Windows: Microsoft Beating on a Dead Horse

[mattmatik]

i think what kills vista is that win xp left little room for improvement. sure for us technology gurus, there are always features that we would like to see, but for the average user, there isnt a lot of incentive to leave windows xp.

I agree and I also think that for folks who follow the development of an OS and who know the advanced workings of an operating system, it's much harder to remove yourself from that knowledge and put yourself in a normal, non-knowledgable user.

[Denis W. Veteran]

Vista has major issues with resource usage. Just ask Brad Wardell or any other IT professional actually testing out Vista trying to see how it will handle "real work". [...]

What are you going to say when it comes out and it is a total letdown? Are you going to ignore the problems and say it is better than XP? Because right now, Vista is a lot worse than XP.

First point: care to share some links where IT professionals or 'professional' journalists indicate the latest Vista build being a resource hog? (on a personal persepective, sure, it's slower than XP. Sure, the memory usage is quite high - but isn't that due to the fact I have 2 GB instead of <1 GB of RAM? in any case the latest build wasn't the slow choppy OS it used to be for the builds in the past.)

If Vista will still remain as a resource hog right through RTM, then hell, it'll be fixed over time - just like what happened to OS X. It was (according to many) a slow crapper from 10.0 to 10.1, but now 10.4 runs well.

Second point: Sure, in the opinion of some who've spent the last few years reading about and testing this new OS from Microsoft that went though a lot of development hell, Vista's a letdown. Guess it'll be the job of you guys to go around and "educate" the average Joe users buying new Vista OEM machines why this OS is "crappy" compared to XP. :p

Vista may not have been what it was, but I'd at least like to see MS offer something new that improves on a couple of things instead of sticking the middle finger at users and saying "Screw you, just keep using XP until we work out something new in a few years." Those who wish to jump ship, go right ahead. If MS loses enough marketshare because of their sloppiness, that's their problem, not ours.

Just look at IE and Firefox. Firefox was enough to get the IE team off their sleeping asses and do something. We may expect to see a repeat of that if Vista turns out to be that crappy.

[phantasmorph]

"Screw you, just keep using XP until we work out something new in a few years."

Um...it's been a few years, and they still have barely managed to improve a couple of things. Arguably Vista is a huge step backwards from XP in many areas, most notably the user interface and usability. I just hope this catastrophe is the reality check that Microsoft needs to get their asses in gear again.

[Computer Guru]

First point: care to share some links where IT professionals or 'professional' journalists indicate the latest Vista build being a resource hog? (on a personal persepective, sure, it's slower than XP. Sure, the memory usage is quite high - but isn't that due to the fact I have 2 GB instead of <1 GB of RAM? in any case the latest build wasn't the slow choppy OS it used to be for the builds in the past.)

Hmmmmm. Perhaps you care to explain this:

http://neosmart.net/blog/archives/191

[ScottKin]

What an odd article - from an equally-odd website!

I remember seeing that logo for NeoSmart somewhere before....ah, now I remember...

IT WAS ON THIS VERY WEBSITE!

I remember the logo as being one that was posted and the poster asking for feedback on the logo - I think it was within the last year or so - for a new website.

Anyhow - the article written has all of the earmarks as an amateurish attemtp at trying to look "smart". The author has absolutely no idea about the importance of "legacy" applications support on the corporate level and the investment that companies make in software and upgrades. Does the author belive that companies are going to dump entire investments in applications just so they can work with the latest software that won't be backwards-compatable?

The author, in all due respect, is writing his article from only one viewpoint and one stance. Such "journalism" is amateurish and can actually make the author look as amateurish as their article.

Microsoft's "backwards compatability" is provided in the same fashion as it was when Windows NT rolled-out - an address "thunk" process which handled memory address and processor translation between the 32-bit long-address and the old-school 16-bit "segment:offset" memory addressing. It was identified as "wowexec" (wow = Windows-on-Windows). Vista uses a similar scheme to provide backwards compatability, just as the 64-bit version of Windows XP does.

Again, as I said before - the author of the article apparently has very little understanding of business processes and decisions. The humorous aspect of this article is the same as every other pre-release analysis of Windows Vista that is on nearly every other technology & IT forum done by amateurs - IT'S NOT EVEN A RELEASED PRODUCT!!!!

To all out there who are thinking of doing a similar article or write-up on Neowin or any other blog or site - remember that we're still more than 6 months away from Vista's actual release. Wrap your hands in duct tape until then and refrain from making silly and/or stupid commentary until the actuall RTM"ed product is in your hands....it only makes you look foolish when the actual product is released and all of the whine-points you made in your article are non-issues.

--ScottKin

[ScottKin]

And I'm so sick of people who don't have a bloody clue about software development in general let alone issues involving backward compatibility making such fanboy comments.

