Intel Supercharges Atom Chips With 16 Cores and Pro Level Features (pcworld.com)

← Back to Stories (view on slashdot.org)

Intel Supercharges Atom Chips With 16 Cores and Pro Level Features (pcworld.com)

Posted by msmash on Tuesday February 21, 2017 @08:00AM from the stepping-up-the-game dept.

Agam Shah, writing for PCWorld: Intel's Atom was mostly known as a low-end chip for mobile devices that underperformed. That may not be the case anymore. The latest Atom C3000 chips announced on Tuesday have up to 16 cores and are more sophisticated than ever. The chips are made for storage arrays, networking equipment, and internet of things devices. The new chips have features found mostly in server chips, including networking, virtualization, and error correction features. [...] A surprising feature in C3000 is RAS (reliability, availability, and serviceability) capabilities, which is mostly found on high-end Xeon chips. The feature corrects data errors on the fly and prevents networking and storage equipment from crashing.

77 comments

Min score:

Reason:

Sort:

Too many cores. by Anonymous Coward · 2017-02-21 08:10 · Score: 0

16 is about 12 too many. Most people can't wrap their brains around more than two, and most applications don't lend themselves to massive parallelization.
1. Re:Too many cores. by Anonymous Coward · 2017-02-21 08:17 · Score: 3, Informative
  
  These chips are designed for very parallelizable applications like file serving, managing multiple simultaneous VPN clients, etc. You want one core per NIC for these to make the best use of AESNI plus one or two more for management tasks.
2. Re:Too many cores. by MightyYar · 2017-02-21 08:19 · Score: 5, Funny
  
  Going from 16 to 4 would be a light version of the chip, and everyone knows that Cores Light is garbage.
  
  --
  W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
3. Re:Too many cores. by ProzacPatient · 2017-02-21 08:37 · Score: 1
  
  Individual applications don't but open your task manager and look at how many processes are running. The more cores you have the more processing power the operating system has to distribute all those processes, and their threads, across. Furthermore some graphics programs and game engines sometimes use upwards of a dozen worker threads.
  CPU power in general hasn't lept by great bounds in the last decade like it did 20 years ago when every new computer would be outdated in a year, so now the best strategy is to add more cores so each individual core isn't as burdened as it would otherwise be, and it will likely be the only strategy when we reach the limits of silicone based CPUs in the near future due to quantum tunneling unless someone comes up with something better. Stacking transistors could also be a possible solution sort of like a CPU skyscraper.
4. Re:Too many cores. by thegarbz · 2017-02-21 08:51 · Score: 2
  
  Most people can't wrap their brains around more than two
  Fortunately this is not a chip for most people, and boy are most people going to get upset if they look at their GPU.
  
  , and most applications don't lend themselves to massive parallelization.
  Actually a lot of the most computationally intensive tasks are embarrassingly parallel. But even if they aren't, who said we need to run one application per physical CPU? What version of DOS are you still running that doesn't allow multi-tasking?
5. Re: Too many cores. by Anonymous Coward · 2017-02-21 09:12 · Score: 0
  
  This is for server and embedded devices use, not desktop for consumer to tinker.
  Virtualization can use the cores and FreeNAS type solutions likely can, too.
6. Re:Too many cores. by Khyber · 2017-02-21 09:23 · Score: 2
  
  "You want one core per NIC for these to make the best use of AESNI plus one or two more for management tasks."
  One Core per NIC? Given the drastically reduced connectivity in the C3000 versus even desktop Pentium line processors in the form of fewer PCI-E lanes, good luck getting one PCI-E lane per NIC, let alone one core per.
  
  --
  Still waiting on Serviscope_minor to wake up to fucking reality and realize that Jessica Price isn't going to fuck him.
7. Re:Too many cores. by Blaskowicz · 2017-02-21 09:23 · Score: 1
  
  You might have a couple hundred tasks, threads or processes that sit there and collectively eat about 0.1% CPU time. Some are daemons waiting all day for you doing something that use them, some are GUI threads that similarly do nothing. They're like that guy who sits in a basement full of military archives and watches TV all day, only doing actual work for a couple minute once every week or month when someone comes to visit him.
8. Re:Too many cores. by SciFurz · 2017-02-21 09:33 · Score: 1
  
  They're like that guy who sits in a basement full of military archives and watches TV all day, only doing actual work for a couple minute once every week or month when someone comes to visit him.
  
