Intel GPU Business in a $3.5 Billion Hole, Jon Peddie Recommends Sell or Kill

[btarunr]

Jon Peddie Research (JPR) provides some of the most authoritative and informative market-research into the PC graphics hardware industry. The firm just published a scathing editorial on the future of Intel AXG (Accelerated Computing Systems and Graphics), the business tasked with development of competitive discrete GPU and HPC compute accelerators for Intel. Founded to much fanfare in 2016 and led by Raja Koduri since 2016; AXG has been in the news for the development of the Xe graphics and compute architecture, particularly with the Xe-HP "Ponte Vecchio" HPC accelerator; and the Arc brand of consumer discrete graphics solutions. JPR reports that Intel has invested several billions of Dollars into AXG, to little avail, with none of its product lines bringing in notable revenues for the company. Xe-LP based iGPUs do not count as they're integrated with client processors, and their revenues are clubbed with CCG (Client Computing Group).

Intel started reporting revenues from the AXG business since Q1-2021, around which time it started selling its first discrete GPUs as the Intel DG1 Xe MAX, based on the same Xe-LP architecture powering its iGPUs. The company's Xe-HPG architecture, designed for high-performance gaming, was marketed as its first definitive answer to NVIDIA GeForce and AMD Radeon. Since Q1-2021, Intel has lost $2.1 billion to AXG, with not much to show for. The JPR article suggests that Intel missed the bus both with its time-to-market and scale.



A sizable launch of Arc "Alchemist" in 2021 or early-2022, in the thick of the GPU supply crisis, would have enabled Intel to cash in on sales to whoever is in the market for a graphics card. With the supply crisis over in the wake of the crypto-currency mining demand for dGPUs, Intel finds Arc "Alchemist" competing with GeForce and Radeon products purely on gaming performance, where its fastest "Alchemist" product matches their mid-range products. Both NVIDIA and AMD are ready to ship their next-generation, which is bound to widen the performance gap with Intel even further. Besides graphics, NVIDIA and AMD are ready with even their next-generation scalar compute products, with NVIDIA's Hopper, and AMD's CDNA3, increasing the performance gap with "Ponte Vecchio."

With the recent axing of the Optane Memory business, which rode on the promise of the pioneering 3D XPoint memory technology that Intel invented, it's open-season on non-performing Intel businesses, especially with CEO Pat Gelsinger seeing favorable outcomes in Washington DC for legislation that makes business favorable for Intel and increases government subsidies for the company. JPR recommends that in light of the losses faced by AXG, Intel should consider selling the entire division off and exiting this market.

The JPR editorial can be read from the source link below.

View at TechPowerUp Main Site | Source

 

[Daven]

I haven’t heard anything lately about Ponte Vecchio and the Aurora supercomputer. What ever happened to this product?

 

[Blaeza]

ROFLcopter

 

[thegnome]

Killing it would be really bad business, much better to sell or keep going (with the hope of competing well and making money).

 

[Bomby569]

They have driver issues, serious driver issues, and to be honest this was to be expected, you had to know nothing of the market to assume anything else. I called it here way before all problems.

But from raw performance the cards don't seem that bad, for a first try the silicon shows an immense progress. AMD had decades of this and couldn't do better against Nvidia then the RX480 at one point. This is money lost if they fell they can't make any progress, it's money invested if they fell they can improve.

But there is clearly a leadership problem making things worst, this launch was absolutely ridiculous at all levels.

 

[mechtech]

Ummmm

RD is always big bucks and time consuming but can pay huge dividends in the end.

Just ask TSMC.

 

[ZoneDymo]

all does not really matter at teh scale they operate, but man do I want them to bring something real.
I honestly feel Intel is too big for itself and internally people cant get stuff done or know who to talk to to get things done.

 

[Bomby569]

I honestly feel Intel is too big for itself and internally people cant get stuff done or know who to talk to to get things done.

The new CEO seems to be just as bad as the previous one.

 

[Fouquin]

Jon Peddie also once stated that he advised Jensen Huang repeatedly to not go into computer graphics while he worked at LSI. So, while I respect JP and his firm, he's not the arbiter of what does and doesn't succeed in the computer graphics industry.

 

[zlobby]

Yep, reality check.

