Why Windows Vista will be better

Thank you Ronjor, interesting reading.

It looks promising anyway.

 

I skimmed through this article. While some interesting points were made. It felt like an out & out sales pitch. So I said to myself. Who wrote this? Couldn't find the authors name anywhere. I wanted to see if the author had any M$ ties. That it's unsigned speaks volumes to me.

 

Thanks ronjor for the name & link.

I'd still like to know if the author is biased towards M$. Either by renumeration & or affiliations. All I found out so far is that M$ likes him enough to link to one of his reviews from an M$ site. http://www.microsoft.com/windows/windowsmedia/press/presscoverage.aspx Under March 2004.

 

I edited my post b4 I saw the ty.

 

In the runup to Vista's launch there will doubtless be many more articles like this - portraying Vista as a giant leap for mankind ("one small step for Microsoft...") while ignoring its downsides which so far include:

No support for DVD drives lacking firmware region coding
Affects old DVD drives (pre RPC-2) but also recent ones if they have been flashed to allow multi-region playback.

New monitor needed to view Hi-Definition content
MS and the MPAA can share the blame for this - monitors (and televisions) will have to be HDCP-complaint to be able to display hi-definition video - those that aren't will receive (degraded) standard quality video instead.

Vista Will Encrypt Hard Drives by Default
While some may see this as a useful security measure, in practice this will also make it very hard (possibly impossible) to install an alternative OS (like Linux) since the installer will not be able to adjust the Windows partition (e.g. to shrink it to make room for a new one) and may not even be able to make any modifications to the hard drive without causing data loss. It will also likely make many image backup/restore utilities obsolete since image restores normally have to be done outside of Windows (hence losing the ability to write encrypted data using the TPM).

Vista may shorten the life of flash memory
To quote from the ExtremeTech article's first page " SuperFetch also takes advantage of external memory devices—plug in that spare 256MB USB key (any size will work, really) and Windows can cache a lot of the working set to it. It's not as fast as your system RAM, but it's much faster than randomly grabbing small bits of data from all over your hard drive."

Using flash memory in this way is a bad idea - flash's lifetime is determined by the number of writes that can be made (typically in the 10,000's) so having Vista automatically use it as a cache (with the frequent updates that may involve) will result in a shortened lifespan.

Vista won't work with many graphics cards and will remove games developers' ability to cater for older systems
From the sixth page: "Perhaps one of the best features of DX10 is the removal of capability bits, or "cap bits." Today, graphics cards don't have to support everything in DirectX 9 to be a "DX9 graphics card." There are lots of optional features, and the drivers have to report to the OS exactly what it can and can't do with a set of cap bits."

Cap bits allow developers to identify what features are present and therefore make it possible for them to produce games that run on older hardware while still exploiting leading edge features. By removing this ability in DirectX 10, Microsoft not only "encourages" developers to produce Vista-only games but also limits the ability of graphics cards manufacturers to add new features without having a totally new version of DirectX which supports them.

Drivers have to be signed for x64 versions of Vista
If you have hardware whose manufacturer goes out of business, forget about third-party updates. This also rules out any user/enthusiast modifications (like the Omega drivers for Nvidia graphics cards) or beta releases (which are often needed for with graphics drivers to fix problems with specific games). The now-mandatory testing and certification by Microsoft for every release will likely increase prices for hardware and software needing drivers.

MSDN Channel9 has an Interview with the Vista Kernel team and when I watched it, the following points came up:

• Windows is hopelessly monolithic - the team mentioned over 5,000 dependencies that had to be charted which creates a massive overhead of testing for any change. This alone greatly restricts what changes can be made suggesting that Vista is more of a rehash than a rewrite. One basic tenent of computer science is that your code should be as modular as possible so changes made in one section do not have an impact elsewhere.
• The Windows Registry, with all its problems, is still going to be a key part of Vista.
• The one genuinely new security enhancement I could pick out was the idea of "trust levels" for individual modules. If structured (and enforced) properly, this could add significantly to security - but can hardly be considered an innovation since some mainframe operating systems (e.g. ICL's VME) had this feature in the 1980's.

Ultimately for an article that finishes off with "It's the unknowns that keep us from giving an early wholehearted recommendation.", it seems to make great efforts to avoid the likely downsides.

 

Thanks Paranoid2000. I too share the point of view that all will not be Rosy with Vista. One thing conspicuously missing from most reviews of Vista are the increases in price of replacement hardware. I built my XP box a little over 2yrs ago. The tower cost less than $600 with all hardware included. Rock solid parts was my aim. With the decrease of prices I could shave at least $100 off that price now. What I read about the hardware requirements for Vista. It will cost at least double maybe more to put together a Vista tower that won't perform as well as my XP box.

