February 6, 2012 6:08 PM
Posted by: Ed Tittel
increasing touchscreen options available for Windows 8
, using touch with Win8 involves a learning curve
OK, so I’m getting onboard for an upcoming “day and date” book on Windows 8 (this means we have to have the book finished far enough in advance of the ship date for Windows 8 that it can be sent to bookstores to arrive in time to be stocked and waiting the day the OS goes into general availability, or GA status). That means I’m starting to dig down into the tools and technology so my co-author and I can get our systems and our acts together to exercise and document its functionality while waiting pensively for the release of the “consumer preview” or “beta release” of Windows 8 that’s due out later this month.
First up in Newegg 19-24 inch touchscreens
This means I’ve ordered a new-generation Intel-based system (with SLAT and UEFI) to host the new OS, and am looking at my options for touchscreens to enable the touch-based Metro interface to do its thing properly on my test machines. I’m also reaching out to various PC manufacturers to lay hands on their all-in-one (touch-based) systems, because I’m learning that the aftermarket for touchscreen monitors starts pretty expensive ($340 -$620 for a 15″ touchscreen monitor, $508-697 for 17″, $549-980 for 19″, and so forth) and is full of “interesting deals” such as an Acer 23″ HD resolution touchscreen for only $329.
Second Newegg item for 19-24 inch touchscreems
A great place to check out what’s available (and the number and range of offerings have exploded since the last time I checked on this in December, 2011, so the market’s obviously getting ready for Windows 8 ) is at Newegg, where they actually have a product category called “Touchscreen monitors.” Here, you can filter those monitors by size (under 15″ (3 total), 15-17″ (55 total), 19-24″ (29), 26-42″ (8), and over 42″ (1)), price, manufacturer, and even by type: 5-wire resistive, AccuTouch, Acoustic Pulse, Capacitive, CarrolTouch, and IntelliTouch [see Note 3 at the end of this blog posting for comments on this terminology; taken straight from the Newegg touch monitors listings, it includes numerous trade names and isn't really descriptive of the touch technology landscape. This will be my one and only use of these terms].
Item 3 in the Newegg 19-24 inch touchscreens
At this point, I can see I will have lots of options from which to choose. I now understand I need to learn more about the types of touch technology and how best to integrate touch into a desktop system for easy access and use. I’ll keep you posted as I zero in on some good choices and let you know about such deals as I can find. I can’t be the only person who needs to figure this stuff out! I’m also casting about for hardware recommendations and war stories in making touchscreen monitors work with Windows 8, so please let me know if you know of or have any to share!
New Material Added on 2/9/2012
Since this blog originally posted on 2/6/2012, I have been in repeated contact with Geoff Walker, of Walker Mobile at www.walkermobile.com. He’s a long-time expert and full-time consultant and writer on mobile computing with a particular emphasis on touch technologies. Check out his Website, where you’ll find a wealth of valuable and informative materials on touch technologies of all kinds. Geoff has also been a guest editor on Touch Technology for Information Display magazine since 2007 (where a search on his name produces over 500 hits), and has contributed numerous stories on touch to other industry publications as well. Geoff has very kindly provided copious feedback for both Episodes 1 and 2 of this blog posting, and I will now recast his remarks here in somewhat abbreviated form, in the form of a series of numbered Notes to follow. They are primarily quoted verbatim or merely shortened in the interests of brevity; where I provide pricing or other information, it’s in square brackets to identify my presence, not Geoff’s.
Note 1: The best touch monitor in the world today comes from Perceptive Pixel: though expensive, it’s without question the very best [$6,000.00!]. The next best is from 3M [around $1,400.00]. The touch performance is phenomenal, as good as Perceptive Pixel, just smaller with a few more limitations. It’s not as expensive as the Perceptive Pixel, but it’s not cheap.