Jon Peddie also once stated that he advised Jensen Huang repeatedly to not go into computer graphics while he worked at LSI. So, while I respect JP and his firm, he's not the arbiter of what does and doesn't succeed in the computer graphics industry.

Or, hear me out, he says some BS and then profit big time on the stock market. Boom!

 

[MikeMurphy]

When a company the size of Intel decides to enter into the dGPU market, it does so. Market timing and all the rest doesn't undermine the business case for entering into the dGPU market. Microsoft didn't abandon gaming when the Xbox360 RROD debacle came about, but rather fixed it at great cost.

Big picture is high performance parallel computing is a big deal and Intel will enter the market. Product pricing can be adjusted to remain competitive.

First gen can be rocky, but Intel has the design chop to solve these problems, even the mediocre DX11 driver issues.

It's coming.

 

[zlobby]

all does not really matter at teh scale they operate, but man do I want them to bring something real.
I honestly feel Intel is too big for itself and internally people cant get stuff done or know who to talk to to get things done.

It's the same with biology. Organisms can grow only so big. After certain extent the energy and complexity for support and living start to grow exponentially, while the organism grow linearly.
Natural selection at work.

 

[MikeMurphy]

The new CEO seems to be just as bad as the previous one.

Gelsinger is a big step up. He's been tasked with a number of legacy issues which take time to resolve. Intel has been badly mismanaged since around 2014.

I'm quite bullish on the big capex fab spend to service non-intel products. I'm glad Intel is moving on from xpoint. I see a bright future for Intel.

 

[Solaris17]

JPR recommends that in light of the losses faced by AXG, Intel should consider selling the entire division off and exiting this market.

Im sorry but what? It almost sounds like they have a vested interest. Intel, S3 or whoever entering the market would be a massive win in both terms of the company in question and consumers. It is confusing to me that they are concerned about Intels $2.xB loss when if they are that good at finance knows that is peanuts to develop a product. Let alone Intels over all capitol this is a drop in the pond.

Maybe I’m just that financially tone deaf but fucking lol.

 

[Bomby569]

Gelsinger is a big step up. He's been tasked with a number of legacy issues which take time to resolve. Intel has been badly mismanaged since around 2014.

I'm quite bullish on the big capex fab spend to service non-intel products. I'm glad Intel is moving on from xpoint. I see a bright future for Intel.

You cannot look at all that surrounded the ARC launch and say there is even a glimpse of leadership, and i'm not talking about performance issues.

 

[ModEl4]

I haven’t heard anything lately about Ponte Vecchio and the Aurora supercomputer. What ever happened to this product?

Intel: Sapphire Rapids With 64 GB of HBM2e, Ponte Vecchio with 408 MB L2 Cache

www.anandtech.com

Intel is expecting to deliver the first batch of hardware to the Argonne National Laboratory by the end of 2021, but this will come with $300m write-off on Intel’s Q4 2021 financials. Intel is expecting to deliver the rest of the machine through 2022.

 

[lexluthermiester]

Jon Peddie Recommends Recommends Sell or Kill​

Typo?

Also, Jon Peddie is not seeing the long-term big-picture plan Intel has. His advice needs to be ignored.

Any company jumping into a new market is going to experience large costs before the return begins paying in. This is the reality of any business venture. 3.5B is big money to many, but not to Intel. Jon Peddie's statement seems more like needless fearmongering than that of wisdom.

Im sorry but what? It almost sounds like they have a vested interest. Intel, S3 or whoever entering the market would be a massive win in both terms of the company in question and consumers. It is confusing to me that they are concerned about Intels $2.xB loss when if they are that good at finance knows that is peanuts to develop a product. Let alone Intels over all capitol this is a drop in the pond.

Exactly!

Maybe I’m just that financially tone deaf

Nope, you're on the money, pun intended.

 

[trsttte]

Probably comes from the same guys who were recommending that Intel should sell their fab business a year or 2 ago, big lol

 

[Vayra86]

But there is clearly a leadership problem making things worst, this launch was absolutely ridiculous at all levels.

This engineer has no business at executive level jobs.

Overpromise and underdeliver is not where you want to be at this level. Its where you absolutely should not be. That is the reason for 'Poor Vega' and now/soon Poor Arc.

