Subject: General Tech | April 20, 2015 - 01:17 PM | Jeremy Hellstrom
Tagged: amd, seamicro, HPC
Just over three years ago AMD purchased SeaMicro for $334 million to give them a way to compete in HPC applications against Intel who had recently bought up QLogic and the InfiniBand interconnect technology. The purchase of SeaMicro included their Freedom Fabric technology which was at that time able to create servers which could use Atom or Xeon chips in the same infrastructure. AMD developed compatibility with their existing Opteron chips and it was thought that this would be a perfect platform to launch Seattle, their hybrid 64bit ARM chips on. Unfortunately with the poor revenue that AMD has seen means that the SeaMicro server division is being cut so they can focus on their other products. Lisa Su obviously has more information that we do on the performance of AMD but it seems counter-intuitive to shut down the only business segment to make positive income, but as The Register points out the $45m which they made is down almost 50% from this time last year. AMD will keep the fabric patents but as of now we do not know if they are looking to sell their server business, license the patents or follow some other business plan.
"Tattered AMD says it's done with its SeaMicro server division, following a grim quarter that saw the ailing chipmaker weather losses beyond the expectations of even the gloomiest of Wall Street analysts."
Here is some more Tech News from around the web:
- AMD: Windows 10 will launch in late July @ The Inquirer
- Exchange Server 2016 will be mostly Cloud Exchange ported back on-premises @ The Register
- MSI Preparing SFF-8639 Adapter Card for Motherboards @ Kitguru
- TSMC to open new fab for 10nm manufacturing in mid-2016, says chairman @ DigiTimes
- GNOME 3.16: The Sleekest Linux Desktop to Date @ Linux.com
- Trials & Tribulations: Installing Gentoo Linux With GNOME & Systemd @ Techgage
- Tech ARP 2015 Mega Giveaway
- NikKTech & FSP Group Spring Worldwide Giveaway
Subject: General Tech | April 20, 2015 - 08:00 AM | Scott Michaud
Tagged: windows 10, microsoft
According to WinBeta, the internal builds of Microsoft's Windows 10 operating system have jumped from the 10060s-range to the 10100s. This mirrors the activity before January's consumer event, which led to the release of 9926. What this likely means is that Microsoft has forked internal development ahead of their BUILD 2015 conference, which takes place between April 29th and May 1st.
I expect that they will release a new, highly-tested build in the 1006x-1008x range to both Fast and Slow rings, like 9926 was, at or around the time of the event. Meanwhile, new and experimental features will land on the 10100 branch. The interesting question is: when will we see that later fork?
If Microsoft dedicates themselves to rapid releases, it might not be too long for users in the Fast ring, or a faster-than-Fast ring that they could potentially announce at the event. With the visibility of BUILD, it would be a good time for them to shake up their release cycle. They really cannot afford to relax quality control any more significantly than they did with 10041 without assurance that Insiders get the message. The journalist attention of the conference would likely do it though.
Alternatively, the released build might be classified as a developer preview that is expected to stick around for a while. If I needed to guess though, I doubt it. As stated earlier, it will probably be a highly QA-tested build for Slow ring users, but I see little reason for Microsoft to throttle down the more enthusiastic users. When 10049 was the last build for Fast users, you could say that they were not wanting to overshadow BUILD. That obviously does not apply after the conference is over, and I cannot see anything else further on the horizon.
That is, of course, unless they are getting cold feet about releasing not-fully-baked builds to the public.
Subject: General Tech, Graphics Cards | April 20, 2015 - 07:30 AM | Scott Michaud
Tagged: Red Hat, Khronos
With a brief blog post, Red Hat has announced that they are now members of the Khronos Group. Red Hat, one of the largest vendors of Linux software and services, would like to influence the direction of OpenGL and the upcoming Vulkan API. Also, apart from Valve, they are one of the only Linux vendors that contributes to the Khronos Group as an organization. I hope that their input counter-balances Apple, Google, and Microsoft, who are each members, in areas that are beneficial to the open-source operating system.
As for now, Red Hat intends to use their membership to propose OpenGL extensions as well as influence Vulkan as previously mentioned. It also seems reasonable that they would push for extensions to Vulkan, which the Khronos Group mentioned would support extensions at GDC, especially if something that they need fails to reach “core” status. While this feels late, I am glad that they at least joined now.
