Intel Declares Independence From PC, Prioritizes Cloud, IoT and 5G Efforts

A week after announcing 12,000 job cuts, Intel CEO Brian Krzanich has shared vision for the company, hinting a shift in its prime focus away from PC business. In a blog post, Krzanich said that the company will be actively growing its data center business. The chip maker also plans to focus on chips and technologies for IoT devices. "The biggest opportunity in the Internet of Things is that it encompasses just about everything in our lives today-- it's ubiquitous," Krzanich said. The company also plans to boost its memory chips business and make a push towards utilizing them in data centers and various cloud services. Intel said that it has made several investments in this field, noting the $16 billion acquisition of Altera last year. The company says it will be playing a big role in the move to 5G connectivity. "Connectivity is fundamental to every one of the cloud-to-thing segments we will drive," he writes.

Over the years, Intel has failed to keep up with Moore's Law, an axiom that semiconductor density will double about every two years. The company previously extended the timeframe to 2.5 years, but Krzanich assures customers that the they are working to make further advances in order to meet the goal. "Moore's Law is fundamentally a law of economics, and Intel will confidently continue to harness its value," Krzanich said. PCWorld has extensively reported on this.

Comments

Meh....

[Anonymous Coward]

Not much of a difference between a so-called PC processor and a server CPU. Mostly marketing speak.

Re:Meh....

[Junta]

That used not to be the case, then it was the case, now it much isn't the case again, particularly for Intel.

Intel's desktop only goes to 4 cores, a small number of PCIe lanes, no ECC memory support, 2 memory channels, single socket, and won't go to very high TDP.

Servers go to much higher core counts, pcie lanes, ecc memory, 4 memory channels, and more sockets, and will drive TDP through the roof to get more powerful if needed.

Re:Meh....

[Anonymous Coward]

Not much of a difference between a so-called PC processor and a server CPU. Mostly marketing speak.

You clearly don't work on server or desktop CPUs.

Increased IO bandwidth, hardware automatic fail over, more cores, support for data-center relevant interfaces, engineered full load, full temperature, 100% duty cycle operation over the lifetime of the device, parity and error correction everywhere inside and out. These are the things that make the server CPUs worthwhile.

     

Re:Meh....

[Austerity Empowers]

There's some difference in caching and virtualization support, the latter of which I really think needs to become a client PC product anyway. Running a HW virtualized OS under linux is very useful for gaming, but right now both nvidia and Intel like to think of HW virtualization as a server feature and charge a bunch of money. But given that Windows is no longer the obvious operating system for either every day use or developer use, HW virtualization is a lot more important.

Re:Meh....

[MightyYar]

Nor is there a big gap between a typical laptop chip and the higher-end "mobile" stuff. Or rather, the gap is quickly diminishing. As for desktops, I'm not sure there is much of a low-end market anymore and the high-end can probably use Xeons or their derivatives just fine.

Gamers might have some issues with this, as their needs are not so much reliability but raw performance.

Re:Meh....

[fustakrakich]

Heh, that kinda is marketspeak for "Intel is moving out of the low margin, already saturated consumer market into something with indefinite growth potential"

Re:Meh....

[TechyImmigrant]

There's some difference in caching and virtualization support, the latter of which I really think needs to become a client PC product anyway. Running a HW virtualized OS under linux is very useful for gaming, but right now both nvidia and Intel like to think of HW virtualization as a server feature and charge a bunch of money. But given that Windows is no longer the obvious operating system for either every day use or developer use, HW virtualization is a lot more important.

My Broadwell NUC has virtualization support. Hardly a server CPU. Yes server SKUs all do the virtualization whereas not all the desktop and low power CPUs do, but if that's what you want, you can get it in a desktop CPU. For me it's great for SW development.

Re:Meh....

[Anonymous Coward]

So to do all these great server CPU things why would they cut so many people? Seems like all this server CPU awesomeness might even require a few more people.

Re:Meh....

[maligor]

Not much of a difference between a so-called PC processor and a server CPU. Mostly marketing speak.

You clearly don't work on server or desktop CPUs.

Increased IO bandwidth, hardware automatic fail over, more cores, support for data-center relevant interfaces, engineered full load, full temperature, 100% duty cycle operation over the lifetime of the device, parity and error correction everywhere inside and out. These are the things that make the server CPUs worthwhile.

   

I will give you increased bandwidth, it's quite expensive to increase bandwidth and the server parts like Xeon E5 and beyond do have increased bandwidth. As do the extreme edition consumer parts which are essentially Xeon E5s.

Core count is really a profit optimization thing, larger chips are more expensive due to taking up more space. Considering the profits and manufacturing technology Intel has, they could easily make consumer products at current prices with higher core counts and make a profit. How do you like weak AMD?

"Parity and error correction inside and out" ... I'll say you refer to ECC, I really doubt there is any difference in this inside the CPU (Aka, registers, cache and whatever else they need memory bits internally) But I'll give you that the Xeons do support ECC DRAM, but realistically this is a market segmentation issue, supporting ECC on consumer products would be trivial.

The lifetime, temperature and duty cycle stuff sounds like it's from some wank marketing handbook. These are primarily affected by manufacturing (defects) and the QA can at best only guarantee the operation of the chip on delivery. I really expect that the only difference between sever parts and consumer parts in this segment is that the server parts have a larger margin on operation tolerances.

Re:Meh....

[DigiShaman]

Basically, the "PC" is going away. The classical Desktop in typical office (tower and monitor) will be the "workstation" outfitted with the Xeon lineup. Everything else will either be embedded chipsets in monitors; think AIO units (All-In-One) running off ARM, or typical BYOD phones and tablets. This is where the industry is headed in the next few years. This is precisely why Windows 8 overshot the paradigm with a GUI geared to content consumption and not multitasking. They screwed up, and Windows 10 is the compromise between Windows 8 and Windows 7. So yeah, there you go.

Re:Meh....

[mattventura]

Probably due to lack of competition from AMD. Why waste money on R&D when your only competitor is still a few laps behind you?

Re:Meh....

[thinkwaitfast]

There are many companies that market embedded chipsets. These are wafer thin (aaahaahahaa) margin products with differentiation coming in percentages of pennies by the millions. They are never seen by consumers, so couldn't care less if your remote control has an ARM, PIC, PPC, ATMEGA, Z80, 6809 or whatever Inside. You push a button, magic happens and your garage opens. That's all that counts. And price, so whenever TI drops quantity pricing and the new redesign requires a new part, the engineer gets the spec sheet and two weeks to deliver a prototype for testing. The software engineers learn the new opcodes, make a few tweaks to the C code and a month later your embedded widget is running on a new processor with reduced chip count because the new processor has more embedded whatever.

Intel exists, almost solely on a completely different business model of high wafer yields due to maniacal focus on a single process good for a couple of years that creates very high margins on each part. How they manage to wade this transition will be interesting.

Re:Meh....

[Anonymous Coward]

Are you sure you aren't just using marketing speak?

Intel defines a Pentium as desktop-not-server (despite whatever server applications that a user might come up with) and defines a E5 Xeon as server-not-desktop (despite whatever desktop applications that a user might come up with).