  Yeah, but he's really going to need them when he has to save America 500 years later with toilet water so it's better to have them and not need them than not having them in the first place.
  
  --
  Write and/or read. https://scifurz.wordpress.com/
9. Re:Too many cores. by Anonymous Coward · 2017-02-21 09:34 · Score: 0
  
  What version of DOS are you still running that doesn't allow multi-tasking?
  5.0 :(
  I just don't trust that Double Space software in the newer version...
10. Re:Too many cores. by arglebargle_xiv · 2017-02-21 10:17 · Score: 1
  
  It's not just too many cores, it's too much everything. They've taken a crappy, underpowered chip that was trimmed to the bone to try and make something that competes with Arm, and are hacking on extras to make it sound more like a Xeon. In which case why not just use any non-Atom CPU, not necessarily a Xeon but just something that isn't as bare-bones as the Atom, and use that. Or an AMD G-series APU.
  Or could this be Intel's trick, that they've taken a Core 2 Mobile CPU, scraped off the Penryn label, reprinted it as Atom++, and are shipping those?
11. Re: Too many cores. by Anonymous Coward · 2017-02-21 10:50 · Score: 0
  
  -brought to you by CarlsJr.
  It's got what plants crave!
12. Re:Too many cores. by Anonymous Coward · 2017-02-21 11:11 · Score: 0
  
  16 shitty devs from 16 shitty countries writing 16 shitty single-threaded apps that don't have a clue about one another think they're the masters of the domain... I think 16 is just fine, thanks.
13. Re:Too many cores. by TeknoHog · 2017-02-21 11:23 · Score: 2
  
  You want one core per NIC for these to make the best use of AESNI plus one or two more for management tasks.
  While we're talking about massive parallelity, I want Knights with NI.
  
  --
  Escher was the first MC and Giger invented the HR department.
14. Re:Too many cores. by TeknoHog · 2017-02-21 11:48 · Score: 1
  
  They've taken a crappy, underpowered chip that was trimmed to the bone to try and make something that competes with Arm, and are hacking on extras to make it sound more like a Xeon.
  So it's like taking Pentium 3 and hacking on extras from Pentium 4 (the actual innovations around the core, not the GHz race) to make Pentium M, then putting several of these on a single die to make the Core series? Not a bad idea.
  
  Or could this be Intel's trick, that they've taken a Core 2 Mobile CPU, scraped off the Penryn label, reprinted it as Atom++, and are shipping those?
  I think this already happened a while ago, in a way. For instance, the original Atoms didn't have out-of-order execution, but the later ones do: https://en.wikipedia.org/wiki/... It looks a bit like the Pentium brand that lives on as the low end of Cores.
  BTW, I have one of the earlier in-order Atoms running happily in a server-ish machine where GPUs do all the heavy lifting. It's perfect for the job, and I guess more Atom cores would be great for a lot of server tasks, at least given enough I/O. Ideally, something like ARM or MIPS would probably be even better, but good luck finding (a) a suitable mobo with all the PCIe slots and (b) AMD/Nvidia binary drivers.
  
  --
  Escher was the first MC and Giger invented the HR department.
15. Re:Too many cores. by Tough+Love · 2017-02-21 12:30 · Score: 1
  
  So it's like taking Pentium 3 and hacking on extras from Pentium 4 (the actual innovations around the core, not the GHz race) to make Pentium M, then putting several of these on a single die to make the Core series?
  It's not like that. It's like inventing a new execution unit primarily intended to preserve high margins on core architecture chips by delivering substantially worse superscalar performance. Billed as power efficient, the latest generation with out of order execution still delivers worse performance at the low power end then celerons. It's a scam. Intel knows what to do to make Atom not suck: just throw it away and use core arch. But the charade is going to continue as long as Intel is able to avoid serious competition with its high margin parts.
  