Note 2: In the Newegg listing, the Acer T231 bmid is a typical consumer-level Win7 touch monitor. It uses camera-based optical touch (2 cameras) from Quanta. It’s limited to two not-very-robust touches (in other words, it barely works on Win7, much less Win8 – read my article on optical touch and you’ll get some hints of why that is). But it’s still a very valid test platform, since almost all of the consumer multi-touch monitors sold for the past two years have used a similar touch system, and all those owners will want to know what to expect when upgrading to Win8. Note that there is NO information whatsoever in the Acer specs about touch! That’s a big red flag, indicating that touch is not something that Acer (or any of the PC OEMs) really cares that much about. Compare Acer’s lack of touch specs with the touch specs of the other monitors I’ve mentioned so far – big difference! The HP L2105tm and the Viewsonic VX2258wm are basically the same as the Acer – all of their touchscreens are camera-based optical from either Quanta or NextWindow.
Note that what I’m saying is that there are NO monitors in the Newegg list that are capable of meeting the Win8 5-touch specification because doing that requires either (a) projected capacitive, which is too expensive for a consumer monitor today, or (b) improved camera-based optical (with at least six cameras), which won’t be ready for another six months or so. [The emphasis on NO in the preceding text is Geoff's.]
Note 3: [About these terms “…5-wire resistive, AccuTouch, Acoustic Pulse, Capacitive, CarrolTouch, and IntelliTouch" (these appear in my original blog, and came straight from the Touchscreen Types category on the Newegg Touchscreen Monitors page).] I strongly recommend that you stop using the Elo brand names for generic touch technologies. It’s confusing to everyone and does nobody any good. Even Elo would like to get rid of them, but it’s very hard to change that kind of thing once you start. Use only the generic technology names, and use them consistently. [Will do, Geoff, and thanks for pointing this out! Fortunately, that was already the case by the time I got to Episode 2.]
February 3, 2012 3:50 PM
Posted by: Ed Tittel
How long can XP remain king of the OS hill?
, Windows 7 hasn't yet eclipsed Windows XP
, Windows XP still on top
Trends don’t always translate into facts, as a quick glance at this “Desktop Top Operating System Share Trend” graph for March 2011 through January 2012 illustrates.
The NetMarketShare trends suggest imminent inflection in December then diverge in January
I must confess that I (and numerous other pundits and online panjandrums) interpreted this graph to mean that Windows 7 would surpass XP last month, but that’s not how it turned out. Instead of each line following its long-established trend, numbers for XP actually jumped a little in the period from December 2011 to January 2012, from 46.52% to 47.19%, while the numbers for Windows 7 dipped slightly from 36.99% to 36.40% in the same period. Go figure!
Thanks to the editors at Tom’s Hardware for bringing this to my attention, in their February 2, 2012, story entitled “Report: Windows XP is Still The Dominant OS.” They speculate — and I agree — that still-lagging economies (especially in Europe, sorely best by the Greek debt crisis and the Euro melt-down) can easily account for the recent stall in the XP-to-Windows-7 cut-over that’s been chugging along for the past two years or better. Likewise, they observe that the pending release of Windows 8 in the third or fourth quarter of 2012 will further complicate matters, and should probably cause the trend lines for both of those OSes to decline further and faster.
One thing’s for sure: it’s still impossible to write off Windows XP, and it’s still around in huge and substantial numbers. With the “absolute retirement date” of this venerable OS slated for April 2014, it’s amazing it’s been able to cling to life and major market share for such a long, long, long time!
February 1, 2012 2:59 PM
Posted by: Ed Tittel
care in selecting Intel Chipset drivers is key
, finding the most current Intel Chipset software and drivers
Last Friday (January 27, 2012) I posted a blog entitled “Intel Driver Update Utility and Intel Chipsets
,” wherein I observed that this useful online tool — which analyzes the systems on which it’s running, and reports on the currency and status of the drivers for Intel components they contain — doesn’t always recommend the most current Intel chipset driver. This observation nagged at me sufficiently that I did some additional research to determine why this might be the case, and how to find and use the most current such drivers for those who choose to use them.
