[Hackaday] Intel discontinues Joule, Galileo, and Edison product lines (UPDATE: And Quark)

[NTMBK]

Then again, there are the products that die with a whimper, their passing marked only by a barely visible press release in an obscure corner of the Internet. Such as this week’s discontinuances from Intel, in a series of PDFs lodged on a document management server announcing the end of their Galileo (PDF), Joule (PDF), and Edison (PDF) lines. The documents in turn set out a timetable for each of the boards, for now they are still available but the last will have shipped by the end of 2017.

http://hackaday.com/2017/06/19/intel-discontinues-joule-galileo-and-edison-product-lines/

 

[Nothingness]

I wonder if that means Intel will definitely abandon x86 for IoT and concentrate on things like Quark S1000 without any horrible x86 core.

 

[SarahKerrigan]

http://hackaday.com/2017/06/19/intel-discontinues-joule-galileo-and-edison-product-lines/

Hmm. Intel suddenly moving to terminate at least part of its embedded product family? I feel like I've seen this movie before. Nine years ago, perhaps.

At least DM&P will be happy.

 

[DrMrLordX]

Feels a bit like Intel dropping mobile Atom actually.

 

[NTMBK]

Feels a bit like Intel dropping mobile Atom actually.

Intel hailed it as the next big thing that would save the company from imploding PC markets, failed spectacularly, then quietly killed off the product line. Nope, I don't see any similarities to their mobile efforts at all

 

[sandorski]

Never even heard of these.

 

[DeathReborn]

They wanted Core i# level markups for a Raspberry Pi competitor, not surprised it failed. They could have halved the price and still made a profit, just not a big one like they are used to.

 

[ninaholic37]

I wonder if that means Intel will definitely abandon x86 for IoT and concentrate on things like Quark S1000 without any horrible x86 core.

Can any Quark run x86 programs? I thought it was supposed to be like the original Pentium, or maybe not quite. Would have been cool to run DOS on a low-powered Quark CPU in 2017. Running FreeDos on my Pentium M at 800MHz works ok, but my CPU is 10.8W TDP at that speed iirc.

 

[Drazick]

We're basically looking at a company that in the last 10-15 years succeeded in a single project (Big x86 Core) and failed all other (GPU's, Other Big CPU's, IOT CPU's Mobile CPU's, Depth Cameras, VR).

On the other corner we have the company which succeeded in (Almost) anything it tried in the past 10-15.

Who are you going to put your money on?

Really, something isn't working there with the way Intel handles project.
It makes me wonder if the Big x86 success isn't an exception.
Could it be just a lack of proper competition which makes us see it as a successful product?

 

[Nothingness]

Can any Quark run x86 programs? I thought it was supposed to be like the original Pentium, or maybe not quite. Would have been cool to run DOS on a low-powered Quark CPU in 2017. Running FreeDos on my Pentium M at 800MHz works ok, but my CPU is 10.8W TDP at that speed iirc.

The latest Quark has no x86 CPU, just some Tensilica core. See this: http://www.cnx-software.com/2017/06/19/intel-quark-s1000-sue-creek-processor-to-support-on-chip-speech-recognition/

 

[FIVR]

We're basically looking at a company that in the last 10-15 years succeeded in a single project (Big x86 Core) and failed all other (GPU's, Other Big CPU's, IOT CPU's Mobile CPU's, Depth Cameras, VR).

On the other corner we have the company which succeeded in (Almost) anything it tried in the past 10-15.

Who are you going to put your money on?

Really, something isn't working there with the way Intel handles project.
It makes me wonder if the Big x86 isn't an exception.
Could it be just a lack of proper competition?

They also have a very successful mass storage division that does make quality SSDs. Otherwise, I pretty much agree with your assessment. They had huge success with their high margin big x86 cores, but have failed to diversify their product line and now they have a real competition in the one area where previously they were dominant.

 

[moinmoin]

Really, something isn't working there with the way Intel handles project.
It makes me wonder if the Big x86 isn't an exception.
Could it be just a lack of proper competition?

It's lack of proper competition in the core market and too much competition in the markets Intel tries to enter. The insane margin they enjoy in the core market essentially ensures that all their new endeavors are bound to fail due to inherent lack of huge profitability (meaning they are always going to be optimized away at some point). On the other hand their reliance on that core market pushes them to expand to lessen that reliance. Paradoxically until the margin drops there is no way for Intel to do it right.

 

[KompuKare]

Paradoxically until the margin drops there is no way for Intel to do it right.

So if someone were to suddenly enter their core market and force their margins down, they'd actually be doing Intel a favour long-term? Strangely, that might actually turn out to be true.

 

[DrMrLordX]

Intel hailed it as the next big thing that would save the company from imploding PC markets, failed spectacularly, then quietly killed off the product line. Nope, I don't see any similarities to their mobile efforts at all

Wonder if they'll make a disco ball out of discarded product this time around.

So if someone were to suddenly enter their core market and force their margins down, they'd actually be doing Intel a favour long-term? Strangely, that might actually turn out to be true.

