Problems? See this thread at archive.org.
Nice! I like playing with underwear too!
-
It's amazing what the 970m has done for the form factor, cooler and faster than the 870m
-
back in the day there actually were hot switching solutions in both red and green camps, but then came along the dawn of automatic switchable graphics, a far more "convenient" solution for the "lazy" computer user
-
-
-
-
-
-
-
u get backpain from 2-3kg? gee, might consider doing some sports
-
-
-
It's a GTX 550M.
Hence my current search for confirmation It does seem the newer Clevos work fine, but I want to be sure.
My current laptop was specced on the basis of needing to run multiple VMs, so the lack of accelerated graphics wasn't too much of an issue. Now that project is over and I've got a lot more free time for gaming so I don't want to commit to a large expense and end up with the same problem. It does seem the newer Clevos work fine, but I want to be sure. -
-
-
What about Windows 10 technical preview? Will windows 7/8 drivers works there? I have a license for Win 7 Pro and also considering Win 10 TP. Also, will they release an image for windows 10 consumer preview? Sorry if I'm slightly off topic
-
-
I meant, will Microsoft release a consumer preview after they showed it to the public? But looks like I will just play safe and go with Win7 Pro
-
Microsoft has a number of programs so you can receive this type of information. Some of them are within the reach of the end-user, although costly (e.g., MSDN), while others are OEM, Solutions Provider and related.
I.e., your best option is to get to know someone at a local Microsoft Solutions Provider, ideally Gold, as they get all sorts of this info.
As an original NT 3.1 Beta Tester at the largest installed base of the first, native NT application, I used to be on several, internal lists until just a few years ago. Today, I don't know much other than what people feed me indirectly, and my past experiences involved with many programs.
I also keep my MCP credentials current, even though I am usually the "Linux expert" on-site (and sometimes even more MCP credentials than most of the Windows staff, beyond my deep internals). I even was offered, repeatedly and by first-hand, personal referrals, to become a "Master" (on-site at Redmond for 3 weeks) on Directory Server, because of my strong NT and AD-LDAP-Kerberos practice at major customers.
But my access finally got yanked around that time too because I was, despite my extensive assistance in AD migration and related LDIF-fu, considered to be working "outside the Windows ecosphere" and could be in a "competitive situation." I.e., I know it came up during the "Master" program consideration, and the last thing "a few people" (and I stress "a few people," as most were professional about it) in Redmond wanted to see was their public "Master" list of Microsoft employees, key Solutions Providers and others "in the ecosystem" along with this single entry of one "(redacted Linux vendor)" employee.
In other worse, I screwed myself by becoming too well-referred with NT and AD history and internals, and especially in one where I cost Microsoft mid-8 figures a few years back, no less on the desktop-related front with my extensive, embedded Linux and Windows background. Someone made the fatal mistake of saying Linux couldn't be a generic, combinational desktop-PoS Kiosk, and 2 days and 1 prototype later (fully skinned and retail themed in ways impossible to do on Windows), I was into a 12 week project that had about 400 people scrambling to undo the "damage" I caused ... literally and I mean literally overnight.
heibk201 likes this. -
If that's the case then NT 6.2 wouldn't share drivers with NT 6.1 tho because of string names, but as you said they do share drivers, so I'm guessing they are modding ini files in the first place already aren't they before releasing the drivers?
and a 12 week project was your only consequence lol?
-
Sorry, to clarify, version string. Version strings are evaluated very differently, often being broken down in the major, minor, revision, build, etc... That's so they can use integer operations on then.
E.g., 6.10 != 6.1 but 6.10 > 6.1, while 6.1.0 = 6.1 and, most importantly here, 6 = { 6.1, 6.1.0, 6.10 }
I.e., now in the case of NT ...
String with NT major version "6" = NT minor versions { 6.0, 6.1, 6.2 }
String with NT major version "10" > NT minor versions { 6.0, 6.1, 6.2 }
Version major, minor, revision, in a string, is extremely common in coding. Look at any INI/INF file, let alone if you disassemble NT binary/library objects. I deal with this a lot because I modify a lot of [open source] code too.