Vista has major issues with resource usage. Just ask Brad Wardell or any other IT professional actually testing out Vista trying to see how it will handle "real work". If you cannot see the problems, then I'm afraid that you are blinded by being an overly optimistic fanboy.

What are you going to say when it comes out and it is a total letdown? Are you going to ignore the problems and say it is better than XP? Because right now, Vista is a lot worse than XP.

The problems with Vista have nothing to do with backwards compatibility. That is just an excuse apologists use to defend MSFT. The problems with the Vista project go much deeper than minor issues with preserving backwards compatibility. I would say that maintaining backwards compatibility only contributes to more potential security flaws and bugs than anything else.

Well, let's see here - I think with my 25+ years in IT/IS/MIS/DP, I think I would classifly as an "IT professional"...and yes, I'm on the Vista Beta.

I use Vista on a daily basis in a real-world environment - my own office. The problem is that we have too many people running memory-meters and figuring that all of that memory that Vista gobbles-up is a memory-leak or poor memory management - when it's actually quite the opposite.

All of that memory that a user sees missing when running Vista is taken-up by prefetch data and the new "SuperFetch" schema that pre-loads logical-next-point data and actually let's Vista run faster in most cases. Vista will flush and relinquish the prefetch and SuperFetch buffers and caches when more memory is neede for applications. I've read Brad Wardell's take on the Pre-Beta 2 release (5381) and I'm suprised that a developer with his reputation and background (I am an ardent fan of Stardock products) would write an article like the one you're probably referring to ( http://www.wincustomize.com/articles.aspx?...&AID=117896&c=1 ), but the author of the article on NeoSmart appears to either have just skimmed Brad's article, didn't understand what Brad was saying or put his own spin on Brad's article and took bits and pieces out-of-context.

Sadly, this article and this thread doesn't "look better" from being posted on Neowin - but it's definitely "unprofessional" - I mean, doesn't Neowin claim that it's a place "where unprofessional journalism looks better"?.

--ScottKin

[Matt T]

Getting back on topic, I have to agree with this article. Windows is a decent OS, but Microsoft tend to rely on old technologies to get the job done, and that really isn't something we should expect from the world's largest software giant/empire with billions of dollars at their disposal. And I know this has been said before, but take EFI as an example - it's faster, it's more efficient, it's modern, yet Microsoft refuse to support it in Vista although Apple, who are a relatively small company compared to Microsoft, added support for it in a simple OS update (please don't mistake this for a flame - I'm just giving an example). Stranger still, EFI is already supported in XP x64 and, even worse, Windows Server 2003, which (correct me if I'm wrong) is what Vista itself is being built on! I'm finding it very odd. :/

And again, I agree that backwards compatibility is what is holding Windows back a lot. Think about it: if Microsoft suddenly removed backwards compatibility from Windows Vista, it will be the developers' responsibility to update their programs to support Vista; Windows is obviously what most developers code for - they wouldn't not update their programs because eventually everyone will move up to Vista and the developers' market share and profit will descend. At least Microsoft could do what Neowave suggested and create some sort of "classic" environment for older applications. Or better yet. create a coding language that won't become outdated as fast as current languages.

[Computer Guru]

And again, I agree that backwards compatibility is what is holding Windows back a lot. Think about it: if Microsoft suddenly removed backwards compatibility from Windows Vista, it will be the developers' responsibility to update their programs to support Vista; Windows is obviously what most developers code for - they wouldn't not update their programs because eventually everyone will move up to Vista and the developers' market share and profit will descend. At least Microsoft could do what Neowave suggested and create some sort of "classic" environment for older applications. Or better yet. create a coding language that won't become outdated as fast as current languages.

The thing is, they have.

.NET Framework is just that. It's a dynamic language. The code developers write today hooks into the framework, which then uses APIs and such to get the deed accomplished - but for some odd reason it doesn't work that way.

No one is suggesting removal of backwards compatibility *cough* ScottKin *cough*, just implementing the compatibility *outside* of the kernel (like the article outlines in depth, and Neowave explains above).

[XerXis]

anyone who call's HIMSELF a computer guru doesn't go high on my ranking. I tend to follow scotkin on this.

Vista uses 500 mb of ram on my computer, that's only a quarter of what I have available, why worry? As said before it's mostly for the superfetch. Apart from that, it's still beta2, there are still unoptimized processes, hell, there is even some debug information left. Vista is stable and fast on my computer, just stop looking at the memory all the time, because it doesn't work that way anymore, and that's what we call innovation.

[kyosuken]

Maybe Microsoft didn't want to add the layer outside of the kernel (as if it was some kind of say vistualisation, the way os9 launch through osx gives that kind of feeling) for performance issues, the way it's compatibility works right now give the same if not more performance as if it were run on the native OS. The only bad thing is that it's not "automatic", and it should be. The other thing i can see is that unless the app was designed for the 9x core, it maybe difficult to differentiate out of the blue the apps built arround the NT core (unless building a huge app compatibility list...).