About the “latest version” of the Intel Chipset Software
Intel proffers a Web page entitled “Do I need to upgrade to the latest version” under the general head of Intel Chipset Software Installation Utility. This page suggests an “If it ain’t broke, don’t fix it” approach to updating the Intel chipset software and its associated collection of drivers. Here’s what that information says, verbatim:
|Intel® Chipset Software Installation Utility
|Do I need to upgrade to the latest version?
|Upgrade to the latest version if you are experiencing an issue listed under Issues Resolved in the latest release notes. The latest version of the Intel® Chipset Software Installation Utility and the release notes are available in Download Center.If you upgrade to the latest version, follow the have-disk installation instructions. The utility cannot install using setup.exe if another version is already installed on your system.
Finding the “latest version” of the Intel Chipset Software
In the Intel Download Center, specify the selections “Chipsets” (Product Family), “Chipset Software” (Product Line), and “Intel Chipset Software Installation Utility” (Product Name) on the initial Find Downloads by product name window that appears. This produces a Find downloads by categorywindow where you can specify your operating system (Windows 7 (32-bit) in my case) and Download Type (I picked “Utilities, Tools, and Examples” because that’s what provides the INF update utility that automatically updates Windows drivers for those willing to bypass the preceding instruction to “follow the have-disk installation instructions” (the chipset driver itself isn’t accessible through Device Manager in any case).
This is where things start getting interesting. Intel’s label for “Latest” driver applies to the file they’ve updated most recently NOT to the file with the highest version number (which actually indicates the most current driver). Here’s a screen shot that illustrates this potential point of confusion very nicely:
Pay careful attention to the date and version columns when selecting driver downloads
Notice that the utility with the higher-numbered version (22.214.171.1242 versus 126.96.36.1990) also has the more current date (8/27/2011 versus 4/21/2011). It’s essential to pay close attention to all of this information if you want to override the recommendations of the Intel Driver Update Utility, and choose the right version to download and install on your PC.
[Note in response to reader comment: You MUST identify the chipset in your PC and search on that particular chipset model to find the drivers and installation software that match it. I don't mean to suggest that the version numbers mentioned in this or the previous posting are prescriptive for all PCs and all readers. Rather, they were meant only as illustrations of the various version numbers that I encountered for the various chipsets on different PCs. Thus, for example the X38 chipset on my production PC gets version 188.8.131.522 mentioned elsewhere in this post. OTOH, my D620 Latitude notebook has a Mobile Intel 945 Express chipset and gets version 184.108.40.2069 instead. As they say on the Internet, YMMV, so be sure to grab only what's relevant for your particular chipsets.]
January 30, 2012 4:02 PM
Posted by: Ed Tittel
stable Windows 8 ARM code could be released to developers soon
, stable Windows 8 developer release for ARM rumored for Feb2012
An interesting rumor popped up on SlashGear this morning, in an article from Chris Davies entitled “Windows 8 on ARM stable release in February tip developers.” It seems that despite earlier issues with the stability of the developer preview for Windows 8 on ARM CPUs, another, much more stable release of Windows 8 will be released in February — again for developers — to coincide with the “customer preview” release of Windows 8 for x86 CPUs that heralds another milestone in the march to RTM (release to manufacturing) and GA (general availability).
Citing sources from CNET, Davies said “…there’s no obvious reason that [an] ARM-vesion of Windows 8 … should be ‘staggered’ from the traditional x86 build.” Furthermore, the same sources opined that “…ARM alternatives to Intel and AMD based Windows 8 machines could ‘undercut them by hundreds of dollars…’” Given the popularity of chips from Texas Instruments and Qualcomm, among others (these two companies showed Windows 8 prototypes at this year’s CES), for smartphones, this could be a huge coup for Microsoft in finally making a dent in that non-PC marketplace.
