On December 15, 2011, Paul Thurrott of Supersite for Windows appeared on Windows Weekly 239 with Mary-Jo Foley (another regular commenter on Microsoft and Windows operating systems and technologies). My specific area of interest here is the company’s release of iOS apps for Skydrive, Lync, and an update for OneNote 1.3 for the iPad (iOS apps for Kinectimals, Bing, and an iOS connector for email are also available as well).
Of course, it’s still the case that the deepest and most effective support for MS applications occur on Windows Mobile platforms (but because this occurs at the OS level) sometimes the non-Windows mobile platforms get such support sooner than major Microsoft releases can manage. As Thurrott points out, Microsoft is not to tightly aligned that it can push out mobile OS releases in synch with features under development in various part of the company. Some Windows enthusiasts (bigots?) have expressed displeasure that iOS in general (and iPad in particular) have gathered features and functions that might not yet have appeared on mobile Windows platforms. That’s why it’s still the case that Xbox Live is better integrated into Windows mobile platforms, as is the entire Windows Office suite.
It’s going to be interesting to see how this unfolds. Given the increasing importance of mobile apps, and the proliferation of platforms (and the inescapable popularity of iOS) it’s ever more likely that MIcrosoft will reach out to competitive platforms at the same time as it makes apps to run on those selfsame platforms.
On Thursday, December 15, MS released a hotfix to “optimize the peformance of AMD Bulldozer CPUs…” for Windows 7 or Windows Server 2008 R2 (KB2592546). Seems that the original code base wasn’t properly set up to take full advantage of multi-threading in the Bulldozer architecture. According to some hurry-up benchmarking work from X-bit Labs, the hotfix could boost performance anywhere from 2-10 percent, depending on the actual application involved — but only if that application is designed to run in multi-threaded fashion.
This morning, however, VR-Zone posted a story entitled “Microsoft Pulls Down the AMD Bulldozer Multi-Threaded Patch,” citing reports of user problems and peformance decreases for the reversal in course. While you can visit the Hotfix page for KB2592546, you’ll immediately notice that the download links on this page are no longer live.
I’m guessing that a development team is working feverishly to address whatever gotchas popped up following the inital drop of this release onto the Microsoft Support pages. I imagine we’ll see a round two on this stuff as soon as it’s judged to be “really working this time.”
[Note on 12/19: HardOCP has since explained that the initial hotfix was incomplete, and got pulled because it demonstrated a negative effect — that is, decreased performance — on enough Bulldozer systems that MS apparently decided it was better off pulling the code than leaving it up for download.]
As anybody who reads this blog regularly knows, I use and endorse Secunia’s excellent Personal Software Inspector (PSI) software on my notebook and desktop PCs. This program takes a look at the OS, applications, and helper software components to check release versions and dates, then compares it to its voluminous database of current OS patches and fixes, application updates, and more, to determine what elements of a particular install are out-of-date and need to be made current. It’s a peachy program (a corporate version called CSI is also available for business and commercial use) and one that I regard as essential in helping me keep my machines secure and up-to-date.
This morning, when I logged in I saw various security update bulletins that induced me to run the PSI scan on my primary desktop, it reported that my Microsoft Silverlight installation (and Google Chrome) needed updates. This struck me as odd because Patch Tuesday just hit yesterday, and I’d already updated that system. As it happens, Silverlight has just been updated to a new major version (5), but this update is not yet being distributed through Windows Update itself. Here’s what clued me into the situation:
It’s weird to find oneself in a situation where a piece of Microsoft software displays this kind of warning, even in the face of Windows Update. That’s how I figured out that Silverlight 5 hasn’t yet fallen under the WU umbrella (as also happens with OS Service Packs, which must be manually downloaded and installed until some time after their official release).
So now that I knew what I was dealing with, I jumped over the the Silverlight page at www.microsoft.com/silverlight, where I beheld the following welcome:
So indeed an upgrade was needed. And once it was applied no more warnings, and everything was once again up-to-date. This is just what PSI is designed to do, and this time it made me extremely happy for it it do it for me: otherwise, I might not have realized Silverlight needs a manual update until after some no-doubt heinous exploit had already been foisted!