The entire product array is whatever-you-want-to-use-it-for and it's Intel who fancifully claims that some are desktop and some are server.

Re:Meh....

[sinij]

Actually, it is a lot more likely that your smartphone, when docked, will be transformed into PC.

Re: Meh....

[Anonymous Coward]

Yes, agreed. In fact Apple has melded some of the cloud syncing between iOS and OSX already. But yes, future phones will connect via a wireless KVM acting as nothing more than a kiosks in public settings; for example, while setting in a plane or at the home office. Perhaps the same wireless tech will charge the battery via induction too. And just current phones, they can always be restored from cloud backup in the event of a hardware failure.

Re:Meh....

[Anonymous Coward]

As someone who built an unRaid server a few months ago with both a linux and windows VMs for serving media, games, and cloud services, I couldn't agree more. What really bothers me is that the h97/z97 chipsets are all limited to a max of 8Gb per memory slots. This is extremely limiting when you are attempting to run multiple VMs.

Sure, Intel will tell you to just buy a high end server class CPU and chipset... the problem with that idea is the total power consumption goes way up considerably with that hardware. I can't buy a low end chip with low power consumption with a decent memory capacity. When I'm running a low end personal server 24-7 that really adds up quickly.

Re:Meh....

[Anonymous Coward]

Well except that intel desktop goes to 6 cores, announced 8, and rumored 10 in the next couple of months. 4 memory channels, and a fair amount of PCIe lanes (40). Just look at the -E line of desktop chips, not the mainstream i7.

Re:Meh....

[Anonymous Coward]

no ECC memory support

O, rly?

Re: Meh....

[Anonymous Coward]

Not quite higher end gaming desktops and servers are closer than you think. For example the x99 chipset is for 6 to 8 core cpus, have 4 memory channels and if you want to , it'll actually work with a xeon and eec memory (the memory controller is on the processor so you'll need the CPU so the eec lanes need connecting to the pins (mb manufacturers don't tell you if they did, but a lot do) and a xeon needs to be used, however it can work).

Re:Meh....

[nycsubway]

That is very true. The majority of business applications, spreadsheets, word processing, etc, can all run pretty well on an average smartphone. All you need is a desktop size screen, a USB and network port and the "CPU" can be embedded in the screen.

What I would like to see someday is a smartphone dock, where you can plug the smartphone into a monitor/mouse/keyboard and use the smartphone like a desktop PC.

Re:Meh....

[Blaskowicz]

You can. If you get the Skylake gen i.e. Pentium G4400, upcoming Celeron G3900, i3 6100 etc. then your memory ceiling is 64GB up from the older 32 GB. You can even get ECC on Celeron/Pentium but at the cost of an expensive motherboard wirh C2xx chipset. (there is IOMMU too)

Re:Meh....

[mrclevesque]

That was definitely off topic, -1

This one too

Your phone is the next PC.

[LWATCDR]

The microcomputer first replaced the low end mini-computer and word processing systems.
Next they replaced larger mini-computers like the PDP-11.
Later they replaced the Super-minicomputers like the VAX.
The cell phone will probably replace the PC.
Imagine a Windows Phone that you put on your desk and hook up a USB 3 connector to your monitor that provided power and a network connection to your phone and video to the display. Mouse and keyboard can be USB connected to the monitor or bluetooth. You could even use the phone's screen as a touch pad.
If Intel can get a good x86 mobile cpu and microsoft may just win the mobile phone market.
If not then it maybe Android or Apple but I doubt that Apple wants to do that since it is not all that interested in the enterprise market.

Re:Your phone is the next PC.

[Junta]

Much hinges on whether MS can deliver a good and *perceived* to be good mobile experience. As it stands, today people prefer to have two distinct devices, in order not to run MS on their phone. Or it could be because MS is being obstinate phones being ARM, limiting the 'desktop' versatility.

Re:Your phone is the next PC.

[thinkwaitfast]

And your watch will be you next phone.

Re:Your phone is the next PC.

[irrational_design]

It was a good post until "microsoft may just win the mobile phone market." This has already been won by Android (and to a certain extent by Apple as well). Microsoft can't win something they have already lost. Especially when they haven't even hardly begun the race when everyone else is already at home watching the highlights on the 11 o'clock news.

Re:Your phone is the next PC.

[NotDrWho]

What kind of FPS can I get running The Division on my phone in 1440P?

Re:Your phone is the next PC.

[npslider]

The car of the future: Intel Inside!

Drive up to 40% faster while using 20% less energy! And with Turbo-boost technology, win those street races every time! Multi-core technology allows you to be more productive on and off the road!

*** Warning: Overclocking may lead to engine overheating and -- Your car has encountered an unknown error and

Re:Your phone is the next PC.

[npslider]

Everyone thought Netscape was king of the hill. Mind you, that was a very different situation, but it shows that any platform that is on top can fall, and underdogs just may take the fallen crown.

Re:Your phone is the next PC.

[npslider]

The problem is... where do you put that huge heat-sink and fan for the Extreme editions of the new Intel Moble Core i9 CPU on your new smartphone?

Re:Your phone is the next PC.

[Anonymous Coward]

Zero. It's a shit game on a shit system. Ignore it and move on.

Re:Your phone is the next PC.

[Anonymous Coward]

Well I would say Apple is the biggest winner, they take more than 90% of the profits from the entire world wide cellphone market.

http://www.wsj.com/articles/apples-share-of-smartphone-industrys-profits-soars-to-92-1436727458

Re:Your phone is the next PC.

[Anonymous Coward]

I have Lumia 920 running Win10. Got it couple of years ago with Win8. Decent performance. Decent specs. I'd say it's better AND reliable than much hyped Galaxy S series (I have used one from that series).

The problem is app eco system. Devs don't make apps because of low marketshare. People don't buy because of lack of apps.

With Xamarin purchase and making it free, devs could start developing cross platform apps with much ease. We'll have to wait and watch...

Re:Your phone is the next PC.

[LWATCDR]

You mean like Nintendo and Sega where the winners in the Console war... until Sony showed up. Then it was Nintendo and Sony until Microsoft showed up...
Or how DEC and IBM where real computers companies while Apple made toys...
Or IBM won the PC war?
It may be over for now but not forever.

Re:Your phone is the next PC.

[dejitaru]

That's the thing. For casual use maybe your fone will be the "next computer" but as long as there's a PC gaming industry, they will keep pushing PC performance boundries.

Re:Your phone is the next PC.

[LWATCDR]

Today probably not all that fast but in two years? Besides gaming and other HPC tasks will not go away but for the I need to run Office and Quicken crowd phones are getting very close to good enough.

Re:Your phone is the next PC.

[macs4all]

I doubt that Apple wants to do that since it is not all that interested in the enterprise market.

That's not exactly true. They just aren't interested in the SERVER market. But in the past few years, they have taken a renewed and serious interest in the enterprise market overall.

Dig down into this deceptively-fluffy page. There's actually a lot of information there.

Re:Your phone is the next PC.