Should be very interesting to see how all this plays out. Stay tuned!
January 27, 2012 7:17 PM
Posted by: Ed Tittel
Intel Driver Update Utility doesn't always recommend the newest drivers
, Intel Driver Update Utility settles for valid rather than current drivers in some cases
- The Intel Driver Update Utility is handy but not always completely current
The Intel Driver Update Utility(IDUU) is a handy-dandy software tool that depends on a “Systems Requirements Lab” active widget from Husdawg to scan PCs for Intel components, and to report on their current update status. It’s pretty useful, and has become part of my normal maintenance routine for checking driver status on the PCs I manage. I also use DriverAgent (though other tools like RadarSync, Driver Detective, Drive Guide, and so forth also do pretty much the same thing) but I’ve observed that Intel is often better at keeping up with its own drivers than third parties, so I’ve come to depend on IDUU to help me keep my Intel drivers as current as can be.
But recently, I noticed something subtle about the language that the IDUU uses to report on drivers it finds, that in turn led me to realize that Intel apparently doesn’t care if a driver is the most current in every case. Rather, it appears to care only that some drivers are “valid,” even when newer drivers may be available. Notice this report block on the Intel Chipset Software Installation Utility:
Now, compare this to the language used for the built-in Realtek RTL8013EL Ethernet GbE adapter on that same motherboard:
In the chipset case, the version is valid, but in the Ethernet adapter’s case the driver is current. “Hmm…” I wondered, “Is there a difference between valid and current?”
The answer turns out to be “Yes.” By researching the most current Series 3 Chipset Driver through a manual search on Intel’s Download pages, I was able to determine that the highest numbered version available was 220.127.116.112. My machine was happily running version 18.104.22.1686 instead. A quick download and install took care of that issue as the preceding chipset screen capture now attests, but this leaves me wondering why the IDUU doesn’t tell its users that a more current version is available. My best guess is that it waits until some critical feature gets introduced in a newer update, and only then instructs its users to update their drivers. That would be a reasonable approach to driver updates where “if it ain’t broke don’t fix it” often prevails as a guiding principle. I can only imagine that’s why Intel also labels the 22.214.171.1242 version as valid, even though it’s also the most current as well.
But if, like me, you want at least some machines to always be running the absolute latest and greatest Intel drivers, if only for test purposes, it’s good to know that when you see “valid” in this tool you should probably go looking for something more current, just in case it’s out there for downloading.
January 25, 2012 8:55 PM
Posted by: Ed Tittel
basic Windows 8 API makes it easy for manufacturers to include sensors in Windows 8 platforms and for developers to put them to work
, Windows 8 offers simple access to sensor data for sophisticated uses
The latest entry in the Building Windows 8 blog is called “Supporting Sensors in Windows 8,” and it comes from Gavin Gear, one of the Microsoft Product Managers on the Windows 8 Device Connectivity team. It tells a fascinating story of how (and why) the basic Windows 8 hardware requirements include so many sensors (an accelerometer, a magnetometer, and a gyroscope, among others) so that application and service developers can count on basic system capabilities when building next generation software of all kinds. The following diagram helps to illustrate how combining such hardware devices together can support smoother and more capable physical functions, not just for the usual purposes (screen rotation and orientation) but also to support lots of other interesting capabilities as well (including game controllers, smart remotes, measurement and data acquisition, and so forth):
Position and motion sensing devices work better in concert than by themselves
Mr. Gear follows up his explanation and discussion with some data traces from an accelerometer to depict why straightforward use of such data doesn’t always produce the best user experience. From there he goes on to explain how MS has worked with hardware manufacturers and device makers to create a single Microsoft driver that can “…work with all Windows-compatible connected over USB and even lower power busses…” The upshot is that any sensor company can construct a Windows 8 compatible package by adhering to “…public standard USB-IF specifications for compliant device firmware.”