It's the difference between earning only a quarter-mil in services, and a sale of 7 figures ... while, more importantly, ensuring the customer doesn't have major costs and issues (security is a big one) in the future. In the end, the customer cut their nearly 9 figure expense in half, with Microsoft eating half -- not just losing sales, but having to subsidize entire support staffs, with external parties, at the customer. But they will have the same issues for the next 6 years, like they did the prior 6 years by forcing their entire business to use desktops with Windows.
I certainly don't do "open standards" based solutions for the money. I could make millions a year in various commissions integrating proprietary, lock-in solutions. As I always joke, you know I'm all-about-the-money if you see me working with Oracle (Microsoft isn't the only one out there).It's a vicious cycle whereby "open source" is used as a threat to get a better deal, even though it locks the vendor into costs long-term. It's business and finance folk only looking at 24, 36 or, best case, 60 months down-the-road. Microsoft aims for 24 months, settles for 60 months. Engineers like myself look at decades, at least 5 years in advance, especially those of us with semiconductor and, far better yet, aerospace (25 year planning) backgrounds. But I do see a lot of business people get fired over this stuff, and blackballed as a result. If they are smart, they leave the company within 3 years -- enough time to "save" money on the sheet, but not long enough for the "future costs" to hit them, hard.
This is not commonly talked about because these are very, very "sensitive" contracts, but extremely common practices, and I've been in so many (all under NDA of course, so I cannot name parties, obviously). Every now and then, the "big ones" hit the media, so everyone knows the "jist" of it, while people like me know all of the gory details. The London Stock Exchange (LSE) finally destroyed Microsoft's final, "last bastion" in FSI. Microsoft's biggest "loss leader" was the London Stock Exchange (LSE), with Microsoft funding Accenture (porting from POSIX/Java to Win32/.NET). Over 5 years, Linux 2.2 to 2.4 to 2.6 response time improved 50x fold over NT, and Accenture had major issues porting.
Once the LSE went down for almost the entire, same day Freddie Mac and Fannie Mae were "nationalized" by the US gov't, that's when they got the boot. Microsoft blamed Accenture, but Accenture was funded by Microsoft (part of the "arrangement"), at a huge loss, so Microsoft didn't lose their "last bastion" of the high-speed, mega-money FSI industry. Sometimes saving money really costs you in the end, especially as the LSE was just getting killed by every Linux-based trading house, which had become the overwhelming majority. Even non-RT Linux was killing heavily RT-modified NT (with 2 patents licensed from a Linux-centric developer house, ironically enough).
Microsoft and .NET have now been regulated to Bloomberg and Reuters-related, non-real-time desktop software in FSI, and they'll never regain that foothold. Oh, you can find plenty of jobs doing that, the majority. But the real money is in the real-time trading, and Windows is no where to be found.
Retail has been a constant battle for them, even controlling Best Buy's distribution at one point -- e.g., that's when Apple products, sans MS Office for Mac, were removed from the shelf overnight (the iPod's timing being their "bargaining chip" to get Mac products back on the shelf within a couple of years), only to lose them. Micros and other industry relationships have attempted to change that, as Microsoft long assumed Windows would get the back-end as much of the front-end by '00, but never planned on a desktop, let alone a better embedded/kiosk platform, solution presenting itself when they only saw Novell as a server-only competitor.
In this case, it was actually the customer's the 2nd time in 7 years they looked to go "standards-compliant," and they came very close as I had Firefox and GNOME completely profiled (mandatory polices -- yes, you can do it, and manage it easily, in Linux with LDAP and/or CM without costly, $300 CAL add-ons for AD) and setup for accessibility limited employees, and it was very locked down compared to Windows. There were huge advantages to the solution (using $99 PoS systems, price at 1K quantity) over Windows, especially in replacement and related support costs outside of software and services (easily $500/year that went away). But ultimately it was a decision they made 6 years earlier that gave them the "reasoning" to stay Windows.