[EndlessNameless]

It's not even hatred of MS that doomed their phones. I tried one in a store, and it was actually pretty usable. Arguably better than the contemporary Android phones.

But they had no apps. Back then, Google Maps was the top dog, and it wasn't on Windows Phone. Maybe one app out of a dozen that I wanted actually ran on it. And I couldn't find replacements for half of them.

If people don't develop for the platform, it's not gonna live long. Even the perception of "no apps" is deadly because it makes people---both devs and users---shy away from that platform.

Microsoft failed to get the jump on mobile with Windows CE, and Windows Phone was good but too late to the party.

I would prefer Windows Phone over iOS (as it's the less-walled garden), but an OS without developer support is worthless.

Re:Your phone is the next PC.

[ranton]

That's the thing. For casual use maybe your fone will be the "next computer" but as long as there's a PC gaming industry, they will keep pushing PC performance boundries.

No reason video acceleration needs to be done in the cell phone. That could happen in some module connected to the monitor / tv. This could easily be the duty of the next generation of consoles, with them simply being external GPUs which give your mobile devices better gaming experiences.

Re:Your phone is the next PC.

[EndlessNameless]

Netscape v Microsoft was a David/Goliath matchup when you look at the companies behind the software. Plus Microsoft used illegal means to promote IE over Navigator.

The phone OS market is entirely different. Both Android and iOS are supported by large corporations. Both are entrenched with an ecosystem of apps dependent on them, and users' money sunk into those apps.

Navigator actually did have some unique plugins, but other browsers eventually implemented their plugin API. And users didn't have their money invested in the platform.

I'd say the chances of upheaval in the phone OS market are nowhere near the same as what it was for the 90s browsers.

I'd rate a compatible-yet-superior implementation like Cyanogen as the best bet, but even that's a long shot. Google has more than enough resources to respond to any serious contenders.

Re:Your phone is the next PC.

[pr0fessor]

Except the mobile phone market hasn't been replaced yet so the race to the end of that particular technology isn't at an end, and no one has finished. If they think they are done and are relaxing eating chips in front of the tube then someone else does have an opportunity.

Re:Your phone is the next PC.

[dejitaru]

True, and I believe I saw some external video cards recently being made. But in terms of performance, I don't think having an external video card receiving data via USB-C or the likes will ever match or exceed one directly plugged in to a motherboard. But who knows.

Re:Your phone is the next PC.

[Anonymous Coward]

No reason video acceleration needs to be done in the cell phone. That could happen in some module connected to the monitor / tv. This could easily be the duty of the next generation of consoles, with them simply being external GPUs which give your mobile devices better gaming experiences.

Do you "use your phone as a PC" people even read what you write? You actually think people are going to go through all that shit to play games? Especially when gamers already have a PC running a 3570k at the low-end that shits all over any mobile processor likely to come out for the next decade. You are dreaming if you think that will be replaced by a phone plugged into a monitor. Lol.

Re:Your phone is the next PC.

[ranton]

True, and I believe I saw some external video cards recently being made. But in terms of performance, I don't think having an external video card receiving data via USB-C or the likes will ever match or exceed one directly plugged in to a motherboard. But who knows.

I would have agreed with that, but after looking up the throughput of PCIe 3.0 and USB-C, it looks like they have very similar speeds. USB-C provides 1.25 GB/s while PCIe 3.0 is about 1 GB/s. There is the distance factor but it appears the speeds could be similar.

Re:Your phone is the next PC.

[Anonymous Coward]

I don't think having an external video card receiving data via USB-C or the likes will ever match or exceed one directly plugged in to a motherboard.

What about CPU speed? And RAM size? And storage space? And storage speed? All the things, not even counting video card power, that make a PC superior to a phone for playing games. Whatever you're smoking that makes you think a phone will replace gaming PCs anytime in the forseeable future, I wish you'd pass some this way!

Re:Your phone is the next PC.

[Anonymous Coward]

Today probably not all that fast but in two years? Besides gaming and other HPC tasks will not go away but for the I need to run Office and Quicken crowd phones are getting very close to good enough.

So when everybody's done spending money on one of these new phones and a monitor that supports this scheme, they get to sit down to an experience that is inferior to a half-decade old Sandybridge laptop they could have bought off of eBay for 50 dollars. Wow. The Future(TM).

Re:Your phone is the next PC.

[JoeMerchant]

Dump the heat into your bloodstream via the watchband... Now, where do you get the power for all that heat?

Re:Your phone is the next PC.

[JoeMerchant]

I know Netscape lost, but did IE really win long term? I mean, from my perspective, not - most people I know only use IE when forced to...

Re:Your phone is the next PC.

[npslider]

Zero point energy collectors.

Re:Your phone is the next PC.

[Holi]

Not USB 3.1, Thunderbolt 3, which at 40 Gbps seems to be fine for current video cards.

Re:Your phone is the next PC.

[Anonymous Coward]

Inexplicably, MS dropped their project to make Windows 10 (Phone) run Android apps. They did learn enough about Android to build a Linux subsystem for Win10 (in the previews now). But why not just run Android apps, at least in the phone?

I have an older (.5GB RAM) Lumia (still says "Nokia" on power-up) 635. It was inexpensive at the time, with better as-delivered usability and features than Android. For an olde farte like me, it's fine in Windows 8, and at least Mail works in 8 on the phone (not so much in 10 on a laptop, where I switched to Thunderbird rather than keep fighting with Mail and Calendar). Given the small memory and storage available, I don't think I'll ever see 10 on the phone, but that's OK - it's old enough that I'll probably be replacing it next time the cell provider switches systems again, soon. Already, I can't use the phone in some expansion service areas, though it's not normally a problem, because it doesn't support (and never will) the new radio bands and VoIP stuff that T-Mobile requires for them.

Re:Your phone is the next PC.

[Anonymous Coward]

It's ok, but realize that thunderbold 40Gbs is uni-directional, and it's only 5GB/s. That equivalent to a 4x PCIe 3.0 slot, where we are running video cards in 16x slots now. That said, I'm sure that a lot can be done to minimize the throughput from CPU to GPU bandwidth issues as well.

Re:Your phone is the next PC.

[Anonymous Coward]

I would hate to have to use my phone as my primary computing device, even if it was docked.

Re:Your phone is the next PC.

[LWATCDR]

Unless streaming games takes over. I also doubt that the Cell phone will take over the CAD/CAM, Video production, or any other HPC tasks. But for the masses that get on facebook, use quicken, and run office a phone maybe good enough. With an X86, monitor, mouse, and keyboard a phone running Windows 10 could also run all those legacy applications that a lot of people can not live without.

Re:Your phone is the next PC.

[Anonymous Coward]

Then it was Nintendo and Sony until Microsoft showed up...

Wait what? Microsoft didn't win even single generation of consoles. last 4 generations were Sony (PSX) Sony (PS2) Nintendo (Wii) and currently it's Sony again.
All Microsoft did in console market was burning through gigantic pile of cash their Office and Windows division generated.

Re:Your phone is the next PC.

[bingoUV]

From blood glucose. Or cholesterol.