  --
  When all you have is a hammer, every problem starts to look like a thumb.
16. Re:Too many cores. by Snard · 2017-02-21 15:02 · Score: 2
  
  Be careful asking for that. They may require a shrubbery.
  
  --
  - Mike
17. Re:Too many cores. by Anonymous Coward · 2017-02-21 17:14 · Score: 0
  
  The only reason to use this is ECC memory. That's it. It's as powerful as a i7 mobile part otherwise.
18. Re:Too many cores. by unixisc · 2017-02-21 18:37 · Score: 1
  
  16 is about 12 too many. Most people can't wrap their brains around more than two, and most applications don't lend themselves to massive parallelization.
  The thing I'm not getting is: is the Atom supposed to be a low power chip, or a low cost chip? I don't see how either of them is satisfied by blowing things up to 16 cores.
19. Re:Too many cores. by Anonymous Coward · 2017-02-21 19:31 · Score: 0
  
  Sounds like your stuck in a Bonnell world while these are silvermont and in general have no clue what you are talking about
20. Re: Too many cores. by Anonymous Coward · 2017-02-22 01:03 · Score: 0
  
  This is only true for those who can't find the details tab when pressing ctrl+shift+esc
21. Re:Too many cores. by Anonymous Coward · 2017-02-22 02:00 · Score: 0
  
  But the charade is going to continue as long as Intel is able to avoid serious competition with its high margin parts.
  
  Embedded/ultra low power CPUs is where Intel does have serious competition. It's the desktop arena where they have a virtual monopoly.
22. Re:Too many cores. by Tough+Love · 2017-02-22 02:08 · Score: 1
  
  But the charade is going to continue as long as Intel is able to avoid serious competition with its high margin parts.
  Embedded/ultra low power CPUs is where Intel does have serious competition. It's the desktop arena where they have a virtual monopoly.
  Yes, and they keep trying to push crappy Atom parts into that market segment to avoid cutting margin on their low power celerons, which are better than Atoms by every measure including power consumption. I thought I already said that?
  
  --
  When all you have is a hammer, every problem starts to look like a thumb.
23. Re:Too many cores. by Anonymous Coward · 2017-02-22 03:36 · Score: 0
  
  You may have done, but we're not talking about desktops here, we're talking about CPUs aimed at the low power markets. If you were talking about desktops, well, OK, but Intel isn't pushing Atoms at the desktop segment, so.... what's the issue?
24. Re: Too many cores. by Anonymous Coward · 2017-02-22 04:36 · Score: 0
  
  Especially desktop machines will benefit from this architecture:
  - one core to run firewall traffic
  - one core to send usability tracking data to Microsoft
  - one core for NSA backdoor server
  - one core for displaying adds 9that is free windows, right)
  - one core for Malicious Software Removal Tool and DRM violations scanner
  - one core for Hate Speech Monitoring Tool and other provided by Thought Police
  - one core for newest social app to update your status without delay
  - one core for IoT monitoring
  - 4 cores for AI assistant ...
  - one core for maleware
  - one core for bitcoin distributed data miner (without your knowledge)
  - one code for your solitaire game
  and one core to rule them all ..
25. Re:Too many cores. by lsatenstein · 2017-02-22 08:15 · Score: 1
  
  Going from 16 to 4 would be a light version of the chip, and everyone knows that Cores Light is garbage.
  I never drank Coors light.,
  
  --
  Leslie Satenstein Montreal Quebec Canada
26. Re:Too many cores. by lsatenstein · 2017-02-22 08:28 · Score: 1
  