They had built their entire intranet on MS IE 6-only tagging and ActiveX (yes, security issues and all -- which is why so many companies get hacked despite firewalls and layer 7 filtering) which many argued against doing 6 years earlier. This meant their "costs" were heavy in development. This meant there was the added, "conflict of interest" type "incentive" for 400 developers to keep Windows the platform, especially when we provided case studies on reducing the number to 40 for Internet standards-based development. So the arguments weren't technical, even if we provided the major TCO reduction in the next 6 years, compared to the former. As Gartner said oh-so-well in 1999 ... no proprietary vendor will offer a way out of lock-in, a customer must choose to do it and stick with it.
In reality, I don't consider Microsoft proprietary. They purposely break compatibility with their own, undocumented, proprietary formats. That's called purposeful "abandonware." People complain about ODF-based Office suites not doing the latest MS Office formats "perfectly," but even MS Office has issues between versions. At the same time, if you use ODF, it works pretty well everywhere, and when compatibility is "broken" in a new version (e.g., 1.0 -> 1.2), it's very well known and documented, with the older version being "read-only" and the new version being created with many warnings. MS Office 2013 at least does a better job at this with very good warnings, but you have to understand it took 3 versions before Microsoft recognized the "Transitional" 2007 and 2010 had only created new, "abandonware" non-standards too that enterprises and, even more so, governments were tired of.
Case-in-point: MS Office 2007, 2010 and 2013 all have different "Transitional" Office OpenXML (OOXML) formats, which are not ISO standard aka "Strict" OOXML. It was so bad with 2010 "breaking" tags with 2007 (let alone 2008 v. 2011 for Mac, with 2008 missing entire functionality versus Windows), that Microsoft now has no less than three (3) "Compatibility Modes" in 2013 -- including for 2007 and 2010, not just 2003 (like 2007 and 2010). Of course, only 2013 offers a "limited, strict" ISO mode, and several tools don't work, and definitely not by default. Which is why I encourage everyone to setup their MS Office to always produce 2003 formats by default, which are the only formats that MS Office 2007, 2010 and 2013 (let alone 2003 itself) actually "agree upon," at least when the newer tools support 2003.
This is why, even with ISO standardization (don't get me started, ISO OpenDocument documents many MS Office standards better than Microsoft's own, "written overnight" 1/10th the size ISO standard reference), the UK and several other countries are finally starting to mandate OpenDocument Format (ODF) instead. Boeing has supported it since Day 1 (surprise, surprise, #1 engineering technical documentation company in the world needs a 25 year format), Corel-WordPerfect from Day 1 (filters to/from ODF work great) along with IBM-Lotus and many others have adopted it over the years for this reason. Microsoft cannot even make MS Office compatible with MS Office, let alone Mac compatibility (especially sending back documents to Windows) can be a real PITA.
Which is why I always joke, "MS Office for Linux would have as many compatibility issues as MS Office for Mac, whereby long-standing (over a decade), deeply documented -- first by OASIS ("the" XML standardization organization) and, later, ISO -- ODF does not, and never will." If you care about compatibility more than 60 months, you have to break from MS IE and Office, which are purposely designed to keep you upgrading and breaking compatibility. Now that said ... Office 365 does end that, because it's "subscription based," and no longer relies on upgrading, so it will eventually get to ISO OOXML (Office 365 on-line has compatibility issues with legacy MS Office and the 3 "Transitional" formats, but I won't go there), plus solves the "Mac user" issue (and will eventually run on Firefox on Linux). Even Google recognizes Office 365 is becoming a threat to their Docs, so they are moving to add more complete ODF support (along with getting "more serious" about Linux support -- duh, your own Android!).
This separates me from being "just a Linux bigot." People call me one of the most objective "trusted advisors" they've ever met, because I know all of the internals and politics, and how to get them the best deals, including long-term TCO. But in the end, they make their own choices, and I hate being right when it costs them dearly.