Re:Your phone is the next PC.

[ShieldW0lf]

Yes. They had the best and most popular browser for a long time.

To be quite frank, browsers really aren't that much more capable than they were in 2000. Take AJAX. You could do the exact same shit with IFRAMEs. Except, your audience were running 300Mhz processors, so really, you couldn't do shit.

Not that improvements haven't been made, but it's really the hardware that made the difference.

Re:Your phone is the next PC.

[JoeMerchant]

I think V8 made some measurable gains (turning a giant steaming pile into a slightly smaller, slightly cooler steaming pile...)

Too bad "we can't be trusted" to develop plugins like the NaCl project anymore. Quick search looks like pNaCl is trying to address that... but I smell Javascript again...

Not really news

[BigU 03C0in]

So Intel, who has focused on the performance enthusiast crowd, now sees dollar signs by trying to put their chips on all your home devices. They're sticking with the one existing market they have a chance at making money from. What a shock.

Seems strange to go from having one main competitor to having many in the embedded world.

Re:Not really news

[npslider]

They see that traditional computing devices are on the decline. They are changing with the times and positioning themselves to ride the wave of smaller devices and emerging technologies.

Re:Not really news

[Anonymous Coward]

really glad that they sold off their security arm because that is one thing idiotsOfThing doesn't need.
ubiquitously unsecured by design.

Re:Not really news

[Rockoon]

The elephant in the room is that this is the year that intel falls behind in semiconductor process technology.

Intel has been enjoying a comfortable lead, but this year TSMC opens up a 10nm fab. Intel recently disclosed that they wont be able to open up a 10nm fab until at least the end of NEXT year.

Dont believe their press-release bullshit hype. Intel forecasts big trouble for itself from here on out, because they just dont have the lead in technology anymore.

Things would be different if Intel was running a Pure-play or IDM business model, but because they arent there is no way that they will be able to keep their fabs busy when Pure-Play and IDM fabs have the same or better process technologies. On top of all this, Intel had locked in lower yields than everyone else in their rush to be first to a 14nm fab.

If you have Intel stock, its time to sell.

Re: Not really news

[Anonymous Coward]

They don't really compete with TSMC in a substantial way though. This is really about the paradigm shift from PCs to tablets and mobile devices more than anything.

Re: Not really news

[Rockoon]

They don't really compete with TSMC in a substantial way though.

When AMD can buy time on TSMC's 10nm fab while Intel can't buy time on anyones fabs... you are obviously just an ignorant person that doesnt understand the business.

It is technically possible for Intel to buy time on another manufacturers fabs, but it would be suicide for that company as they will lose their Pure-play or IDM status.

Re:Not really news

[Mashiki]

They're on a decline in the Americas and parts of Europe. That's mainly because your average person doesn't need a desktop to check the news or their email. That doesn't hold true of course to programmers, gamers, game developers and so on. It also doesn't hold true in many parts of Europe, Asia or the Middle East or Africa either, where the PC is increasing marketshare because there is no saturation.

"Sounds like a great idea!" said AMD

[NotDrWho]

"Don't worry, we'll be happy to take over."

Re:"Sounds like a great idea!" said AMD

[Anonymous Coward]

This truly sounds like a good opportunity for AMD to take some market share from Intel. If Intel ends the research and development by massive layoffs, spends billions on buying crappy companies (McAfee, Altera, etc..) and focuses just on trendy fads (cloud, experience, and IoT), they are doomed. The current Intel CEO seems to be following Microsoft's footprints from market dominance to quick destruction.

Re:"Sounds like a great idea!" said AMD

[ranton]

This truly sounds like a good opportunity for AMD to take some market share from Intel. If Intel ends the research and development by massive layoffs, spends billions on buying crappy companies (McAfee, Altera, etc..) and focuses just on trendy fads (cloud, experience, and IoT), they are doomed. The current Intel CEO seems to be following Microsoft's footprints from market dominance to quick destruction.

Yeah, and companies that focused on fads such as personal computers in the early 80's were doomed as well. We have a lot to learn from visionaries like IBM's John Akers who dismissed PC's as a fad best handled by companies with no vision, like Microsoft and Intel.

Re:"Sounds like a great idea!" said AMD

[Impy the Impiuos Imp]

The point was laying off your own core competency engineers and buying up companies to plug together some quick business model has had...problems in the pase, as demonstrated over and over and over.

This shop is closed due to richness

[Anonymous Coward]

Everybody wants to be Apple. How long will it take for them to realize that Apple has nothing up their sleeves? Smartphone sales are already flattening. How many chips can you sell for one smartphone, Intel? How much CPU power can you put into ever thinner cases? The ARM CPUs you're trying to compete with sell for how much? The cloud. Don't make me laugh. You want to be in a buyer's market? IoT will be made in China. It already is. They will not buy Intel chips at a premium. It is a shame that AMD has lost its mojo. They would have kept Intel focused.

Trumped

[Anonymous Coward]

"Moore's Law is fundamentally a law of economics, and Intel will confidently continue to harness its value," Krzanich said.

Sorry, but Laws of physics inevitably trump "laws" of economics.

Re:Trumped

[Anonymous Coward]

Absolutely true, but try to convince any MBA or economist.

Re:Trumped

[Bob_Who]

I agree, but we might need to find a better verb than "trump".

As for economics, laying of 12,000 people really sucks when its so easy to low ball the next graduating class and quietly hire others, as needed. Its a human resource style that interchanges employees as if they were silicon chips and not carbon based life forms.

Re:Trumped

[Rockoon]

As for economics, laying of 12,000 people really sucks when its so easy to low ball the next graduating class and quietly hire others, as needed.

Thats not what Intel is doing. They are laying off people because their own forecasts are grim. They lost the lead in fab process, and by the time they finally get a 10nm fab up and running they will only be catching up to the two other semiconductor manufacturers than have already had 10nm fabs running off chips for a year.

If you have Intel stock, its time to sell.

Hyp

[Alomex]

The internet of things is mostly hype, as people with nests at home have learned . Sure you can talk to your thermostat at home via the internet, but why would you want to?

Re:Hyp

[dugancent]

Exactly. I'm been disconnecting things from the Internet lately, not connecting.

Once the lovely wears off, you're left with why. I haven't found a good answer yet.

Re:Hyp

[TheGratefulNet]

until IoT runs its course, it will sustain a bunch of businesses that make connected (and spying) toys.

after enough bad things happen, the fad will die down and remote access and remote control will still be here (which is good) but hopefully the spying and datamining will stop.

that's my hope.

I like iot, but I don't like the spying and phone-homing. I DIY it so I know my own stuff isn't doing anything I didn't tell it to. most people don't DIY it, though.

Re:Hyp

[Anonymous Coward]

It's not hype, it's the best way to get pwned! Waze users recently pwned with ghost trackers. The cloud is a surveillance state. Do not participate.