It may have been easier for Apple because OSX and OS9 don't share at all the same core, but Vista afterall is still based on the NT line...

This is my 2 cents on the matter anyway, as long as devs are not going to use .NET as a primary platform, we will be stuck with backward compatibility for a long long time...

[SaLiVa]

I think the article is right. Vista will probably be a success - A huge one I might add, but after that - There's very little chance. Microsoft made so many promises - So many promises that were in the end, unfulfilled (Of course, we have to wait for the final build to pass judgement). Consumers are in the end quite confused as to whether or not they should upgrade or stick to what they have. The marketing of this OS is not so great, and people still believe that it's just an improvement over the last OS. My main concern and my reason for this is because they're still continuing the Windows trademark.

I hope after Vista, Microsoft will use another name - Doing so will change the Microsoft image, not just as a leading software house, but as a creative powerhouse in the industry.

[.Neo]

Emulating a 'classic' Windows XP and before environment in an upcoming Windows OS may sound like a good idea - but as much as an advantage it may be, there's the problem of a speed hit in the emulated OS. At least for now, the compatibility settings in XP/Vista are nothing more than changing a few settings instead of emulating an older core.

The speed hit in Classic is only 30%, which isn't even noticeable on the latest PPC models anymore due to the advancements in hardware over the last years and the fact that Mac OS 9 hasn't been updated since 2000/2001. So basically a 2006 PowerMac G5 minus the 30% speed lost is still way faster than any Mac in 2001.

Same will happen almost instantly to Windows ME and below and in time to Windows NT 5.x.

[5Horizons]

...And I know this has been said before, but take EFI as an example - it's faster, it's more efficient, it's modern, yet Microsoft refuse to support it in Vista

The x86-64 version will support EFI. I'd guess that almost every computer sold with Vista pre-loaded will have the x86-64 extensions now that Intel has finally gotten the ball rolling on supporting x64, so this really isn't an issue.

Or better yet. create a coding language that won't become outdated as fast as current languages.

Are you kidding me? C and C++ have been around for years and they're STILL not outdated. Language has absolutely nothing to do with Windows' backwards compatibility.

Anyhow - the article written has all of the earmarks as an amateurish attemtp at trying to look "smart". The author has absolutely no idea about the importance of "legacy" applications support on the corporate level and the investment that companies make in software and upgrades. Does the author belive that companies are going to dump entire investments in applications just so they can work with the latest software that won't be backwards-compatable?

Well said.

And again, backwards compatibility is not a big issue that's holding Vista back. I'm still waiting for someone to post articles that prove otherwise...

[Fourjays Veteran]

And again, backwards compatibility is not a big issue that's holding Vista back. I'm still waiting for someone to post articles that prove otherwise...

On reflection, I'm not sure holding back is the right term really. It is part of the problem maybe, and even then it isn't really "holding back". Surely though, backwards compatibility in its current state causes problems with developing new features? MS can't just add a nice new feature, they have to make sure it doesn't stuff anything else up at the same time... I doubt it can be good for security, stability or reliability either.

[+Warwagon MVC]

Emulating a 'classic' Windows XP and before environment in an upcoming Windows OS may sound like a good idea - but as much as an advantage it may be, there's the problem of a speed hit in the emulated OS. At least for now, the compatibility settings in XP/Vista are nothing more than changing a few settings instead of emulating an older core.

Well we do have dual core systems now so if they were to incoperate it I don't see a speed problem.

[libertas83]

As ScottKin said, businesses require full backwards compatibility and not some emulation that will be much slower on most systems. Remember corporate user's often do not get the latest PCs and are often stuck on crappy machines. Also, a person's position in the company tends to lend weight to who gets the new cool fast computers. So emulation will not work at all because it will be very slow.

Don't forget that when Vista is released, 99% of all apps will be legacy apps thus 99% of apps need backwards compatibility to function. Vista uses a new rendering system (DWM) that creates two copies of the Window in memory to render GDI apps. Why? So we can run all our basic programs we expect to run since they aren't DirectX apps.

Why is Explorer and other apps not using .NET Framework? Cause they are still creating the framework and making changes. MS developers have admitted on Channel 9 how stressful and crazy it was to develop during the Longhorn days because they would have rewrite everything constantly.

Anyway, back to the main points in the article. Consumers and businesses only use the software and don't care about the techie stuff. They just want to run their software and new stuff is cool, but they just want a smoother more stable OS. Longhorn promises, get over it! It is not important because half of the promises can be done by 3rd party developers. They still have 2 of the 3 pillars of Longhorn in there.