Hmm, can't think of anybody doing that.

 

[formulav8]

Could it be just a lack of proper competition which makes us see it as a successful product?

IMO its CEO and leadership. I'm glad I'm not depending on my Intel stock anymore. AMD WILL take money from them in an ever increasing way. I know Intel has process issues but KL stinks overall and feel they could have done better personally.

 

[NTMBK]

And the other shoe drops: Intel kills off Quark as well https://www.anandtech.com/show/13888/intel-discontinues-quark-socs-and-microcontrollers

 

[DrMrLordX]

I wonder if that means Intel will definitely abandon x86 for IoT and concentrate on things like Quark S1000 without any horrible x86 core.

About that . . .

And the other shoe drops: Intel kills off Quark as well https://www.anandtech.com/show/13888/intel-discontinues-quark-socs-and-microcontrollers

Sad, but not surprising.

 

[frozentundra123456]

We're basically looking at a company that in the last 10-15 years succeeded in a single project (Big x86 Core) and failed all other (GPU's, Other Big CPU's, IOT CPU's Mobile CPU's, Depth Cameras, VR).

On the other corner we have the company which succeeded in (Almost) anything it tried in the past 10-15.

Who are you going to put your money on?

Really, something isn't working there with the way Intel handles project.
It makes me wonder if the Big x86 success isn't an exception.
Could it be just a lack of proper competition which makes us see it as a successful product?

If you are referring to AMD, I would have to disagree. What have they succeeded at really? Would you call Vishera a "success"? How about Fusion? They claimed it was "the Future", but now their most successful cpus dont even have an igpu. Tablets, phones, IOT? Dont see much from AMD there. Basically they succeeded in the consoles and finally have become competitive in cpus, mainly because they offer more cores for less money. I would only call their gpu division a "success" in the most loose definition of the term, since they have been without a top end dgpu for what, 2 years now, and are still well behind nVidia in performance per watt.

 

[Ajay]

And the other shoe drops: Intel kills off Quark as well https://www.anandtech.com/show/13888/intel-discontinues-quark-socs-and-microcontrollers

Geez, what is left of Intel's drive to IOT?

 

[Shivansps]

Geez, what is left of Intel's drive to IOT?

Atoms, C and E series mostly, even Z as used in LattePanda boards, but is not the same as a Quark.

I think this is a good move, Intel tried to go too wide and lost focus, they should concentrate on what X86 can do, and X86 has no place on a Arduino-like board.

 

[NTMBK]

Geez, what is left of Intel's drive to IOT?

Ah, but they'll always have the drone light shows

Honestly, it feels like they are clearing out Brian Krzanich's legacy, so that the new CEO has a clean slate.

 

[Nothingness]

Honestly, it feels like they are clearing out Brian Krzanich's legacy, so that the new CEO has a clean slate.

Quark was not a BK legacy but rather a legacy from Otellini.

I hope that Intel have now understood that x86 everywhere is not the way to go.

 

[moinmoin]

I hope that Intel have now understood that x86 everywhere is not the way to go.

Now Intel can earnestly fight again for their CPUs to not become commodities.

 

[Topweasel]

So if someone were to suddenly enter their core market and force their margins down, they'd actually be doing Intel a favour long-term? Strangely, that might actually turn out to be true.

Oddly enough I did a recent write up when talking about the differences on how Zen 2 and Intel's monolithic dies and their over capacity 14nm demand would impact each other. It's kind of funny, with this quote the suggestion would be to fight AMD on prices with consumer CPU sales. But while Intel is losing margin no matter what, it would be in Intel's best interest to have AMD become the defacto desktop part to deincentivize desktop and to a lesser degree laptop Intel CPU sales to allow Intel to focus on more SL and Cannon Lake server CPU wafers to keep profits up.

It might be smart for Intel to take a margin and profit hit to maintain market share (and make any of their emerging or growing markets either look less bad or more good). But my guess is they will refuse to compete with AMD on value on the consumer end to keep as much server business as possible.

 

[NTMBK]

Oddly enough I did a recent write up when talking about the differences on how Zen 2 and Intel's monolithic dies and their over capacity 14nm demand would impact each other. It's kind of funny, with this quote the suggestion would be to fight AMD on prices with consumer CPU sales. But while Intel is losing margin no matter what, it would be in Intel's best interest to have AMD become the defacto desktop part to deincentivize desktop and to a lesser degree laptop Intel CPU sales to allow Intel to focus on more SL and Cannon Lake server CPU wafers to keep profits up.

It might be smart for Intel to take a margin and profit hit to maintain market share (and make any of their emerging or growing markets either look less bad or more good). But my guess is they will refuse to compete with AMD on value on the consumer end to keep as much server business as possible.

See also AMD's recent entry to the Chromebook market. It's exactly the sort of market with razor-thin margins that will get low priority from Intel, and hence is likely to be hit by shortages... leaving a potential opening for AMD.