Re:Hyp

[b0bby]

I got an Ecobee, mostly for the fact that I can tell the fan to run a certain number of minutes per hour and it has a remote sensor so it can tell that my upstairs is getting too hot (thermostat is on the ground floor). The remote access is a plus; I can set it to vacation mode after I have left, for example. Not worth doing it just to get that, but once you're buying a little wifi computer for your wall, why not have access remotely?

Re:Hyp

[JoeMerchant]

I have a different "IoT" thermostat, and my best use for it is to set it in "vacation mode" from my phone, after I've already left on vacation and forgot to do it before I left. It's also handy to have the thermostat interface on my phone so I don't have to walk my lazy butt over to the thermostat to interact with it. The temperature/humidity and system operation logging and graphing functions are kind of informative to look at / compare how long the system runs when set at 74 vs 76 degrees - a couple of degrees can make a huge difference in daily run time, or almost none, depending on the weather outside.

Re:Hyp

[JoeMerchant]

I've got one "connected" light switch - it turns on at dusk, off at 10pm - outside lights... and they can be controlled from our phones. Works, works well, I'd probably install a half dozen more of the things around the house if they weren't $50 each.

Re:Hyp

[Anonymous Coward]

I have a $20 motion sensor switch that does all that and does not need to be controlled by the phone. It just turns on when someone is walking past it. AND is does not need Apple or Google account to work.

Re:Hyp

[Anonymous Coward]

It COULD be a great idea it IoT, however it will be a total bollocks up mess.

This is because so many of the players will be trying to use you data for marketing, they will want the device to phone home.

Companies have somehow come to believe they are entitled to your data.

Until that is gone, the IoT can not be trusted, it is simply a spy in your home.

The IoT has the ability to empower disabled people, make their lives significantly easier, but they won't because they won't be a big enough market share for advertisers to be interested in buying data from the makers. The sooner Marketing/Advertising disappears the better.

Re:Hyp

[Anonymous Coward]

I keep thinking the purpose of IoT is to create a bigger DDoS platform, but not sure how it should be conscripted -- in-border or outside-border.

Re:Hyp

[thegarbz]

Your specific applications for IoT are mostly hype. Replace IoT with "sensor network" which is what it was called in the 90s and you'll realise your entire life has benefited from IoT already. It's more than just some overpriced thermostat and a lightbulb which you can control from your phone.

Re:Hyp

[eumoria]

The "Internet of Things" is a buzzword used by middle management to make it sound like they're doing something and innovating. Very few people will ever use any simple appliance on the internet outside of HVAC. Home automation has been around a long time with very little adoption adding a swanky term/acronym to it doesn't change adoption rates. Also remember when the PC was dead in 2008? Oh, right. There are tons of people in positions where they're asked to come up with a new way for the company to make money and/or seem like they're "cutting edge." The "Internet of Home Appliances that Don't Require the Internet" is just another one in the cycle.

What's this mean for gaming PCs?

[the_skywise]

Obviously the market is slowing down and that's reflected in my own PC builds as I've gone longer and longer between refreshes as the technology held up longer. But if they're going to shift development to appliances it sounds like they're giving up on the enthusiast market altogether. I can be happy with AMD too but that's gonna affect the whole market of motherboard manufacturers as well.

Also I think the IoT market has reached saturation. I don't want or need internet access to my refrigerator and how much innovation do you need for thermostat chips?

Likewise cloud tech - You can switch to targetting cloud based machines but they were all pretty much on Intel already and the market for that is far more limited than the consumer market.

Basically... Intel just surrendered.

Re:What's this mean for gaming PCs?

[Anonymous Coward]

Maybe Moore's Law is no longer applicable ? Perhaps it's like a sprinter: one can keep the pace only for so long

Re:What's this mean for gaming PCs?

[the_skywise]

Oh sure - I didn't think they'd maintain Moore's law forever but they don't seem to have any other ideas to get around that or to further the PC market.

Make 'em faster.
Make 'em faster.
Make 'em faster.
Oh... we can't do that anymore? Oh well... let's do something else.

memory/mb bandwidth is still a problem and intel hasn't really dug into that area yet. Nor have they looked at getting away from the aged x86 instruction set and using some new sets that are more conducive to performance. (But muh Microsoft and Linux! Well you're already giving up on that market and going elsewhere...)

Re:What's this mean for gaming PCs?

[DaveMikulec]

Indeed. I built a new Haswell based machine a couple years back for gaming and it still has more power than I need. I don't see myself building another any time in the foreseeable future.

Re:What's this mean for gaming PCs?

[Anonymous Coward]

1975-2015. That's a pretty long marathon in my book.

Re:What's this mean for gaming PCs?

[Anonymous Coward]

Intel moved away from x86 years (decades?) ago.
Yeah, the compiler still emits that, but it's all translated in hardware for backwards compatibility purposes.
No real reason to change that unless you want to eke out the last ounce of performance per watt.
And having internal instruction set translation means that it's easier to make changes to the design without having to redo your compiler every single time.
The memory bandwidth problem is something that will probably never be satisfactorily solved.
It's the largest problem in computer engineering and other than adding memories all over the place that the programmer has to use explicitly, no one has a clue what to do.
If someone solves that they will be very rich.
Computer architecture is starting to look like a solved problem.
Every single processor sold today has the same basic structure, and even the different instruction sets are largely the same.
There are some interesting fringe research projects but those will probably never become mainstream because the fundamental problem (memory bandwidth) remains.

Re:What's this mean for gaming PCs?

[ChrisMaple]

I'd buy a new machine, but the significant performance increase I'm looking for just isn't available. I'd like to see the single-thread instruction rate for moderate-sized programs (something that can be held in on-chip cache) double. There are optimizations that can be done to allow one core to go really fast, and I don't think they're being done.

Re:"Ubiquitous"

[zenlessyank]

Yet you are the first one to not ignore him. ;)

Not impressed by IoT efforts so far

[werepants]

Most of what Intel is pushing (Edison, Galileo, etc) in terms of IoT hardware is more expensive than existing solutions (Raspberry Pi, Arduino) which is pretty antithetical to many IoT applications where what you want is low cost. So I'm a bit skeptical that they are really serious about moving into that space, or at least if they are, they need to up their game. The Pi Zero, or the recently Kickstarted C.H.I.P. ($9 all-in-one computer with wifi and li-poly charging circuit) are much more aligned with IoT applications.

To get most business cases to close, you have big problems to address with respect to cost, power, and connectivity, and from what I'm seeing Intel isn't really fielding anything competitive.

Re:Not impressed by IoT efforts so far

[macs4all]

Most of what Intel is pushing (Edison, Galileo, etc) in terms of IoT hardware is more expensive than existing solutions (Raspberry Pi, Arduino) which is pretty antithetical to many IoT applications where what you want is low cost. So I'm a bit skeptical that they are really serious about moving into that space, or at least if they are, they need to up their game. The Pi Zero, or the recently Kickstarted C.H.I.P. ($9 all-in-one computer with wifi and li-poly charging circuit) are much more aligned with IoT applications.

To get most business cases to close, you have big problems to address with respect to cost, power, and connectivity, and from what I'm seeing Intel isn't really fielding anything competitive.