  DIrect comparison (mips * clock frequency) is not a true measure.
  There was a posting that doubted AMDs performance against equivalent Intel chips. The author used a frequency * mips rate to indicate throughput.
  Here are some explanations about the difference between Ryzen and older CPU chip techolologies.
  How to explain the difference? Here is my take. First of all, a 3 ghz frequency means that there are usually 6gig of clock ticks. Some instructions take multiple clock ticks. When cpus are implemented on 14nm cell sizes, there is room within the die for many many more cells. Cells that could be used to do things better.
  For example , suppose that a “long integer” multiply/divide requires 20+ clock ticks with one model CPU, and because of design improvements and cell size reductions, the same operation on the Ryzen may only require 16+ clock ticks. That instruction’s execution is improved by 20%.
  Consider parallel operations within the cpu chip. To achieve more parallelism, you need to allocate more logic to the multiplier/divider circuitry. And that is probably what has happened. (Smaller micro-circuit sizes allows more logic space available within the CPU die).
  You bought a 16core cpu chip. Ask yourself how many internal (additional reserved) cores are present within the chip for the chip to do it’s work? These reserved cores could be working in parallel to implement an instruction, thus completing a complex instruction in fewer clock ticks. (AES encrypt)
  Go through all the instructions that can be optimized by smarter circuitry, and you have the explanation why the AMD chips are more performing than ever.
  If there are more logic elements, then consider that a “register” shift instruction could be optimized to three or four clock ticks, irrespective of the shift amount. Typically, the shifter hardware moves the bits, according to a specified amount.
  I am willing to bet that chip for chip, the AMD has many more transistors and gates for logic to support parallel sub-instruction processing than found in older designs. More parallelism used to support fewer clock-ticks.
  Instructions within a CPU are also within a pipeline queue. I do not know the chip internals, but there may be up to 10+ instructions in the input queue that are at various stages of being decoded. The queue is flushed if an interrupt instruction is received.
  Taken all together, fewer clock ticks to decode an instruction, perhaps saving some of the queue contents during an interrupt may be the major reason the AMD chips even with slower clock frequencies are faster than older chip designs. Older designs need to boost clock speeds by up to 25% over the AMD chips, to arrive at par.
  There was a posting that doubted AMDs performance against equivalent Intel chips.
  Here are some additional explanations about the difference. How to explain the difference ?
  Here is my take. First of all, a 3 ghz frequency means that there are usually 6gig of clock ticks. Some instructions take multiple clock ticks. When cpus are implemented on 14nm cell sizes, there is room within the die for many many more cells. Cells that could be set aside to do things better.
  For example , suppose that a “long integer” multiply/divide requires 20+ clock ticks with one model CPU, and because of design improvements and cell size reductions, the same operation on the Ryzen may only require 16+ clock ticks. That instruction’s execution is improved by 20%.
  Consider parallel operations within the cpu chip. To achieve more parallelism, you need to allocate more logic to the multiplier/divider circuitry. And that is probably what has happened. (Smaller micro-circuit sizes allows more logic space available within the CPU die).
  You bought a 16core cpu chip. Ask yourself how many internal (additional reserved) cores are present within the chip for the chip to do it’s work? These reserved cores could be working in parallel to implement an in
  
  --
  Leslie Satenstein Montreal Quebec Canada
27. Re: Too many cores. by Anonymous Coward · 2017-02-22 15:09 · Score: 0
  
  Lol, at first I thought "shrubbery" was some kind of technical term.
28. Re:Too many cores. by Anonymous Coward · 2017-02-24 02:23 · Score: 0
  
  Neither of you are talking about desktops so why keep bringing it up? All the dude is saying is that Intel is just doing its Intel thing of undercutting markets it doesnt dominate with shit.
Next up by DontBeAMoran · 2017-02-21 08:14 · Score: 1

Apple switching to 16-cores A16M ARM processors for their new Macs.

--
#DeleteFacebook
1. Re:Next up by ckatko · 2017-02-21 08:16 · Score: 1
  
  More like Apple switching to the "Brand new" i3 processor for their newest generation of outdated powerbooks.
2. Re:Next up by unixisc · 2017-02-21 18:40 · Score: 1
  
  Apple switching to 16-cores A16M ARM processors for their new Macs.
  This would make a lot of sense. Is the A16M from their A line of CPUs? They might as well use the same CPU in their Macs as they use in iPhones, so that their Macs can leverage iPhone apps. There is no compelling reason for Macs to be x64 (the way there is for Windows)
3. Re:Next up by Anonymous Coward · 2017-02-21 18:59 · Score: 0
  
  Their slogan is less is more, so they are just doing what people want. Their customers can't handle a two button mouse, why would they be able to harness any real CPU power either?
4. Re:Next up by DontBeAMoran · 2017-02-22 01:35 · Score: 1
  
  AFAIK they're up to "A10" now, I just imagined a new model number that was "16" as in sixteen cores and I added M at the end for Mac.
  But yes, increasing their profits (Tim Cook would never lower Mac prices, even though he doesn't seem to care about Macs at all) could be done by using their own ARM CPUs instead of paying Intel for their overpriced CPUs.
  Tim Cook seems to think an iPad Pro can replace a computer, so the next logical step is to replace the computer hardware with iPad hardware.
  Bonus points if they put multiple ARM CPUs or custom ARM CPUs with more cores into their Macs.
  