In my last blog — entitled “Heck No, We Won’t Go! (from Windows 7 to Windows 8, that is)” — I argued that given how happy most folks are with Windows 7, and how little non-touch machines gain from the upgrade, that many satisfied Windows 7 users are unlikely to upgrade existing machines to the new OS after it ships. Unsurprisingly, not everybody agrees with me: I’ve received email (and one comment) from a handful of readers advancing these arguments:
- A certain class of PC user is inclined to use the latest and greatest, no matter what it may be in the interests of keeping up
- Another class of PC user (including your humble author) is required to work with (and in some cases like mine, write about) the latest Windows version even now, as it’s in pre-beta status
I have to concede that some people will indeed do the upgrade for all kinds of good reasons. But my argument is more along the lines of “those who don’t HAVE to probably won’t” rather than “nobody’s going to do it at all.” Of course, I also cheerfully confess that my previous post’s headline is a bit lurid and exaggerated, so I am probably overdue for a little “constructive feedback.”
For another interesting take on the interested in upgrading side of this discussion, check out Matt Egan’s very nice article for PC Advisor this morning (12/12/2011) entitled “Why 2012 will be the year of Windows 8.” His argument is worth reading, but my thumbnail sketch is that he believes that users are more interested in shared and common access to services and data, and that Windows 8’s ability to integrate apps and info across tablets, smartphones, and desktop is an unbeatable proposition that is also necessary to keep MS In the same league as Apple and Google.
FWIW, I agree. But I also observe that the only way you get these advanced Windows 8 features is on new PCs with UEFI BIOS and with iX class CPUs with SLAT (or the AMD equivalent). Thus I believe this is a case where I can have my cake and eat it, too — because my previous argument is about UPGRADING from Windows 7 to Windows 8, not about buying new systems that support the full panoply of Windows 8’s advanced features. I stick by my argument having now had more time to realize that without these razzle-dazzle features, users will either stick with Windows 7 on older hardware or bite the bullet and buy new Windows 8 capable systems. I’m already speccing-out one such system myself! I’ll probably upgrade (or dual boot) some of my older systems with Win7/8 just to see how that works, and to better understand how far behind the curve pre-2010 PCs will be in a Windows 8 world.
Here’s a cheery bit of news from our friends over at MaximumPC. The story’s headline “IDC: Happy Windows 7 PC Users Won’t Switch to Windows 8” pretty much says it all. Quoting from ZDNet’s Mary-Jo Foley, the operative phrase is that “Windows 8 will be ‘largely irrelevant’ to traditional PC users.” The interior quotes are from an actual IDC report, which costs $3,500 and should therefore explain why none of us underpaid Windows hounds have the necessary scratch to obtain and quote directly from the real source. It’s entitled “Worldwide System Infrastructure Software 2012 Top 10 Predictions” from which IDC VP Al Gillen tweeted the net-net items on December 2:
01. Customer Face Confusing Choices as Virt, Cloud System SW and Information Automation SW Converge.
02. Private Clouds Will Grow Like Gangbusters, One Use Case at a Time.
03. 2012 Will Be VMware’s Last Year as King ot the Hill. Competition Starts Squeezing Hard in 2013. No More Green Field for VMware.
04. Operational Complexity Will Drive Demand for Predictive Analytics and APM.
05: Consumerization of IT Will Create New Management Challenges and Solutions.
06. Platform as a Service Will Ramp Up Slowly Due to Lock-in Fears.
07. Battle Royale Will Be Waged to Establish Linux Kernel of Cloud Computing.
08. Enterprises Will Reconsider Benefits of Infrastructure Heterogeneity.
09. There will be Layers for the Masses, Stacks for the Few.
10. Expect big success for WS8, Win Client 8 a skip-over for desktop users. Uphill battle for WC8 mobile space.
Item 10, of course, is what caught MJF’s eye and produced this magnificent quote from the actual report: “Windows 8 will be largely irrelevant to the users of traditional PCs, and we expect effectively no upgrade activity from Windows 7 to Windows 8 in that form factor.” IDC believes that Windows 8 will become generally available no later than August 2012, at which point PCs with Win8 pre-installed may be sold. I take this to mean that while new PCs may come bearing Win8, users won’t be lining up at midnight to buy the upgrade media at retail outlets, or clogging the Internet to download the new OS when the floodgates open.