In fact, the thing I hate in the IT world is that "job security" phrase. No, you should document yourself out of a job. That's how you grow and get even bigger jobs. If you wonder what is wrong with America as a "producer," instead of the consumer, it's that people think that way -- instead of focusing on efficiency, reliability and ... my personal favorite ... willing to forgo familiarity to see "what else exists."Last edited: Dec 31, 2014heibk201 likes this. -
There are a lot of driver issues in general, because Microsoft didn't ship a "generic" XHCI driver. I could really go deep into this, but ... to the original point ...
The lack of a SP2 for Windows 7 is the main problem for installers today. Microsoft expects everyone to be an OEM and build a WinPE and other, installer environments with the drivers necessary to install a system. This is getting beyond unhelpful for the typical Windows technician. You literally have to know how to do OEM and/or enterprise-type pre-installation and deployments, modifying images, etc... to do Windows. It's ironic that Linux is now 10x easier to teach in this regard (first-hand dealing with both), but that's another story (especially on imaging v. automated install/configuration -- imaging is not faster, and definitely not better, "in general").
So -- back to the end-user argument -- Microsoft really needs to continue to release updated SPs/SRs so people don't have to know those details, just to re-install. They cannot expect end-users to be IT departments with desktop deployment specialists.
As far as everything else ... more off-topic ...
I don't "hate" Microsoft. I "understand" Microsoft. That's the difference. I've been a long, long time insider. I'm well respected for my objective views based on specific, in-depth knowledge of why the "technical decisions" are not always "technical." This includes Microsoft's own, horrendous record on both MS Office file format compatibility (with itself, including in the alleged "ISO" standard era) as well as security (which is a long story on differences between Windows and Linux library approaches, with pluses and minuses to each).
But the NT boot loaders and general MS flexibility in boot has been pretty much laughable. They always assume they are the only OS on the system, and even have conflicts with their own solutions at times. People assume otherwise, but when you support a lot of platforms and solutions, you see a lot of major differences and resulting limitations in the NT approaches.
Understand my first really "bad taste" of Microsoft as a MCP was back in 1999, when I stumbled upon issues with Exchange's SMTP/RFC822 solution. A 3rd party package was crashing an Exchange server. The MS Solution Provider was blaming my DNS and MX records (how the heck can MX records crash a server?). After sending various malformed RFC822 headers to it, I was able to crash it at will. I reported it to the MS Solutions Provider managing the server, who immediately read me the riot act, and threatened me. By 2003, the same issues were now allowing black hats to compromise and take over Exchange servers, not merely crash them.
Early 2003 was also SQL Slammer, and I was working at a Fortune 20 company. Two fixes unpatched the patch that would have prevented SQL Slammer, and it was only because I had the print version of Network Computing -- with the original article that confirmed what I had already told management from various, prior release notes (all on-line versions were yanked) -- that I saved a lot of people's jobs. It was the first time I had ever seen MS and their Solution Providers throw their own MCPs "under-the-bus," and I really was able to "push back" with detailed information -- in private, of course -- that countered their public statements.
It was then that Microsoft finally started taking security seriously. This was also a Gates move, because before then, security teams at Microsoft were basically "shut out" of meetings. Although the first new "Security Czar" they hired was canned within 6 months, not long after he admitted -- publicly -- that no version of Windows had ever been designed for the Internet. He was a bit too honest.
Not actually until 2009, per Gates himself (long story). Most everything else was faux, including "Shared Source," which everyone saw through. IP is IP, it's all about how you "share" it when it comes to true interoperability.
In fact, Microsoft went to Red Hat first, for an interoperability agreement, because MS Solution Providers were continually losing deals because they could not also support Linux. Red Hat shot it down when Microsoft pulled out the "IP license" contract, so Microsoft was forced to Novell-SuSE. Go forward a few years, after that debacle (and, sadly, the end of Novell-SuSE which are now 2, separate Attachmate companies with a lot of former SuSE people becoming fellow Red Hat colleagues), and Microsoft was back to asking ... as originally ... Red Hat. Red Hat refused the IP agreement again, and Microsoft finally caved on an "interoperability agreement" without an IP agreement.