  --
  #DeleteFacebook
5. Re:Next up by unixisc · 2017-02-22 02:43 · Score: 1
  
  Profits would be one - and the main - thing definitely, but it would also give Apple a simpler ecosystem to maintain. It would be them doing what Microsoft tried in Windows 8, but better: merge the OS X and iOS code bases to the binary level, and eliminate the need for fat binaries. It would also make available to the Mac just about every app that's available for iOS. While some of them - such as Vonage or Lyft - maybe unsuitable for Macs, others, such as WhatsApp, could be very useful - enabling people at home to type their messages in the comfort of their laptop instead of struggling w/ their Mac keyboards
6. Re:Next up by MightyYar · 2017-02-22 03:32 · Score: 1
  
  Apple could get most of those benefits by "simply" requiring submissions to the App Store be universal binaries compiled for Intel.
  For me, the availability of Windows in virtualization trumps the ability to run iOS apps outside of virtualization. Frankly, most of them would be bizarre on a huge laptop screen - like viewing a mobile website on a desktop computer. The only thing keeping me on Mac is its "universal" status. I can run Windows, Mac, various unixes. I can compile X11 apps, develop for iOS or Android, etc, etc. A move to ARM can only happen if Windows becomes available for ARM, at least for my purposes.
  
  --
  W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
With a new improved higher memory limit of 17GB!! by Anonymous Coward · 2017-02-21 08:25 · Score: 0

assholes... gimmick bars instead of touch screens... less memory than other computers.... IO port musical chairs... EOL perfectly fine computers via OS upgrades...
How about bout' just fix your fucking OS and browser :/
Re:With a new improved higher memory limit of 17GB by Anonymous Coward · 2017-02-21 08:32 · Score: 0

Who the hell actually uses touch screen laptops? I can't figure out the point.
Sounds Nice by youngone · 2017-02-21 08:32 · Score: 1

These chips sound great.
I have just finished building a pfSense firewall/proxy/router etc based on a PC Engines APU SOC. This has 4 AMD cores and 4 GB of RAM and is plenty powerful enough for what I need.
I guess 16 cores with all those extra bells and whistles would be nice for bigger customers than mine, but to be honest, the box I put together is plenty good enough, and the price was right.
Maybe the SME type market is not where Intel wants to play.
1. Re:Sounds Nice by Anonymous Coward · 2017-02-21 12:53 · Score: 0
  
  Dunno what kind of traffic you are looking to route but that seems like an overkill. Or is it, AMD APUs are pretty bad at bandwidth actually?
  My pfSense box back in the late 2000s had crappy Realtek cards PCI cards (regular PCI) for WAN (replaced them after every other lightning storm, even after the surge protector) and a good server 3COM with Intel PCI card for LAN and a Pentium 233 MMX (overclocked to 250Mhz and slightly overclocked PCI) and 256 Mb of memory and was running a good 100Mbps load-balancing per connection and failover over 2 WAN connections to different providers after tuning IRQs and interrupts a bit. For a home user that was more than enough especially since the hardware was bought in 1997 and continued operating well into 2012. It is currently stored in my grandma's attic, and most likely still in working condition but got replaced with a $50 Mikrotik that uses a smidge of the power.
2. Re:Sounds Nice by Agripa · 2017-02-24 02:54 · Score: 1
  
  Dunno what kind of traffic you are looking to route but that seems like an overkill. Or is it, AMD APUs are pretty bad at bandwidth actually?
  