Pretty slick already, but MS also worked to control power consumption and sensor performance as well by enabling sensor processing to occur at the hardware level without involving the CPU, and by building filtering mechanisms to restrict data flow (and event rates) to speeds that won’t exceed the processing stack’s handling capacity (this also helps reduce consumption of CPU cycles).
The blog concludes with a discussion of how sensors can play into Windows 8′s new “Metro-style” apps, using a new sensor API included as part of the Windows 8 runtime environment (aka WinRT). Code examples show the APIs to be simple and straightforward, making them easy for developers to put to work. It should be interesting to see how easy access to sensors and location data helps raise the bar for Windows 8 apps over the years and months ahead. There’s even a product reference to a specialty sensor part built into last year’s Windows 8 Developer Preview slate PC from Samsung, now that developers can purchase such parts on the open market, to help get this phenomenon rolling.
January 23, 2012 2:51 PM
Posted by: Ed Tittel
MS cert exams continue to change and evolve
, MS cert exams deal with issues of fairness and challenge
As part of my daily due diligence in writing this blog, I keep up with the postings on Microsoft’s Born to Learn blog. A recent post (1/17/2012) from Andrew Bettany entitled “Have MSTS/MCITP exams got easier?” caught my eye this morning, as much or more because of the comments to this post as for the post itself. Be sure to check it out if you’ve got a Microsoft exam in your near-term future.
But what really got my wheels turning were the concepts of “fairness” and “challenge” raised in a comment by Microsoft’s lead psychometrician, Liberty Munson, especially in light of my recent brokering of contact between herself, an exam developer, and a friend and colleague who had taken the 70-662 Microsoft Exchange Server 2010, Configuring exam no less than four times without managing to pass it on any attempt. As somebody who’s actively engaged in consulting with major Fortune 1000 companies to help them transition hundreds of thousands of email users into Exchange 2010, I have trouble accepting the notion that he simply doesn’t understand the platform and its capabilities. In that light, consider what Ms. Munson has to say in this posting:
Thanks for great post! It shouldn’t surprise anyone that people who pass exams tend to think they are easy while those that fail believe that they are too hard. The key is to make sure that those who fail also believe that they were “fair assessments” of their skills while those who pass also found them to be “challenging” (or “easy but challenging”). That’s nuance of “easiness” that is often overlooked by candidates and hiring managers. Exams should be “easy” for qualified candidates because we’re assessing skills that they have, but they should also be “challenging” and fair assessments of skills. We rarely get past the “it’s too easy” part of the conversation to understand the deeper nuance…Was it challenging? Was it fair? Does it differentiate you as someone who knows the technology? It’s possible that our exams haven’t been challenging–that this is assumed in the “it’s too easy” statement–but there’s a deeper question here that needs to be asked.
Enough with the philosophical aspects of this conversation :)…over the last year, we’ve made changes to our item writing guidelines and use of item types to increase the percieved difficulty of our exams because we understand that the conversation generally stops at “it’s too easy”…certain item types and item writing strategies can be leveraged to make exams feel more challenging without requiring more skills or knowledge. However, as we introduce new certifications, you’ll find that we’ll be requiring deeper knowledge and higher level skills to address comments that “it’s too easy.” Psychometrically, I regularly monitor the difficulty of the exams and ensure that it’s at an appropriate level for the target audience and programmatic goals/definitions of what it means to hold the credential. Changes are made when and where necessary to meet these targets.
I’d love to start hearing people talk about our exams as “challenging and fair assessments” as we continue to improve the ability of our certifications to differentiate people who really know their stuff from those who don’t or just sorta do. Andrew–Thanks for starting to get the word out!
Clearly, some of the ongoing renovation of existing MS exams and development of new ones incorporates great concepts and intentions on the part of the people who work with SMEs to put those exams together, and to measure and monitor them for statistical relevance to the populations being tested. But “fairness” is a difficult concept to get right, especially when exam questions veer into the realms of what my aggrieved friend and colleague perceived as mostly “administrivia” that real-world admins would simply look up if they had to deal with such seldom-used (or rarely-needed) features and functions. In his case, he’d readily concede that the 70-662 exam is challenging, but also vehemently claims that it’s grossly unfair.