 

[phanbuey]

JPR Claims AMD Tiptoes Towards Workstation Exit

We've talked about AMD's server roadmap on multiple occasions this year while simultaneously covering the company's ramp of its FirePro graphics cards. With all the Bobcat, Llano, and Bulldozer chatter, the quiet passing of AMD's workstation offering has gone all but unnoticed. That, at least...

hothardware.com

…

Let’s see. I think they’re pretty committed at this point.

 

[TheoneandonlyMrK]

JPR Claims AMD Tiptoes Towards Workstation Exit

We've talked about AMD's server roadmap on multiple occasions this year while simultaneously covering the company's ramp of its FirePro graphics cards. With all the Bobcat, Llano, and Bulldozer chatter, the quiet passing of AMD's workstation offering has gone all but unnoticed. That, at least...

hothardware.com

…

Let’s see. I think they’re pretty committed at this point.

Wtaf has that got to do with Intel's GPU business?!.

 

[Vayra86]

Gelsinger is a big step up. He's been tasked with a number of legacy issues which take time to resolve. Intel has been badly mismanaged since around 2014.

I'm quite bullish on the big capex fab spend to service non-intel products. I'm glad Intel is moving on from xpoint. I see a bright future for Intel.

Gelsinger is certainly active. Whether it amounts to anything remains to be seen..

Im sorry but what? It almost sounds like they have a vested interest. Intel, S3 or whoever entering the market would be a massive win in both terms of the company in question and consumers. It is confusing to me that they are concerned about Intels $2.xB loss when if they are that good at finance knows that is peanuts to develop a product. Let alone Intels over all capitol this is a drop in the pond.

Maybe I’m just that financially tone deaf but fucking lol.

Yes you are. If you arent leading in some way or another in GPU you arent playing. We have proof in AMD. Doing both is powerful if it works, but doing one (CPU/GPU) is already challenging enough. It echoes in everything. Even Nvidia cant handle it proper.

And now here we have Intel, who havent been making strides in GPU despite previous attempts and is notorious for tunnel vision company culture; late to market and timing just after decades of economic optimism that has recently turned stale.

Intel might need to just fight for survival of its CPU business soon. They still have no answer to enterprise EPYC and their answer to all markets underneath is yet another monolithic giant with horrible TDP. Big little is not an escape, just postponing eventual total takeover. They either shrink/cut and connect or they die. 'Death' being a slow repeat of AMDs history. Sell fab, etc.

 

[lexluthermiester]

Wtaf has that got to do with Intel's GPU business?!.

Not a single thing.

 

[r9]

From the interviews I've seen it looked to me like they made peace on how this first generation of ARC gonna go down which I think it's the right way to go about it.
Only reason to cancel this would have been to if they actually believed that they will just blow these GPU companies out of the water, companies that been the cutting edge for decades and that's just delusional.
So they are bad in DX11 which means old games that gonna run well anyways so it doesn't really matter and their DX12 has solid performance.
They also claim that they'll price it based on DX11 performance so we might be able to get same performance for less if we accept that there will be bugs to deal with.
Another thing that I think it's positive is that the number of transistors of ARC 770 is around 6800 non-XT and I think that's where they'll land with performance in DX12 and if that's true they are actually on par and it's not like they are using twice the silicon to match it's performance as that way they'll never be a success.
And lastly the driver thing well that just takes time it's plain physics there is no way around it. So they can only better with time.

 

[MikeMurphy]

Gelsinger is certainly active. Whether it amounts to anything remains to be seen..


Yes you are. If you arent leading in some way or another in GPU you arent playing. We have proof in AMD. Doing both is powerful if it works, but doing one (CPU/GPU) is already challenging enough. It echoes in everything. Even Nvidia cant handle it proper.

And now here we have Intel, who havent been making strides in GPU despite previous attempts and is notorious for tunnel vision company culture; late to market and timing just after decades of economic optimism that has recently turned stale.

Intel might need to just fight for survival of its CPU business soon. They still have no answer to enterprise EPYC and their answer to all markets underneath is yet another monolithic giant with horrible TDP. Big little is not an escape, just postponing eventual total takeover. They either shrink/cut and connect or they die. 'Death' being a slow repeat of AMDs history. Sell fab, etc.

Big datacenter is all moving to ARM64. It has nothing to do with AMD, but rather less expensive and more secure ARM64 custom silicon. Intel x86 CPU business has a lot of headwinds at present which is why Intel is now aggressively moving to develop other sources of revenue, which honestly should have started a long time ago - but it didn't.