At the enterprise level, of course, things will lag further behind. With many organizations in the middle of migrating to Windows 7, or having just completed same in the past 12 months, enterprises aren’t exactly eager to gird their loins and “do it again.” It’s typical for early enterprise adopters to jump on new OSes about a year after release, and for the main pack to slink onto the platform a year or two after that.
And with most regular, non-enterprise users pretty happy with Windows 7 on current PCs, there aren’t a lot of obvious and compelling reasons visible just yet to propel them to upgrade. This should be an interesting launch to watch–especially for Microsoft!
OK, I cheerfully confess that I glossed over the November 21 Building Windows 8 blog posting by Christa St. Pierre of the Windows 8 Setup and Deployment team, thinking that because I’ve gone through a couple of Windows 8 pre-release installs that I didn’t really need to read it from stem to stern. I was wrong, not because I had some experience and thus also, some appreciation for how it works and how it compares to Windows 7, but because that decision to “just skip it” deprived me of some valuable background information and insight into how the Windows 8 install process has been redesigned. Perhaps I should have paid closer attention to that posting’s title: “Improving the setup experience.”
But thanks to Jon Brodkin over at Ars Technica (one of my favorite sources for inside information, insight, and breaking Windows news and rumors) I started to understand more about what was going on with this part of the Windows 8 experience. The title of his article, in fact, points to some incredibly salient points–namely “Windows 8 gets faster installation, 11-click upgrade for casual users.” So, plowing through this piece I discovered information about how the Windows 8 process has been substantially streamlined so that what required multiple downloads and 4 programs to complete, and involved working through 60 screens of information, is now condensed into a single program that might require working only 11 screens. By itself, that’s pretty remarkable (and this is coming from somebody who’s had to write up detailed install descriptions and instructions for every version of Windows since 3.1 way back in the mid-1990s).
There’s also a fascinating discuss of the upgrade process (Windows 7 to Windows 8, in this case) that explains how the number of files and applications installed can affect overall install completion time. Because Windows 7 stages files for upgrading, it ends up copying files twice during that process, even though the bulk of such moves start and end in the very same folder! Simply put, Windows 8 moves what it has to more expeditiously at the folder lever when moves are required, skips unnecessary moves, and uses hard links to move things logically in the file system without actually moving stuff around on disk. This cuts upgrade time for large complex installations from 188 minutes to 46 minutes on systems with 430K files and 90 apps (about a 75% improvement) and from 513 minutes to 52 minutes on systems with 1.44M files and 120 apps (almost 90% better). Wow!
The blog goes on to explain how Web delivery has been optimized, through better compression, elimination of duplicate OS files, and smarter download behaviors. It closes with a nice overview of the Windows 8 ADK (Assessment and Deployment Kit, which replaces the old Windows Automated Installation Kit or WAIK), along with some best practices info on building Answer Files to automate bulk and remote installs.
This post is definitely worth reading, and pondering carefully, as you start thinking about test lab, pilot, and ultimately, production roll-outs of Windows 8. Be sure to check it out!
The word on the street is apparently out, and it’s claiming that we’ll see a large-scale beta release of Windows 8 sometime in January or February 2012.This is not too ridiculous to believe, by any means, considering that the developer preview made its debut around September 14. Among many publications proffering such guesstimates, PC World’s story “Windows 8 Beta Expected in Early 2012” gets my vote.
I’m glad to have an approximate time-frame for this upcoming release, because I want to build a new PC with a UEFI BIOS and a touchscreen in time to install the Windows 8 beta shortly after it’s let go. Looks like I’ve still got some time left to put together a new configuration. I’m guessing a lot of other readers who like to keep up with Microsoft’s release schedule may be thinking the same thing!