The last time that Intel was seriously in the Embedded game was with the 8051. And even that had some ridiculous design quirks compared to microcontrollers from Motorola, NEC, ST, etc.

Re:Not impressed by IoT efforts so far

[Anonymous Coward]

You're thinking of the wrong IoT.

Tinkerboards like Edison, Gallileo, RPi, and Arduino are part of IoT, but they're not a major one.

The biggest part of IoT is in prebuilt black-box devices where Intel can ship a huge load of cheapo, bulk Atom SoC's to a company that makes the "Things" in the IoT. Like thermostats or security camera DVR's or home network gear (not just routers, but small NASes and other single-function devices).

The IoT is misnamed. It should be the IoSYDN: The Internet of Shit You Don't Need. It will quickly degrade into the Internet Ghetto of Broken Shit You Don't Want. Intel probably wants to cash in on the gold rush and get out before that happens, but I think they may already be too late.

Re:Not impressed by IoT efforts so far

[werepants]

Even then, though, why get an Atom when you could get something like the TI MSP430? Low power, cheap eval boards, dirt cheap in bulk, and already fairly established. Intel is just putting products out there with the idea that anything with the Intel name on it will sell, but they aren't offering anything better suited to the markets they claim to be targeting. IoT is going to be big, but it's going to be stuff so cheap as to be disposable - think Amazon's buttons, stuff like that. I don't see Intel having much success there.

Re:Not impressed by IoT efforts so far

[Cassini2]

The issue with Edison and Galileo is that they tend to be crippled platforms. It's been a while since I last looked at them, but I always ask the same questions. Does it have Ethernet? WiFi? 5V/3.3V serial port? Bunch of pins for miscellaneous I/O? SATA port or SD card? Enough RAM to run a standard version of Linux? HDMI and keyboard ports for testing?

The latest Raspberry Pi has most of that. However, there always seems to be a catch with Intel products. Intel doesn't want the Edison or Galileo to compete in the PC market, and as a result, they are crippled platforms.

Re:Not impressed by IoT efforts so far

[drinkypoo]

Even then, though, why get an Atom when you could get something like the TI MSP430? Low power, cheap eval boards, dirt cheap in bulk, and already fairly established.

The MSP430 is a bit pathetic, because it never managed to build a community, probably in turn because their dev env is garbage. Why would you even say MSP instead of atmega?

Re:Not impressed by IoT efforts so far

[werepants]

Well, mainly pointing that out because parent poster was arguing that Pi/Arduino weren't for anything that would be rolled out en masse. Whereas the MSP430 always struck me as a more industry-focused solution, with the dev kit focused more on that environment than making things easy for the kid in his garage, as with Arduino. That said, certainly atmels and pics are their ilk are ubiquitous - and the overall point is that's all you need for most IoT applications, and the Atom or anything else Intel is bringing to the table doesn't really offer anything compelling.

Ugh, not this crap again

[s.petry]

Yeah yeah, the PC is dead we know. I have heard this same thing repeated countless times over the last couple decades. In order to make such a claim you have to also claim that our thin client ability has magically evolved so that we can all work on a 4" screen. Oh wait, that has not happened. So people are once again claiming "MAGIC IS REAL" and "REALITY IS FOR PEOPLE WHO CAN'T FANTASIZE LIKE WE DO!".

Could you claim "hey, lots of people work with smaller screens?" Sure. Many people can only afford to browse the internet on a phone too, but given the option would they want to have a nice big screen? Immediate response? Local processing power? Yup, they sure would. Your dichotomy is false, time to go back to the liar's drawing board.

Also, give it a rest already! IoT is not a "thing", so stop making it out to be one. We don't want your chips implanted in our hands, and the ability to have everything we do controlled and tracked with biometrics. We like having cash so that people can't track all of our purchases and use that data as a weapon against us. Sorry, but your psychologically targeted ad is not a benefit to me or society, it's a benefit to you and a weapon against myself and society.

Strangely, this whole subject ignores the fact that our datacenters are packed to the brim with PCs. Not high end graphics PCs, but the same processor and chipsets runs the majority of businesses today AND personal computers. But the PC is dead? OMG! RIP SATA and SAS, we really miss you USB and PCI, we'll never be the same without you DDR 1+. Lot the sobbing ensue! No really, they said it was DEAD!

You people that can't handle reality really get on my nerves sometimes.

Re:Ugh, not this crap again

[LWATCDR]

No but it will be a different form factor. Did you read what I wrote?
Just what would you call a phone with an x86 running Windows using a full sized monitor, keyboard, and mouse? It is just a small form factor PC.

Re:Ugh, not this crap again

[coastwalker]

Agreed that PC's are not going away. The shift in development spending towards datacenters makes sense if you believe that market growth will be in datacenters. This is likely given that the world is saturated with smartphones (and pcs) which are access terminals to applications like social media which run in datacenters.

Re:Ugh, not this crap again

[tburkhol]

Just what would you call a phone with an x86 running Windows using a full sized monitor, keyboard, and mouse?

Useless.

Phone and tablet OSes are designed to do one thing at a time. Even if you plug in a real monitor and keyboard, you're going to have trouble dragging a table from your spreadsheet into a word processor on your phone.

I'm not saying that a phone will never have enough compute power to equal a usable desktop, I'm saying that the interface on the phone is not appropriate for desktop use, nor vice-versa. I thought we all learned that from Apple and the early Windows tablets.

So now you say, what about a computer that uses a phone-like UI when it's untethered, then switches to a desktop-like UI when it's near its base. Maybe. When the phone has enough power to run photoshop filters while I watch. Or to compile a decent sized codebase. I'd take a computer capable of doing real computing, small enough to fit in my pocket, and with a wireless connection to a 30 Hz, 30" monitor. The thing is that real compute power and long battery life are kind of opposite ends of the design spectrum.

Re:Ugh, not this crap again

[LWATCDR]

"So now you say, what about a computer that uses a phone-like UI when it's untethered, then switches to a desktop-like UI when it's near its base. "
You mean like Windows 10?
"When the phone has enough power to run photoshop filters while I watch. Or to compile a decent sized codebase."
The first may be sooner than you think thank to GPUs,CUDA, and OpenCL.
The second well not that many people compile code. Of course I say that with VS 2015 running on my 32GB Xeon equipped PC with an SSD. I am really talking about the Facebook, Quicken, TurboTax, Office class of users vs high end users.
" I'd take a computer capable of doing real computing, small enough to fit in my pocket, and with a wireless connection to a 30 Hz, 30" monitor. The thing is that real compute power and long battery life are kind of opposite ends of the design spectrum."
Well since a modern iPhone would spank a Cray-1 I would say that may change.
How about this as a solution. When running on battery you use a single core and throttle it down to a lower clockspeed and even a lower voltage. When tethered it spins up all cores and clocks?
Power is really less of an issue than heat IMHO. Frankly I see you more attaching it to the monitor with USB 3.1 and the monitor supplies power and a wired network connection to the phone.

More to the point

[H3lldr0p]

What good does it do you? What problem does it solve and does it solve the problem in an economically advantageous manner?