I symthpathize with both parties in this kind of encounter. On the one hand, having created hundreds of practice tests for the certification exams covered in my many Exam Cram and study guide certification books, I get how difficult it is to create sound, meaningful and relevant exam items for candidates to ponder and learn from. On the other hand, I also really feel for my friend’s situation where his consulting company requires him to pass the 70-662 exam so he can present the “right credentials” to his many clients, but where he feels that the exam is more of a random check on obscure or little-used Exchange features and functions, than a meaningful assessment of his skills and knowledge of the platform.
This is a tough situation, and one that requires not only that both sides recognize each other’s motivations and priorities, but also that the testing party provide enough information to testees to let them understand what they must learn to get past the exam, and that testees take this seriously enough to make a concerted effort to learn and master the necessary skills, knowledge, and even administrivia details to get over the hump. Usually, taking a Microsoft Official Curriculum class, reading the books, and acing numerous practice exams is enough to guarantee a good result. But when this tried-and-true method fails, it’s time to start asking why and trying to reason one’s way into the proper information set to figure out what’s wrong. My colleague also believes that MS should provide testees with more detailed feedback about questions answered incorrectly rather than simply flagging certain concept areas where the test-taker failed to meet minimum score requirements.
All this goes to show that certification exams really aren’t easy to create, and sometimes can be incredibly difficult to pass as well. It should be interesting to see how this situation plays out, and how the two parties will find some kind of rapprochement.
January 20, 2012 4:14 PM
Posted by: Ed Tittel
MS unveils Win8 logo requirements. Win8 logo devices to include lots of interesting and useful features
In his most recent monthly column for WindowsITPro entitled “The Rigor of Windows 8 Hardware Certification Requirements, …” Windowsmeister Paul Thurrott provides a snapshot of the current requirements that Microsoft intends to levy on those seeking to put the Windows 8 logo decal on their PCs, notebooks, tablets, and smartphones. It’s an interesting and detailed collection of info, based in turn on Rafael Rivera’s equally interesting WithinWindows blog entitled “Windows 8 Secrets: PC and Device Requirements.” Both are worth reading in their entirety; here I’ll summarize some of the most interesting high points.
A mock logo for Windows 8
- 5-point digitizers: Windows 8 touch PCs must support at least 5 touch points (five fingers’ worth, guaranteeing support for complex multi-touch swipes and gestures).
- NFC “touch marks”: Near field communications lets devices interact with each other when they’re so close as to be nearly touching, and are key to emerging technologies for payment transfers especially from smartphones to cash registers (or computers acting like same).
- 5 required hardware buttons: devices must provide buttons for power, rotation lock, the Windows key, and volume up/down controls. Also, devices that belong to a domain that lack keyboards must support Windows Key + Power as an alternative to the Ctrl+Alt+Del key sequence (the infamous “three-fingered salute” for those who remember DOS and early Windows versions).
- Minimum component requirements include 10 GB of free storage space after Win8 is installed, support for UEFI (see my 9/23/2011 blog for more info), WLAN, Bluetooth and LE networking, Direct3D 10 graphics with a WDDM 1.2 driver, 1366×768 screen resolution or better, 720p camera, an ambient light sensor, a magnetometer and an accelerometer, a gyroscope, at least one USB 2.0 controller and exposed port (or better), and audio speakers. Sounds like a decent smartphone or tablet, eh?
- Graphic drivers must be upgradeable without requiring a reboot (easier to do thanks to dropping XDDM drivers and keeping only WDDM drivers for Windows 8).