  Using an AMD APU for this application *is* overkill but they are inexpensive, their motherboards are inexpensive, and they draw a reasonably low amount of power.
Finally ECC-RAM by Anonymous Coward · 2017-02-21 08:37 · Score: 1

Can we have this in desktop processors too now, please? Bit errors in RAM are an underestimated source of data corruption and crashes.
1. Re: Finally ECC-RAM by Anonymous Coward · 2017-02-21 10:20 · Score: 0
  
  That would break the gap between consumer desktop systems and Xeon processor workstations. ECC memory is one markup point, then there is the clock speed, cache size, number of cores and number of sockets.
2. Re:Finally ECC-RAM by Blaskowicz · 2017-02-21 10:37 · Score: 1
  
  I'd like if the voltage sensors worked under linux. I wanted to blame libsensors but I was told to blame desktop motherboards. You can run memtest once in a while as a weak alternative to ECC on a desktop but you can't diagnose the power supply without Windows.
3. Re: Finally ECC-RAM by Lost+Race · 2017-02-21 16:16 · Score: 1
  
  Core i3, Pentium, and Celeron chips all support ECC. Good luck finding a motherboard with all 72 data pins wired though.
4. Re:Finally ECC-RAM by l20502 · 2017-02-22 00:58 · Score: 1
  
  Sometimes is just a matter of downloading sensors-detect from git and running that instead of the packaged version.
  Also on my current motherboard both the manufacturers windows utility and lm-sensors show mostly bogus values because they read raw values, only the bios setup screen shows the correct values.
5. Re:Finally ECC-RAM by Blaskowicz · 2017-02-22 07:40 · Score: 1
  
  Wow, running "vanilla" sensors-detect did work, I think I had not tried that for a looong time. (Motherboard is old). You have to understand what the hell it displays and prompts you. It gave me the name of a kernel module for the Super I/O chip, you load it and everything is there albeit the voltage are almost all wrong, like they think they're measuring 3.3 volts voltages. One is at "0.10 volt", maybe that's the mysterious -12V. First one seems to be an accurate vcore, second one maybe memory controller voltage or such. Mobo temps are accurate, fan speeds are there. Thanks.
  Too bad e.g. +5V and +12V aren't labeled and aren't showing properly calibrated values (they'd do with Windows and Speedfan). They're unlabeled and fluctuate around "+3.2V". I don't know if there's a good reason for that - does wiring depend on motherboard? Anyway, it's very close to everything working, but with fairly nonsensical values still. Only lacking a small final step. So I can understand the monitoring is not enabled by default (save for CPU temp and GPU temp, which are separate and consistent across different machines, so they're enabled by default)
  So there's my little report, however boring it might be.
6. Re: Finally ECC-RAM by Agripa · 2017-02-24 02:59 · Score: 1
  
  Core i3, Pentium, and Celeron chips all support ECC. Good luck finding a motherboard with all 72 data pins wired though.
  They are not that difficult to find. Look for LGA1150 and LGA1151 workstation/server motherboards which use C series (or I guess the X99?) south bridges.
7. Re: Finally ECC-RAM by Lost+Race · 2017-02-24 04:38 · Score: 1
  
  I've had pretty good results with the Asrock H87WSA-DL "workstation" board. Very reasonably priced compared to "server" boards.
McFillem by Tablizer · 2017-02-21 08:44 · Score: 1

They'll just fill up more cores with @#&$% McAfee scans; they own half the company. It's like a dog breeding company also selling pooper-scoopers.

--
Table-ized A.I.
1. Re:McFillem by Anonymous Coward · 2017-02-21 19:09 · Score: 0
  
  And they are also truing to sneak that mcafee crap via every Adobe flash update, just in case user forgots to uncheck a checkbox in download page. No wonder why their software did not sell, as it seems to be designed for eating CPU cycles and making PC slow in a hope that user would buy a new computer.
Re:Sounds Nice - C2000 "fix"?? by Anonymous Coward · 2017-02-21 08:46 · Score: 0

Be careful of what you ask for - see this regarding the gear, including Cisco's, with the prior gen C2000 Atom's bricking themselves after 18 months:
https://www.theregister.co.uk/2017/02/06/cisco_intel_decline_to_link_product_warning_to_faulty_chip/
RO
UltimateTarget for Atom: Cars and Trucks by Anonymous Coward · 2017-02-21 08:49 · Score: 5, Insightful