As far as I can see, it doesn't. Would it be nice to have my fridge tell me everything in it? Certainly. Would it be nice if it printed out a list of everything it is now missing? Again, yes. But so what? That saves me all of five minutes. Because I am certainly not going to leave the next part, the shopping, to some third party like Amazon or Google or WalMart. This isn't me being a luddite. This is me just being practical.

The fridge isn't going to know my bank account status and I'm sure as hell not letting anyone else in there to check that I can afford groceries this week. Not while computer security is where it is. All of that is just begging to be attacked and broken and fixed and attacked and broken and fixed and so forth. I've had my debit card replaced five times in the last three years thanks to various hacks. It's a goddam headache is what it is. And now these geniuses want me to network _everything_ in my house? To what purpose?

Re:More to the point

[Anonymous Coward]

To what purpose?

Big data.

Re:More to the point

[JoeMerchant]

My "connected" thermostat has sensors in multiple rooms, that enables better control of temperature in the whole house, not just the one place the thermostat is mounted. Since installing it, we are more comfortable (less over and under cooling at various times), and the electric bill dropped enough to "pay back" cost of the thermostat and extra sensors in just about 1 year.

All in all, I think it is economically advantageous to us, and probably would be for most people in a typical 3+ bedroom house, especially if you have challenging variations in heat/cold distribution around the house.

Re:More to the point

[JanneM]

But what does the IoT bit have to do with it, though? Could just have a local thermostat with sensors in every room. Optionally passively download prices from your electricity provider if you want to get really fancy with scheduling your water heater.

You don't need the thing to be accessible over the net, and it most certainly doesn't need to synchronize all your usage data with some IT company somewhere. Our fridge manages to save us quite a lot of money simply by learning what time of day we're likely to open the door and when we are not, and adjust itself accordingly. No need for a connection of any kind.

"Things" are fine. "Internet of" is what people are questioning.

Re:More to the point

[Holi]

I want that, I just want the brains that control it to be in my house, not on some company's server who are taking all that data about my house and selling it to data miners.

Re:More to the point

[JoeMerchant]

Yep, if there were a non "cloud dependent" alternative, I would have bought that. At the moment, I needed these features and this is all that's on offer.

Spot me $25M and I'll start a company to develop a cloud free thermostat, it should be on the market within 48 months after the money hits my account - I promise.

Re:More to the point

[JoeMerchant]

I hear you, and agree - I really want mine to offer the connectivity it does and just "serve" the data from itself instead of the cloud - perhaps with an option where the company compensates me $5/month for sharing my data with them for as long as they are that interested in collecting it at that price.

"The cloud" makes configuration a little simpler for people who want to see what temperature it is in their house when they are away - but not enough to force the whole thing to run through their servers with no alternative to bypass them.

Re:More to the point

[Anonymous Coward]

>The fridge isn't going to know my bank account status and I'm sure as hell not letting anyone else in there to check that I can afford groceries this week.

WTF. If I have to worry whether I can afford groceries ("this week" or ever) I'd really reexamine my life.

Otherwise I agree.

Sort of wrong

[BlueCoder]

Yes it should refocus on servers. But rather it should be the "home server" market. Maybe even home cluster.

The market for desktops has mostly gone away, replaced by laptops and tablets. These people only needed a PC for spread sheets, simple word processing, and running a web browser.

The gaming machines market is the same if not bigger.

Professional content creation workstations are bigger than ever.

Same money is being spent; just in different form factors.

There is one area that isn't being exploited and marketed enough. The private cloud; i.e. home servers.

Particular when mixed with virtual machines it's something that needs to happen more. Store all your media at home in one place. Use the online cloud only for immediate stuff and for backups. There is huge potential for streaming. A home server can do transcoding on the fly and a dozen other things all at the same time.

It's more about the demise of the low end desktop which in the least has been replaces by cigar box systems.

Re:Sort of wrong

[davidshewitt]

It would be awesome to see an emphasis on home servers and server clusters. I run owncloud and I recently clustered it (two web server VM's, 3 mysql galera VM's; two pfSense/HAProxy VM's). All of this runs on hardware with Intel chips (one Xeon E3, one i7, one NUC - all Haswell). I might even get another server I have so many VM's. :-)

Unfortunately, the biggest barrier to entry for running a home server is the internet connection, not the hardware. Dynamic IP's can cause your cloud to be unavailable until dynamic DNS updates and the cache gets cleared. My encrypted offsite backups to dropbox slow my upload speed almost to a halt (and downloads from my server when I'm not on my network). I'm already paying way too much money for a static IP.

Re:Sort of wrong

[bluegutang]

A "home server" is just a PC. Servers in the "cloud" are optimized for extremely heavy workloads. But when a server serves just one home, the workload is super light, and any computer can handle it.

Re:Sort of wrong

[Anonymous Coward]

Tablets are fundamentally "consumption" devices. It's worrying to see desktop PCs being downplayed. They provide the seeds for so many programmers, engineers and so much future innovation.

Tablets and phones are fine for what they are... but they are mindless consumption devices. Laptops are slightly better, but not much.

Re:Sort of wrong

[thegarbz]

Only in the USA. In much of the rest of the world the home server is whatever users the least power, quite different from "a PC".

The home server is turning into an appliance. These appliances feature ARM quite heavily.

Re:Sort of wrong

[thegarbz]

Only in the USA. In much of the rest of the world the home server is whatever users the least power, quite different from "a PC".

The home server is turning into an appliance, powered by ARM.

Re:Sort of wrong

[Anonymous Coward]

Home server is backtesting the stock market. It needs to be beefy. But if one isn't privacy anxious it is more economical to move it public cloud.

Re:Sort of wrong

[drinkypoo]

Yes it should refocus on servers. But rather it should be the "home server" market. Maybe even home cluster.

There is no money in that whatsoever, because people expect to get those machines cheap, and they are not disappointed. There are craploads of tiny little computers which do that job already, and they are already cheap.

Dethrone Krzanich, quick!

[Anonymous Coward]

...

This Pretty Much Ensures AMD Is My Next Purchase

[zenlessyank]

My trusty i7 920 @ 3.8 Ghz was pretty much the plateau for Intel chips. Sure I could 'upgrade' to a 4.5 ghz for $1800 but really no reason to. I kept waiting for a 5 Ghz stock chip to upgrade to, but it never materialized. So, goodbye Intel. It was fun. Go ZEN.

lost before they started

[Gravis Zero]

Intel has already lost the embedded game in every way. Intel's chips suck too much power, cost too much and are too closed off when you compare them to the competition. Their obsession with x86 has made their failure a forgone conclusion.

those prices though

[Anonymous Coward]

It's like they don't want people to switch from AMD, with their CPUs from 4 cycles ago still being more expensive than AMDs. Haswell i5s should be under $200 by now. What's the hold-up?

Re:those prices though

[Rockoon]

Haswell i5s should be under $200 by now. What's the hold-up?

They havent actually gotten good yields from either of their last 2 shrinks and the next shrink isnt scheduled until the end of 2017. Meanwhile TSMC opens up a 10nm fab this year.