 

• New HDCP compliant monitor requirement. As you note, I don't think you can really pin the blame for this one on Microsoft. HDCP is an industry-wide nightmare in the making for high-def consumer electronics that is being forced by MPAA and other content owners. The next-gen DVD formats are about to screw over the HDTV early adopters as well since the absence of a HDMI or DVI/HDCP interface will result in down-rezzed content. Analog HD component inputs and/or pre-HDCP DVI inputs will result in an EDTV signal rather than full 720p/1080i/1080p glory. Moreover, according to FiringSquad don't plan on being able to upgrade your PC with a next-gen drive and expect HD playback with a new monitor either. Apparently, not a single graphics board has the necessary keys to support HDCP even though many GPUs have been marketed as "HDCP capable". Nor can you let monitor makers off the hook either, since they have known for several years of this impending situation as well. The high-def content protection schemes are a train wreck in the making from all accounts, with essentially no one company bearing full blame nor completely escaping partial liability.
  
  
• Harddrive encryption. Valid points. However, unlikely to cause insurmountable problems. Active windows partitions will likely be resizeable via Windows utilities. And, of course, inactive partitions can clearly be deleted and reconfigured by any number of utilities. Many people choose to do clean partitioning, or utilize entirely separate physical drives, for alternative OS installs in any event. As to imaging, I do not know enough as to the details to comment fully... but perhaps a bit-for-bit copy, albeit encrypted, may still render a valid and usable image. On the whole, however, I believe the support for optional full-drive encryption constitutes a net positive.
  
  
• SuperFetch and USB drives. I share your concern here as well. I will not likely make use of this feature. Thankfully, according to Paul Thurott's WinSuperSite this feature is optional as is illustrated in screenshot which follows my commentary
  
  
• Cap bits support. Unless I am mistaken, cap bits will still technically be supported in the API (as is likely required for backward compatibility). However, new graphics boards that seek the DirectX 10 compatibility certification and logo will be required to support all DirectX 10 mandatory capabilities, thus rendering the issue of GPU cap bits moot in the eyes of future game developers. Again, I see this as a net positive. I believe history tends to show that Microsoft was typically ahead of the game developers and, often, even the GPU manufacturers themselves. So, often it was Microsoft encouraging devs and GPU mfg to advance their algorithmic support... not the other way around. In fact, DirectX specs have tended to be designed largely in partnership with collective vision and implicit acceptance/encouragement of the GPU manufacturers. Cap bits presented a maze of backward compatibility nightmares for game developers. I doubt few will lament their demise.
  
  
• Kernel-mode driver signing. Much driver development has been shifted to a user-mode driver model where feasible, thus rendering the signing issue moot for a sizeable subset of functionality. For the remainder of kernel-level code, I confess that have no real qualms in the new arrangement. I am willing to accept some nominal decline in driver support and/or nominal increase in driver development costs if it means validated kernel code and a resultant sizeable increase in overall system stability and security.
  
  
• Channel 9 Interview. I also watched that interview and it was somewhat informative although often vague on specifics as well. They seem genuinely interested in addressing many of the issues you noted as best they can while not completely destroying backward compatibility. The system inter-dependency issue in particular seemed one in which the particular developer was especially passionate in attacking. Neither he (nor I) harbor any illusions that all non-essential dependencies will be eliminated overnight. He, nevertheless, did seem to be fairly well versed in many of the root causes of the bulk of the dependencies as a result of his code reviews and automated tool analytics.
  
  
• Performance concerns / hardware requirements. Undoubtedly there will be some intrinsic elevation of overall system requirements. This is, unfortunately, the nature of the information technology beast. However, do not forget that for the most part, the Aero Glass interface which is largely responsible for the often discussed heightened system requirements is not a mandatory component of the operating system. The GUI can be scaled back to a very XP-like display mode which I understand has hardware requirements and performance characteristic quite in line with XP.

.

 

Good article!!!

Thanks

 

Good article even for little fish like me

BUT:

Agree with P2000 (as usual pithy and illuminating posting!! )

The Windows Registry, with all its problems, is still going to be a key part of Vista. SECURITY!! The initial attempts at Vista were aimed at getting rid of the registry. (and now ...including the MAC OS ...we drift further and further from UNIX)

Hard Drive encryption: to the too simple minded like me this could be seen as a potential "lockout" strategy by MS in the guise of seccurity

The article seems to point to less and less support for 3rd party goodies

Looks like we'll all be shelling out for bigger and better hardware?

"too many unknowns...."

Just my simple observations

 

That's yet another in a very long list of reasons why i continue to harp at Microsoft ridiculously silly neglect toward 98's and Me's instead of revisiting those again and completely re-write with the modifications that could win them some real support again.

From most everyone i've talked with over this, Apple is going to be the chief MASTER in their plans when they decide for a new OS system and certainly never Microsoft's Vista's. How can consumers and users trust a lame OS maker who are chickens with long yellow lines running down their back, scared to death of their own work and products. Scared they are of facing 98 and Me again with new full rewrites like they should in order to reach some parallels on the order of XP, and then even add other improvements to surpass maybe even XP? Of course not, in keeping with their famous stupidity they choose to avoid them like the black plague and announce ending support completely just like they think they are going to do for XP.