They'd like a spot at the table of the massive amount of chips that will be needed in autonomous cars and other AI driven machines. Parrelization works very well in processing data from lots of sensors. I suspect this is just the beginning.
1. Re:UltimateTarget for Atom: Cars and Trucks by Anonymous Coward · 2017-02-21 12:14 · Score: 0
  
  I could be wrong, but I think GPUs are better suited for most of the processing required in those applications.
2. Re:UltimateTarget for Atom: Cars and Trucks by Anonymous Coward · 2017-02-22 01:27 · Score: 0
  
  No salesperson is going to say that to the shareholders or potential customers.
3. Re:UltimateTarget for Atom: Cars and Trucks by Anonymous Coward · 2017-02-22 03:19 · Score: 0
  
  Who wouldn't? Automotive is a multi-billion dollar market.
  However, I've recently looked at Intel's autonomous car activities. They're run out of Karlsruhe (Germany), and they're recruiting pretty aggressively, but I'm not exactly impressed. They seem to have a preference for FPGA's. On the other hand, you have NVidia with the Tegra line, and they're getting more traction. Qualcomm/NXP is another competitor with a more believable pitch; NXP was already a chief supplier to the Automotive sector and Qualcomm adds valuable wireless technology.
Imagine these in a beowulf cluster (n/t) by Anonymous Coward · 2017-02-21 08:49 · Score: 0

Really, wouldn't that be awesome - to finally beat those old overclocked celerons from 1998!
The Real Question: by Anonymous Coward · 2017-02-21 09:01 · Score: 1

Will these last longer than 18 months?
Thank you AMD by Anonymous Coward · 2017-02-21 09:07 · Score: 0

We might finally see Intel put out some CPUs with more than a 5% increase in performance and for less than a kidney.
Re:With a new improved higher memory limit of 17GB by Anonymous Coward · 2017-02-21 09:13 · Score: 0

So its a laptop and a tablet combined. Surface Pro is probably the only one that's been done right though...
Re:With a new improved higher memory limit of 17GB by Anonymous Coward · 2017-02-21 09:15 · Score: 0

Police do. Toughbooks mostly.
Re:With a new improved higher memory limit of 17GB by Blaskowicz · 2017-02-21 09:27 · Score: 1

I bet there are some people who have a touch screen laptop but don't know the screen can be touched.
Netbook and this please. by Anonymous Coward · 2017-02-21 09:28 · Score: 0

This on a netbook with the virtualization especially would be great. Hell, laptops too.
Just stop with the fucking annoying graphics card hijinks so I can use Linux 4.10 virtual GPU to share it properly.
1. Re:Netbook and this please. by ctilsie242 · 2017-02-21 09:58 · Score: 4, Interesting
  
  I would be happy with a mini rack/blade system about the size of a HP MicroServer, with the ability to add 4-8 of these CPU boards in it as blades, each board having a SSD big enough to load a Linux distro or ESXi... and one has a nice CPU farm that doesn't take much space, but can run a lot of lightweight VMs very economically.
2. Re:Netbook and this please. by Anonymous Coward · 2017-02-21 14:25 · Score: 0
  
  You can pretty much already do that with Intel's Xeon D processors: https://www.supermicro.com/products/nfo/Xeon-D.cfm
Re:With a new improved higher memory limit of 17GB by Anonymous Coward · 2017-02-21 09:29 · Score: 0

I've seen people who badly need them but either don't know it or don't know enough to find a cheap enough option they can use. Specifically people scoring cricket sitting at an outdoor table, using a laptop to control the scoreboard where they have to click click click using a mouse and spending seconds they don't really have hundreds of times a day playing "I can't see the mouse pointer" when with a touchscreen they could just tap what they want. And no, they can't fix the rubbish software and just use a keyboard! A tablet on a stand might be an acceptable substitute but getting one with a real usb port is going to blow the budget just as quickly and probably be less stable and more hassle anyway.
Re: With a new improved higher memory limit of 17G by Anonymous Coward · 2017-02-21 10:18 · Score: 0

I think it's the other way round now. Users press everything with glass and a picture to see if it is a touchscreen or not.
If anything needs to be a touchscreen it is 3D monitors.
Re:With a new improved higher memory limit of 17GB by avandesande · 2017-02-21 10:18 · Score: 1

No problem with touch screen laptops as long as they include a bag of cheetos.