The rest is history, especially with some key moves Gates made in Microsoft in 2009, despite his reduced authority. I am very, very critical of Gates and some very poor decisions he's made in his time at Microsoft (don't get me started), but several of his late moves were actually very "eye opening." His decision in 2009 to actually push others in Microsoft to finally stop "actively fighting" open source, under countless guises, was a good one. Of course Microsoft continues to court several of my colleagues, and most of them have turned down job offers (very appealing ones, overall compensation-wise), largely because of the continuing IP "issues."
But every company with a proprietary product line has their IP "issues." Only companies like Canonical and Red Hat, which are GPL-centric and make no money on IP, are free from those. Yes, Red Hat continues to register patents, but purely defensive ones, via their Patent Promise shared with an ever increasing number of companies. The legality of the Patent Promise has been covered many times now, as a company cannot promise to share it freely, perpetually, and then yank the terms.
So understand my "views" of Microsoft are from the standpoint of a Microsoft Certified Professional (MCP), and not "hating." It's called being a responsible peer. I don't like the 99% of people who just complain about Microsoft endlessly, over money, etc... They just drown out legitimate concerns of peers. My concern is always how difficult Microsoft makes things for its own professionals, sometimes expensing them.
But yes, eventually I had established myself as a broad expert, so I was brought in as an integrator, and am now a leading consultant in the open source infrastructure space having done most everything from embedded to large scale integrations, plus some custom development here and there. I.e., most people who are "open source" experts are generally "experts" on all sorts of "technologies" and not just "products." E.g., if you are a deep master of Samba services, you actually understand how CIFS/SMB works at the lowest levels. LDAP is extremely useful for AD (where 99% of AD "architects" don't understand it to their detriment), etc...
That's why I am heavily and I stress heavily utilized even by Windows departments. I know exactly why something was done, especially back in the '90s and still affects things today, and the continued hacks and workarounds.
Again, the entire NT3-5 NTLDR was extremely limiting. The Windows Pre-installation Environment (WinPE) for NT5 helped a bit, but it still had a lot of limitations. Even the NT6 BOOTMGR today has a number of limitations that no other OS has.
I.e., even the uEFI firmware is far more flexible than NT6.
More compatible? I think you mean more sustainable? Microsoft is losing money on a lot of their desktop solutions, and a "subscription" model would help them return to profitability in those sectors. That's just smart business.
But yes, Microsoft has had to finally take interoperability seriously. They had to. Part of the reason is legacy x86, which allowed their developers to ignore things like data alignment and various issues. With the Internet generation, everything is streamed bytes and network order now. That has forced them to change their approach.
It also helps that .NET is based directly on licensed Java code (originally 1.1 which Microsoft "won" the rights to continue using, even if they could no longer call it "Java," then 1.4+ with the re-license for .NET 2+), which is very POSIX-like and more portable. So the more .NET "infiltrates" Win32/Win64, the more it helps too. That's always been the view from the POSIX (UNIX/Linux) world, more .NET = more portability.
Unfortunately there is just a lot of Win32/x86-only out there. I.e., .NET libraries and features with a C/C++ dependency on a library that is only available for Win32/x86. That's why Mono failed, utterly. I think Mono did a great job of exposing how inter-mingled Win32 with the GUI, such as the task management being tied to WinForms, and requiring the WINE emulator to emulate portions of the Win32 executive, if one built for WinForms (instead of GTK#, which was cross-platform).Last edited: Jan 1, 2015heibk201 likes this. -
-
-
-
-
-
Sager NP8651 / Clevo P650SE with GTX 970m First Look
Discussion in 'Sager/Clevo Reviews & Owners' Lounges' started by HTWingNut, Nov 5, 2014.