Subject: General Tech, Graphics Cards, Processors | April 19, 2015 - 02:08 PM | Scott Michaud
Tagged: moores law, Intel
While he was the director of research and development at Fairchild Semiconductor, Gordon E. Moore predicted that the number of components in an integrated circuits would double every year. Later, this time-step would slow to every two years; you can occasionally hear people talk about eighteen months too, but I am not sure who derived that number. In a few years, he would go on to found Intel with Robert Noyce, where they spend tens of billions of dollars annually to keep up with the prophecy.
It works out for the most part, but we have been running into physical issues over the last few years though. One major issue is that, with our process technology dipping into the single- and low double-digit nanometers, we are running out of physical atoms to manipulate. The distance between silicon atoms in a solid at room temperature is about 0.5nm; a 14nm product has features containing about 28 atoms, give or take a few in rounding error.
It has been a good fifty years since the start of Moore's Law. Humanity has been developing plans for how to cope with the eventual end of silicon lithography process shrinks. We will probably transition to smaller atoms and molecules and later consider alternative technologies like photonic crystals, which routes light in the hundreds of terahertz through a series of waveguides that make up an integrated circuit. Another interesting thought: will these technologies fall in line with Moore's Law in some way?
Subject: General Tech | April 17, 2015 - 07:00 AM | Scott Michaud
Tagged: windows xp, windows, microsoft, google, EoL, chrome
It has been a year since Microsoft cut off extended support for Windows XP including Internet Explorer security updates for the platform. Yeah, I know, it doesn't feel like it. Other browser vendors announced that they would continue to target the retired OS after Microsoft washed their hands of it. At the time, Google said they would give at least 12 months support, which brings us to yesterday.
Now Google is extending their commitment to the end of the year. They did not say that it was a hard deadline for their customers, but they also did not add an “at least” qualifier this time. The browser vendor wants people to upgrade and admits that they cannot genuinely provide a secure experience if a known issue bites everyone at the OS level. You can keep training the guard at the door, but if your window falls out, mind the pun, then it is still dangerous to be inside.
Granted, we have not seen a major attack on XP over the last year. You would have to think that, even if the attacks are sophisticated, some of the victims would have noticed and reported it to someone. Still, I wonder how it keeps surviving, especially since I would have thought that at least one vulnerability in the last twelve Patch Tuesdays could be ported back to it.
Maybe it is too small of a target?
Subject: General Tech | April 16, 2015 - 05:21 PM | Scott Michaud
Tagged: windows 10, virtual desktops, microsoft
Virtual Desktops is one of the largest interface changes for Windows 10. It is an organizational feature, which allows users to group relevant windows together. This puts them out of sight when you don't want them but also easily accessible when you do, and with the programs that you would want to use alongside them.
Microsoft is wondering about one specific area though: should running applications be visible in other desktops? Some believe that a virtual desktop should only show icons for applications in the current context, and alt+tab should behave the same. It makes sense, because otherwise you're exposing the clutter in the taskbar and alt+tab, which are some of the most visible places. Others want everything to be visible at all times, or alt+tab-able to at all time. I guess this is to reduce the clutter on the desktop, without touching the clutter otherwise?
Microsoft has stated clearly that both cases will be user-selectable because there is enough usage on both sides. A user can have virtual desktops affect the taskbar or not, or alt-tab or not, independently. The question they are trying to solve right now is “What should the default be?”
If you are a Windows 10 Technical Preview user, and you care about the adoption of Virtual Desktops, Microsoft might push you into one or the other camp. Later, they will pop up a notification to ask your opinion once they feel you have used it long enough.
This is one area that your vote will influence Windows 10 in a very specific way.
Subject: General Tech | April 16, 2015 - 04:08 PM | Scott Michaud
Tagged: ea, free to play, battlefield
Battlefield Heroes, Battlefield Play4Free, Need for Speed World, and FIFA World are set to disappear in three months. According to the site for one of these games, Battlefield Heroes, that corresponds to Tuesday, July 14th, 2015. The other three might be plus or minus a little bit, but it sounds like it will be a simultaneous shutdown. Refunds will not be provided, even if you have premium credit remaining unspent on your account. I guess that means that you should burn whatever you have left over while the servers are still running; EA will not accept new accounts or payments though.
The announcement, made by Executive Vice President of EA Studios, Patrick Söderlund, assures players that other EA free-to-play titles are growing in player base and content. He uses Star Wars: The Old Republic as an example, which will receive updates over the year, including ones that build on its narrative.
