs.

 

[GunShot]

I mean... when you have a site like this one that does not even fact-check their sources BEFORE flying off the hip with obvious hatred and refuse to redacted its piece days later, though:

Intel Arc GPU effectively cancelled: 'decision has been made'

Read more: https://www.tweaktown.com/news/8839...y-cancelled-decision-has-been-made/index.html

Intel might need to send to these clowns A C&D doublet.

 

[Nanochip]

Persistence is key. I hope they persist, tune the drivers, and release battlemage on time.

 

[zlobby]

Persistence is key. I hope they persist, tune the drivers, and release battlemage on time.

Hoping is coping. Even though I despise intel I too would like them to create more pressure on those lazy AMD and nvidia.

 

[lexluthermiester]

You’re right. They won’t pull out after spending millions. They will pull out after spending billions! Just look at Itanium. But at least Itanium had a good run for awhile. Not so sure about these GPUs.

As you said, Itanium actually made money. Not a lot, but it did break-even after a few years. Your point isn't one as you contradicted yourself.

 

[Bwaze]

I think it's still possible for Intel to "cancel" most of the discrete ARC lineup and continue to develop the next generation. And release just a limited amount of this more or less failed generation cards, just so they can tell the investors it went "according to plan".

 

[ModEl4]

Hoping is coping. Even though I despise intel I too would like them to create more pressure on those lazy AMD and nvidia.

greedy not lazy, I'm sure they are busy as a bee trying to maximize their profit!

 

[SOAREVERSOR]

Rumors are just that, rumors!

 

[Daven]

As you said, Itanium actually made money. Not a lot, but it did break-even after a few years. Your point isn't one as you contradicted yourself.

Spending on development and making money on said development are two separate but related things. I just commented on the former and made no comment on the later. You cannot contradict yourself when only talking about one side of the equation.

Edit: withdrawn. I did comment on Itanium having a good run.

 

[Space Lynx]

if Intel cancels ARC them might as well cancel Intel alltogether

they have no logical reason to cancel ARC, they just got a 20-30 billion dollar blank check from tax payers this summer. they can play around a little if they like and see where it goes.

must be nice living in a corporate welfare state, that prides itself at night before bed about being about the free markets, while simultaneously denying the welfare state for the core things humans need to be affordable: transportation, food, healthcare, education.

 

[Bones]

Intel has been profitable for decades. Surely it can't be a financial decision to cancel ARC. It will take a couple of more years of experience to succeed and stabilize along Nvidia/AMD. But executives and stock holders aren't in this for the longer run. It is always about short term profit.

You are both, right and wrong about that.
At least some investors will stay in to ride it out while others will be in it for quick gains as you put it.

Speaking of how you put it, was stated as an absolute amongst ALL owners of stock regardless but that's not correct at all, there are investors and then there are investors.

What I'm seeing is it depends on how things go.

The number of investors involved based on whether they are short or long term investors will have an effect too. If there is a majority of short investors that want to bail then it's trouble for sure, if there is a majority of long term investors they'll tend to (But not neccesarily) ride it out, however the REAL factor out of these two investment camps is....
Who has the most money.

Even a single investor, regardless of which side they are on (Short or long term) can sway things - Just ask Buffett about that.
He IS the type if something isn't profitable, he'll do away with it without any hesitation about it.

 

[TheoneandonlyMrK]

I mean... when you have a site like this one that does not even fact-check their sources BEFORE flying off the hip with obvious hatred and refuse to redacted its piece days later, though:



Intel might need to send to these clowns A C&D doublet.

No just a Arc GPU on sale and available for purchase.

 

[Haile Selassie]

Well, exiting other business means a lot of things. Intel is heavily invested in SoC, FPGA and other ASICs not used in personal computers.

 

[iO]

Yea, that means nothing.
Just look at Optane, denied rumours in March but then killed it in July.

And just last week Gelsinger himself talked about killing of other divisions while focusing on it's core business...

 

[Punkenjoy]

Even the worst case scenario, which is that Arc is a huge flop, and Intel only earns back a fraction of their investment is still better than cancelling the whole thing, not even having the chance to earn back anything.

Have you ever heard of the Sunken cost fallacy?

If they continue the program, they have to continue to spend en R&D, pay for mask at fabs, pay for the drivers teams to support all those sku, etc.

some of the cost can be shifted to the iGPU but not all. At some point it's not worth it if they don't plan to invest enough to compete.

I think Intel have 2 flaw here.
1). they wanted to become a market leader right away and went big from start and faced many issue upfront. I think they should have try at least to beat AMD iGPU to start with and start small and every gen, get a bit bigger while learning the process and improving the drivers. But intel can't stand to not be the top performer.