- 2-second resume(does not apply to ARM-based devices yet) from Standby (S3) to “resume complete” status. Rivera speculates “Microsoft simply doesn’t have enough data in this space” to impose the same restriction on ARM as on x86; Thurrott speculates that “it will be added in a future release, such as Windows 9.”
Veeeeeeeeeeeeeeery interesting. Be sure to check out the original posts, too!
January 18, 2012 4:18 PM
Posted by: Ed Tittel
ReFS offers interesting new file system capabilities
, Resilient File System or ReFS to supplement New Technology File System or NTFS in Windows 8
Building Windows 8 strikes again, this time with a 1/16/2012 blog from Surendra Verma, development manager for the Windows 8 Storage and File system team. It’s entitled “Building the next generation file system for Windows: ReFS.” The ReFS acronym is expanded in the FAQ that follows the blog post to mean “Resilient File System;” that same FAQ also documents some awe-inspiring maximum file system attributes.
What’s New About ReFS?
Verma takes several cuts at answering this question without necessary addressing it completely explicitly, so I’m doing my best to read between the lines here. It appears that ReFS comes with a new engine for file access and management, and also includes enhanced verification and auto-correction facilities. In Verma’s words, ReFS is optimized for “extreme scale” to “use scalable structures for everything.” Interestingly, ReFS volume will never be taken offline: even when corruption occurs, unaffected volume elements will remain available and accessible. And a new resiliency architecture (when used in concert with Storage Spaces) will help to protect and preserve volume contents. Verma’s list of key features also highlights plenty of new capability:
- Metadata integrity with checksums
- Integrity streams providing optional user data integrity
- Allocate on write transactional model for robust disk updates (also known as copy on write)
- Large volume, file and directory sizes
- Storage pooling and virtualization makes file system creation and management easy
- Data striping for performance (bandwidth can be managed) and redundancy for fault tolerance
- Disk scrubbing for protection against latent disk errors
- Resiliency to corruptions with “salvage” for maximum volume availability in all cases
- Shared storage pools across machines for additional failure tolerance and load balancing
Some changes underneath, but not much change on top
What Remains in Common between ReFS and NTFS?
Not surprisingly, Microsoft has also elected to “maintain a high degree of compatibility with a subset of NTFS features that are widely adopted” although they do intend to “deprecate other … [features] … that provide limited value at the cost of system complexity and footprint.” This means that BitLocker encryption, access-control lists (ACLs), the USN journal, change notifications, symbolic links, junction, mount and reparse points, volume snapshots, file IDs, and oplocks will remain the same as those used in NTFS.
However, there will be no conversion utility to transform NTFS formatted volumes to ReFS. Instead Microsoft advocates copying data from NTFS into ReFS to make the change, and also to allow the resulting ReFS volumes, file collections, and data sets to enjoy the benefits of the new ReFS structures, maximum sizes, and resiliency. For the time being, ReFS will not be used to boot volumes in Windows 8 (though I’ve seen some discussions of making ReFS volumes bootable for the upcoming version of Windows Server due out late 2012 or early 2013).
What’s Gone from ReFS?
Verma does address this as “What semantics or features of NTFS are no longer supported on ReFS?”
The NTFS features we have chosen to not support in ReFS are: named streams, object IDs, short names, compression, file level encryption (EFS), user data transactions, sparse, hard-links, extended attributes, and quotas.
This is an interesting grab-bag of features, among which there are some obvious items: EFS was never that popular, and had sufficient problems to make lots of enterprise users steer clear, short names have been passe for 10 years or more, and named streams (aka “alternate streams”) have posed some clear and present security dangers. Extended attribute elements (also including object IDs) support the distributed link tracking service, which can be convenient but also had some problems, as did sparse files and hard links. It’s a little surprising to see file compression and quotas going away, and user data transaction handling was never implemented much anyway.
Once again, it’s “in with the old, and out with the new.” This should give Mark Russinovich and his colleagues something interesting to update in the next edition of Windows Internals, which should give me an opportunity to understand the impact of these changes in more detail.