I don't give Vista a 1 in 100 chance of going anyplace but the scrap heap for the majority of business and home owners. Of course the ritzy few will always plop down those demanded dollars just so they can pretend they are up with the times.

[mini-rant concluded]

 

Windows Vista is definitly a no-go for me. In my book, Windows XP is the last version of Windows that will be installed on my personal computers. Microsoft aims too much at nice and facy stuff. They just don't come with good new technologies.

For example. The upcoming KDE4 desktop will have almost the same eye-candy as Windows Vista, however the KDE developers managed to get the recource usage even 30% less then the KDE3.x series. Quite amazing, I would have liked to see this kind of innovation by Microsoft as well.

 

Good points from you also Alec - just a few I'd like to expand on:

What is likely to worsen the situation here is that most people will receive Vista pre-installed on a new system from a major player like Dell. They typically no longer receive the original CD/DVDs removing their ability to reinstall and taylor the system to their preferences (including removing all the pre-installed garbage that Dell, etc receive payment for). This will mean that the only method for trialling Linux would be via a Live CD (Secure Startup permitting), wiping the existing Vista install or adding a new hard disk (assuming that TPM encryption doesn't automatically lock that up also). At least, the encryption seems to be an option but the main question is whether it is enabled by default on OEM setups.

Being selectable does remove one concern, though the dialog really does need a prominent warning that flash memory could have a shortened lifespan if used in this way.

Didn't DirectX 9 have minimum (i.e. mandatory) standards for pixel shaders (via SM 2.0) with options (SM 2.1, 3.0) being added later? (reading from Features of the Next DirectX). DirectX 10 may be setting a higher bar (be interesting to see how many current high-end cards meet it) but it will surely have optional features added later on.

At the very least, cap bits allowed a developer to check for basic features and, if not supported, to give an appropriate error message. Given the wide disparity in capabilities between bottom- and high-end graphics chipsets (which is likely to continue), the alternatives would either be to write for the lowest common denominator or to include a detailed specification list on the game package (e.g. "Graphics card supporting Tessellator Enhancements and Unified Shader Model required.") and leave the choice up to Joe Consumer.

This feature is one that appears to need co-operation from other vendors to work. Given the questionable record by many of the top tier manufacturers so far, how many are likely to write drivers that will work in user-mode? However the key issue I see here is user choice - x64 Vista users are having it removed.

The only real solution to this I can see would be a complete from-the-bottom rewrite with a compatbility mode to handle older software. This would have been most easily done with the move to 64-bit processors (which would have required many such changes anyway), but this does now seem to be an opportunity lost with some key architectural "features" (e.g. the Registry) being carried across and even made more complex (the installation of 32-bit programs to "Program Files (x64)" - what happens if you want to install software in another folder entirely?).

While some of XP's slowness compared to Windows 2000 was due to the UI (see Infoworld: Waiting for Windows XP), there look to be other factors at play also. However the Tech Report: Windows performance comparison focused more on graphics and gaming and produced almost equal results on high-end machines, with Win2K having the lead on low-end. It will be interesting to see comparative benchmarks with Vista, but I'm sceptical that disabling the UI alone will cure all performance ills.

 

It will only be better if it runs with less bugs and issues that the OS its supposed to replace. It simply wont happen, companies are not going to move to vista for a LONG time (im thinking a few years), only the clueless home users and small businesses who get sucked in by the PR, marketing and advertising will jump on the bandwagon.

Edit, of course there all those phreaks who can't live without the features

 

There is one fly in the ointment for Windows XP users with regard to Vista - Microsoft can force an upgrade by withdrawing the Product Activation service for XP. Even those with validated copies would then face the risk of having functionality disabled if they carried out one hardware upgrade too many so would feel pressured into purchasing a Vista upgrade "just in case".

I should point out that Microsoft do address this in their Product Activation FAQ in the following section:

Will Microsoft use activation to force me to upgrade? In other words, will Microsoft ever stop giving out activation codes for any of the products that require activation?

No, Microsoft will not use activation as a tool to force people to upgrade. Activation is merely an anti-piracy tool, nothing else.

Microsoft will also support the activation of Windows XP throughout its life and will likely provide an update that turns activation off at the end of the product's lifecycle so users would no longer be required to activate the product.

However the cynical may wish to highlight the phrase "will likely provide an update" as being a less than 100% guarantee, especially if Vista sales do poorly.

 

I said it before and I`ll say it again. Vista will not get anywhere near my machine. I`ll stay with XP till Gates stops help with it. Once that occurs it`s Linux or to a Mac system.

 

Indeed. You're not alone with that view.

 

I won't be switching. If I switch to anything it will be to Ubuntu Linux. I've been using their LiveCD an awful lot lately. I have been considering dumping Windows, as my primary OS, for a long time - but Ubuntu Linux is the first thing that has really tempted me.