2. Intel just grew too fast and hired way too much. It take a lot of time to build a team and get good at doing something and adding too much people complicate things, It's much easier to add 1-2 guy at the time in a well running teams than trippling the teams with news guys from everywhere.

I still think that if intel cancel Arc, that is a mistake for them. They need that to stay relevant in the future.

 

[lexluthermiester]

I think it's still possible for Intel to "cancel" most of the discrete ARC lineup and continue to develop the next generation. And release just a limited amount of this more or less failed generation cards, just so they can tell the investors it went "according to plan".

That is not only a silly notion but makes no sense considering that Intel has already started ramping up stock production.

Yea, that means nothing.
Just look at Optane, denied rumours in March but then killed it in July.

And just last week Gelsinger himself talked about killing of other divisions while focusing on it's core business...

Meritless opinion.

 

[ZoneDymo]

Meritless opinion.

Meritless? They mention two examples of what Intel has done and said...those are facts.
And if their opinion is based on the facts they mentioned, its not meritless.

 

[lexluthermiester]

Meritless?

Completely.

I mentions to examples of what Intel has done and said...those are facts.
And if their opinion is based on the facts they mentioned its not meritless.

You're missing some context. Opinions based on flawed perceptions of fact does not make for factual information. It's called spin and only knuckle-draggers and tree-swinging poo-flingers will buy it..

 

[GunShot]

No just a Arc GPU on sale and available for purchase.

I'm not understanding your comment.

But, whatever it means, TT, CB, etc. did not need to take things personal, and lose their integrity at the same time (obviously, they did here for no damn apparent reason but only *feeling entitled* to Intel's products, first, but Intel had chosen Linus instead for its Arc exclusive revealing ) and activate an Intel smear campaign since the early part of the summer.

 

[Tropick]

That is not only a silly notion but makes no sense considering that Intel has already started ramping up stock production.


Meritless opinion.

Meritless opinion? Those were all things that actually happened. You feeling okay?

Intel would absolutely scale back a project if the board decided it wasn't going to be profitable enough.. Doesn't matter how great it would be to have a third competitor, if the shareholders aren't seeing dollar signs there's always going to be a chance the division will face intense scrutiny.

only knuckle-draggers and tree-swinging poo-flingers will buy it..

How eloquently put

 

[phanbuey]

If Raja is refuting it then it must be true.

 

[Dragokar]

Well, Larrabee 2 still might happen. They could ship the whole ARC lineup and some Battlemage GPUs and cut it off there for retail. Shift it to data center and computing, which I believe they do great in.

Xeon Phi aka Knights xxxx series wasn't that bad as a Larrabee offspring. The main focus is to bring Sapphire Rapids finally. All other things might shift in one way or the other, and it wouldn't be the first time to axe something before it really started.

Itanium also made money because HP paid for its life support.

 

[lexluthermiester]

Intel would absolutely scale back a project if the board decided it wasn't going to be profitable enough..

While that seems like reasonable logic, it is in fact meritless. The Intel board is not going to cancel a product line this far into the production process. If they were going to cancel it, they would have done so during the R&D phase like they did with Larrabee. They are well passed that phase and are in the production phase. The limited release of GPU's that seem to have hit the Chinese market were a test and seems to have sold through very well. The Intel board is happy with the progress of the ARC lineup. Any nonsense to the contrary is just jibber-jabber from and for nay-sayers.

How eloquently put

You like that?

Well, Larrabee 2 still might happen.

No, it won't.

 

[Tropick]

While that seems like reasonable logic, it is in fact meritless. The Intel board is not going to cancel a product line this far into the production process. If they were going to cancel it, they would have done so during the R&D phase like they did with Larrabee. They are well passed that phase and are in the production phase. The limited release of GPU's that seem to have hit the Chinese market were a test and seems to have sold through very well. The Intel board is happy with the progress of the ARC lineup. Any nonsense to the contrary is just jibber-jabber from and for nay-sayers.


You like that?

Whatever you say, man. Just don't be shocked if they decide to pull the consumer dGPU div. in favor of pursuing existing business lines. It's absolutely reasonable to worry about the continuance of consumer ARC when they've already proven they're capable of getting rid of established divisions (Optane), even if it means writing off over $500 million dollars of existing inventory. Gelsinger is trying (to his commendation) to restructure Intel and pull off a return to form even if that means streamlining the business.

You like that?

No, I don't, and neither does anybody else on the forum. Personal attacks and name calling only make you look unhinged and childish. If you want to call people out for being ridiculous it's a good idea to make sure you don't look ridiculous first. I'm surprised the mods continue to put up with your inflammatory behavior.