With the impending release of the CompTIA Storage+ credential on January 18, 2012, the Storage Networking Industry Association (SNIA) has announced the withdrawal and impending retirement of its SNIA Certified Storage Professional (SCSP) credential the day before. In fact the SCSP home page currently starts out this way:
In checking out the latest hub-bub on the Building Windows 8 blog this morning, I encountered a new post entitled “Enabling large disks and large sectors in Windows 8.” It’s the work of Bryan Matthew, a program manager on the Windows 8 Storage & File System team, and it’s definitely worth a read-through, not just for its explanation of how the file system in Windows 8 will support disks larger than 2.2 TB directly (without requiring an add-in driver like Seagate’s beyond-2TB software) but also for its explanation of where the technical limitations come from on BIOS-based systems that set the ceiling at 2.2 TB.
A big part of the answer to the support question involves another invocation of UEFI (Unified Extensible Firmware Interface) which is also the key to Microsoft’s boot-time security protection for Windows 8 as well. Here’s what Matthew has to say about UEFI and Windows 8: “Beginning with Windows 8, multiple new capabilities within Windows will necessitate UEFI. The combination of UEFI firmware + GPT partitioning + LBA allows Windows to fully address very large capacity disks with ease.” Although this is nice, it is also another indication that BIOS-based PCs–which represent over 99 percent of the current installed base, including every system I myself own–are heading for second-class status in the brave new Windows 8 world.
I’ve recently started speccing out a UEFI based system with a touchscreen to put a Windows 8 test system together, and I’ve learned that the number of UEFI-based motherboards is still very small (though it’s growing by leaps and bounds every month: when I first checked on this in September, I could find only 8 UEFI mobos on Newegg; this morning, the site lists 28 AMD-based mobos and 43 Intel-based mobos). Though I can understand and appreciate the technical advantages of UEFI, I wonder if the world is ready to make a mass migration to a new firmware architecture simply to take best advantage of Windows 8.
I’ve got two newish notebook and an equal number of newish desktop PCs and I’m not planning to junk them for at least another 2-3 years. Will the need for UEFI to obtain the best Windows 8 experience force me to change my mind? I don’t know yet. But I do know this is gonna be interesting!
I’m the reasonably proud owner of a Dell M11X (Alienware) mini-laptop with an i7 26xx processor, 8 GB RAM, and an OCZ Agility 3 SSD. It runs fast enough for me, and has enough horsepower to do everything I ask of it with aplomb, including some occasionally heavy-duty protocol trace capture and analysis (Wireshark rules!). But yesterday, when I turned it on, I couldn’t help but notice that its built-in Intel WiMax 6250 module had gone bye-bye.
When the laptop booted up, it reported that it couldn’t find the WiMax module. So, I ran the Intel WiMax connection utility to try to kick it into presence. No dice. Next, I read the help file to learn that I should probably duck into Programs and Features to run a “repair” operation on the WiMax connection software. And while that went without a hitch, there was still no WiMax module to be found after my next reboot.
Then, I went into Device Manager and, sure enough, no WiMax device appeared under the Network adapters heading, in contradiction to normal behavior. So I dug into the Dell Technical Support site to see if I could find any joy from their help and advice. Zilch.
After that I remembered seeing the WiMax module as I removed the unit’s conventional HD a couple of weeks ago to replace it with the SSD it’s now using. “Could I have jiggled an antenna wire, or loosening the module somehow?” I wondered out loud. Popped open the hatch, and saw nothing amiss with the WiMax 6250 module upon visual inspection but pushed on the antenna connections and made sure the module was tightly socketed anyway.
I’m not sure if this fixed the problem, or if the WiMax module had simply taken a short vacation, but when I rebooted the machine–Presto! The WiMax module appeared, and established its usual rockin’ connection to Clear Communications. Problem solved, with not too much muss or fuss, or time wasted.
What it reminded me of is the old adage about troubleshooting: always look at what else has changed recently on a system when components go kerflooey. Again, I can’t be 100% sure that opening the case and checking the WiMax module fixed my problem. But performing those actions did result in a return to normal function, so I’m not inclined to look this gift horse too deeply in the mouth.
I was kinda glad it turned out that way, because I’d started to investigate the Dell Technical Support channels I would have had to traverse had more serious fixes been required. The only phone number I could find was for premium, added-charge support. I’m not sure my same-day hardware replacement support warranty gets me that high up the food chain, and I guess I save the effort involved in climbing that learning curve for another day!