But this is one of the problems with “free-to-play” and other genres with an authoritative online component: timed life. These titles are not going to be critiqued and celebrated in University courses over the upcoming decades and centuries. It is a difficult balance. On the one hand, it is interesting to create a universe that evolves and possibly mutates based on player interaction. Those experiences disappear though (and so does the money you tie into it). That doesn't mean “don't play them” -- just be aware what you're investing your time, money, and self-expression into.
Subject: General Tech | April 16, 2015 - 02:36 PM | Ken Addison
Tagged: x99-soc, video, Skylake, Samsung, podcast, nvidia, msi, motorola, Moto E, Intel, GTAV, gs30, gigabyte, Broadwell, amd, 840 evo
PC Perspective Podcast #345 - 04/16/2015
Join us this week as we discuss the MSI GS30 Shadow, Gigabyte X99-SOC, Skylake Leaks and more!
The URL for the podcast is: http://pcper.com/podcast - Share with your friends!
- iTunes - Subscribe to the podcast directly through the Store
- RSS - Subscribe through your regular RSS reader
- MP3 - Direct download link to the MP3 file
Hosts: Jeremy Hellstrom, Josh Walrath, Allyn Malventano, and Sebastian Peak
Program length: 1:20:07
Subject: General Tech | April 16, 2015 - 01:47 PM | Jeremy Hellstrom
Tagged: nvidia, maxwell, Nouveau, open source
The initial benchmarks that Phoronix and other Linux tech sites ran on the new Maxwell cards from NVIDIA were using the proprietary binary drivers, the same as with all AMD cards. Unlike AMD who have always released signed binary-only firmware blobs which could not be reverse engineered and modified for use, previous generations of NVIDIA cards did not require signed firmware images, only a lot of dev work. Maxwell is locked down and the current open sourced Nouveau driver can now only be used to set up display outputs, it has no ability to use the card for hardware accelerated graphics.
NVIDIA states that the new requirement is to prevent shady characters from modifying slower cards to look like new Maxwell GPUs but Phoronix feels that they have gone overboard. It is sad to see NVIDIA taking a step backwards in supporting the open source community, mimicking AMD's procedure of only offering binary-only firmware blobs though AMD is at least updating the blobs and open source driver relatively frequently. That said, the new Maxwell cards do perform very well with the proprietary Linux driver so users should not feel they have to avoid NVIDA; unfortunately for developers the same is not true.
"While NVIDIA's new GeForce GTX 900 series is dominating for Linux gamers with excellent performance with their $1000+ GPU as well as great Linux OpenGL/OpenCL performance out of their lower-cost GPUs with excellent power efficiency, that's only when using the proprietary driver... NVIDIA's newer GTX 900 / Maxwell hardware is less open-source friendly than their previous generations of hardware. "
Here is some more Tech News from around the web:
- iFixit: Apple's 12in MacBook is near impossible to repair @ The Inquirer
- Talk is FINNISHed: Nokia to buy Alcatel Lucent for €15.6 billion @ The Register
- Businesses are not using apps from the Windows Store @ The Inquirer
- Calling Out a GAO Report That Says In-Flight Wi-Fi Lets Hackers Access Avionics @ Slashdot
- Kwikset SmartCode 916 Electronic Deadbolt @ Benchmark Reviews
- Win an MSI GS60 Ghost Pro laptop @ KitGuru
Subject: General Tech, Storage | April 16, 2015 - 06:47 AM | Ken Addison
Tagged: uefi, SSD 750, PCI-E 3.0, NVMe, Intel, ACHI, 750 series
UPDATE: ASUS has pointed us towards a poll they are running to gauge what platforms people are most anxious for NVMe Boot support on. So if you have an ASUS board and are interested in buying an Intel SSD 750 Series, head to their poll to voice your opinion!
Last week, the Intel 750 Series SSD was unveiled the the public as the first consumer SSD to feature the NVMe or Non-Volatile Memory Express interface. NVMe was designed from the ground up for flash storage, and provides significant advantages in latency and potential top transfer rates over the aging AHCI standard. Check out our review of the Intel SSD 750 Series to find out why this is such an important step forward for storage technology.
Even if you aren't necessarily concerned about the merits of a new storage interface, the throughput numbers from the 750 series are hard to ignore. With peak speeds over 2.5GB/s read and 1.5GB/s write, it's hard not to be interested in this new drive.
However, all this new speed doesn't come without a few complications. NVMe is an all-new standard which means it might not be supported on all platforms. Intel themselves only point to official support for Z97 and X99 chipsets. In order to get a better idea of the landscape of NVMe compatibility, I took it amongst myself to start testing the add-in card version of the 750 Series in just about every modern motherboard I could get my hands on at the office.