--
love is just extroverted narcissism
But the Atom is still only 25% of the power by Anonymous Coward · 2017-02-21 11:34 · Score: 0

Hate to break it to anyone looking at those 16 cores thinking they are getting 16 real CPU cores. Just like it's predecessor the 8-core model, if you go look at performance benchmarks you will find that the core's are only 1/4 as powerful. The aggregate performance of a 16 core Atom is the same as a i7 quad-core mobile part.
I suspect that Intel just puts one ALU on each core instead of the three integer ALU's available on an i7. (When in Hyper threaded mode, only one ALU is given to each core, and the first thread of each core has the rest of the core's logic while the second virtual core only has the second ALU and nothing else.)
The other problem is that 16-cores don't help you very much unless the underlying software is designed to take advantage of it. So this might work out fine as a file/web server running ZFS on FreeNAS, but it's utterly not useful running Windows or Linux as a file server because Windows is too top-heavy to do file-serving (database connection handling on the other hand is good) and Linux software is afraid of threads, due to wanting everything to run in user-space, so even ZFS is useless on Linux. Linux doesn't like to use hardware acceleration features because it breaks the user-space sandbox, so a Linux file server will be kneecapped by the network card drivers.
Ultimately the primary use for "Atom" performance devices are meant to be x86-64 Windows/FreeBSD competition for the ARM chips running 32-bit Linux in NAS devices. Unfortunately the vast majority of NAS devices are rubbish-tier SoC's, and if the power supply dies in the device, you've lost ALL your data. At least in an x86-64 PC-based system you can replace the parts with off-the-shelf parts. I keep wanting to buy one of these, but I look at the price/performance and realize that the energy costs alone of running one of these instead of 6 external drives doesn't look good.
If you want to use these for file servers, they are a better choice as long as you have 3+ computers that use it. Otherwise you are better off with 3 USB-interfaced external drives that can do 480Mbps each instead of one file server that can do 1Gbps max, regardless of the drives. The real problem is that only FreeNAS really is of any use with these, due to caching and reliability. Most Linux-based and Windows-based NAS systems are totalled if you have to pull the drives, because the setup data is immutable, so replacing drives > 1TB is unreliable, you need to have 6-drive (2x3 or 3x2) RAID configurations if you're not using ZFS to get the kind of reliability you get from 3 drives in ZFS.
A whinging bunch of nerds who have no idea by Anonymous Coward · 2017-02-21 11:39 · Score: 0

These are chips for appliances - storage, network, nvr, etc.
These are NOT designed for people to use in desktops.
So who gives a fuck if the cores are slow compared to a Pentium d? These things can saturate 10gbps network, and are meant for low/medium level IO without costing heaps.
You idiots are fucking clowns who need to grow the fuck up.
Re: With a new improved higher memory limit of 17G by Anonymous Coward · 2017-02-21 12:10 · Score: 0

Doctors and medical professionals.
Re: With a new improved higher memory limit of 17G by Anonymous Coward · 2017-02-21 12:19 · Score: 1

Virtual breasts, right?
bugfixes ? by sxpert · 2017-02-21 20:38 · Score: 1

hope they fixed the clock issues plaguing the C2000 !
NEVER too many cores ! Re:Too many cores. by Anonymous Coward · 2017-02-22 02:49 · Score: 0

What utter nonsense.. MORE cores and the memory to support them mean xen or vmware can run that many more VM(s), containers, threads etc..and pin certain OS(s)/Applications to certain sets of cores for more performance... works great for the sloppy admin style of running the machine I see often among users ie running 32 apps all CPU and memory hogs all open and bitching about why performance is sucky on a 2-4 core...
LOTS of us out here running personal/private clouds..
anon
Re: With a new improved higher memory limit of 17G by Anonymous Coward · 2017-02-22 22:21 · Score: 0

Right.