Intel is now as fragile as Motorola was.

Just like Edison

[Anonymous Coward]

Curious to see how many of those failed IA32 chips show up in their magical new IoT efforts.

Re:"Ubiquitous"

[Bob_Who]

Whomever smelt it dealt it...

Logic much?

[s.petry]

If I put a PC in a 4x24x22 frame, or a 8x12x12 frame, or a 2x2x2 frame, it's still a PC. Why? The easy but hard to fathom might be that the instruction set the chip uses, etc.. is all called PC. Probably more importantly, it's not a usable PC because of the form factor. It's usable because of the peripheral devices we use to access it. I'm still connecting an external FULL sized keyboard, some type of tracking device (mouse), and at least one reasonable sized (24") Monitor to be able to use the PC.

The PC being dead, as TFA claimed and you tried to back, is absolutely not the same thing as "We have a new form factor for a chassis".

Re:Logic much?

[LWATCDR]

I didn't say the PC was dead I said the phone would replace your PC. It is a bit different than just a new chassis since it will also be your phone and have long battery life and fit in your pocket. Replacing the PC does not mean that PC will go away completely. IBM still makes minicomputers and while the company formally known as SUN probably calls them servers they are closer to mini computers in many ways.
Tradition desktops will still be used by gamers, for CAD/CAM, and other HPC uses but I can see the phone running Windows 10 and USB 3.1 replacing many low end PCs and home PCs

Declaring the course of their death spiral

[Anonymous Coward]

Moving from de-facto monopoly position on expensive cpus to competing with dozens of companies producing cheapest possible crap for IoT devices. Yeah, that's a good plan.

Intel shifting focus away from PC business ..

[tetraverse]

"The chip maker also plans to focus on chips and technologies for IoT devices"

Before connecting 'things' to the Internet, how about fixing the defective MMU that comes in most WinTEL personal computers.

Re: "Ubiquitous"

[sexconker]

Ignore lists are found on just about every SJW site. Revently, Twitter and Reddit both added/extended tools for SJWs to ignore users who don't agree with them.

Re:This Pretty Much Ensures AMD Is My Next Purchas

[Anonymous Coward]

You think Ghz is the end-all, be-all measure of CPU performance? You're an idiot.

IPC has been creeping up, and new instructions add performance for the niches that require it.

AMD and Intel have both been subject to the same, modest performance gains.

And Zen isn't going to be a miracle. It *might* catch AMD up to Intel. (Which is enough for me... in case of ties, the underdog gets my money.)

But don't act like Zen is going to be a game-changer. With some preliminary leaks, it looks like a solid step forward, but it's not the huge leap AMD needs to compete with Intel in the performance segment or win the mainstream/mobile segments.

Re:This Pretty Much Ensures AMD Is My Next Purchas

[Anonymous Coward]

My trusty i7 920 @ 3.8 Ghz was pretty much the plateau for Intel chips. Sure I could 'upgrade' to a 4.5 ghz for $1800 but really no reason to. I kept waiting for a 5 Ghz stock chip to upgrade to, but it never materialized. So, goodbye Intel. It was fun. Go ZEN.

Socket 1366 Xeon 5650 / 5660 / 5675 / 5680 / 5690 is your friend. They cost $50-100 on eBay as datacenters shovel them out onto the surplus market.

1) Update BIOS with the 920 still in it.
2) Drop in 6-core Xeon
3) Reboot.
4) Enjoy 6 cores, slightly better overclockability (4.0 is trivial, 4.2 pretty easy, and some have hit 4.4-4.5 for daily use, but that's pretty much the limit of the technology regardless of which chip you start with), at slightly lower temperatures relative to the 920.

Cheapest upgrade ever. The X58/LGA1366 is the longest I've been on a platform. It was, indeed, a high point of Intel's development.

Re:"Ubiquitous"

[ITRambo]

Big words are hard, aren't they?

The one-trick pony tries again...

[Rob_Bryerton]

As we've seen time & time again, Intel is a one-trick pony. Granted, they do that one trick very well. But, in order to have any hope for success in any venture other than x86, they really need to spin off a separate company and allow it to run independently.

These large multi-billion $ corps are simply unable to innovate due to their corporate cultures and stifling hierarchies. All innovation happens with the small, nimble guys who, once they have a viable product, get gobbled up by the big guys.

Re: "Ubiquitous"

[MobileTatsu-NJG]

I don't think he heard you.

China will take over

[Anonymous Coward]

Now that AMD has signed over the X86 architecture to a Chinese state-owned company methinks, as Intel has decided to exit the PC scene, China is taking over

Might as well ... Intel is not providing us anything truly exciting for the past 10 years

AMD stock price

[Anonymous Coward]

And AMD's stock price just doubled.

Re:Hype

[Anonymous Coward]

If I could control MY Nest thermostat using MY computer(s), without having to use Google's servers, I might consider one. Likewise all the other xxxxx (with Cloud!) devices. Until then, IoT in the home for me is a non-starter. I.Don't.Need an internet-connected refrigerator, washer, thermostat, or even light switch.

Re:This Pretty Much Ensures AMD Is My Next Purchas

[Anonymous Coward]

My trusty i7 920 @ 3.8 Ghz was pretty much the plateau for Intel chips. Sure I could 'upgrade' to a 4.5 ghz for $1800 but really no reason to.

You're way out of touch with modern CPU performance. Modern CPUs have better performance per clock cycle, thanks in part to things like newer SSE2 instructions. Anyway, I've filtered https://cpubenchmark.net/common_cpus.html a bit for you:

• Intel Core i7-5820K @ 3.30GHz 12,987 $389.99 (33.3 passmark/$)
• Intel Core i7-4790K @ 4.00GHz 11,206 $339.99 (32.9 passmark/$)
• AMD FX-9590 Eight-Core 10,261 $224.95 (45.6 passmark/$)
• Intel Core i7-4790 @ 3.60GHz 10,019 $305.00 (32.8 passmark/$)
• AMD FX-8350 Eight-Core 8,945 $149.99 (59.6 passmark/$)
• Intel Core i7-920 @ 2.67GHz 4,994

I've included your CPU for performance comparison. It's your CPU only has half of the performance of a modern CPU that only costs $225 - $305. Also note that your i7-920 only supports up to 24GB of DDR3, so upgrading to a new CPU will also mean doubling your RAM speed.

A quick price search shows that you could double your machine's performance for about for $425 on NewEgg (FX-9590 + ASRock 990FX Extreme9 + 32GB DDR4). It should only cost you about $80 more to stick with Intel. You can probably reuse your old case / keyboard / monitor / mouse, and if your GPU won't fit in that motherboard, then you really need a new one anyway.

p.s. The FX-9590 is only 48% faster than your CPU in single-threaded tests, but the i7-4790 is 97% faster than yours in single-threaded.

Re:This Pretty Much Ensures AMD Is My Next Purchas

[ChrisMaple]

You've neglected to include the effects of a 3.8 GHz clock on the i7-920, and newer chips tend to be more resistant to overclocking. Your point stands, but the advantages of the newer chips are smaller.