Tuesday, November 27, 2007

Update: Re-Testing Vista w/2GB RAM, Office 2003

Many of our members have requested that we re-test Vista SP1 with 2GB of RAM instead of the 1GB we used in our original tests. So, without further delay, we present our revised results table:

image

Figure 1 - Revised OfficeBench Completion Times (Seconds)

Analysis: By providing Vista (SP1) with an additional 1GB of RAM (that's a total of 2GB for those of you keeping score) we managed to achieve a "whopping" 4% improvement in OfficeBench throughput.

Note: We added the Windows XP (SP3) results to the chart to add further context to the Vista results. As before, all tests were conducted on the same Dell XPS M1710 system w/2GHz Core 2 Duo CPU and DDR-2 667MHz RAM.

A few members voiced their concerns over the use of Office 2007 under Vista. They suggested we re-test using Office 2003 on both Vista and Windows XP. Here are the results:

image

Figure 2 - Office 2003 vs. Office 2007 Completion Times

Analysis: Moving from Office 2007 to Office 2003 definitely improved Vista's showing. Instead of being over 2x slower than XP on the same OfficeBench workload, Vista is now "only" 1.8x slower.

To quote Darth Vader: "Impressive...most impressive."

60 comments:

Anonymous said...

I'm curious about Windows Server 2008 (RC0) performance. Could You please check that system?

Anonymous said...

Sadly, this won´t make moaning to stop, but to enhace, because they complain like kids and vista is their new toy.

"Don't say that my toy is slouch, you mean, i'm going to tell mommy that you aren't running vista with 8 gb and a dual four core like Microsoft daddy wants, so people doesn't notice is a hog!"

If it were any other OS, they will say "look, we knew all the time, don't dear to choose out of monopoly, let's better spend money like good consumist"

Then Monopoly guys tells that open source is a cancer, when Vista is a metastasis on memory and resources.

Good review, but unblinded people knowed this, and blind people doesn't care, they like blindness.

I would not care, if microsoft policies and pressure where not so harmful to plural development.

Anonymous said...

Any suspicions as to why Vista is so sluggish? I.E., what component is sucking up CPU cycles, RAM, and the hard disk?

It would be very interesting if you could re-run the tests with Vista with the following conditions set:
- Aero and DWM off (i.e., the "Windows Classic" visual style)
- Indexing service disabled
- Defender disabled
- Shadow copy and automated backups off

Anonymous said...

Exactly what does Office Bench test? It seems strange that even game performance isn't at 50% of XP whether GPU limited or CPU limited. Is there a link with more details of the benchmark?

Research Staff said...

OfficeBench uses OLE Automation to drive Microsoft Word, Excel, PowerPoint and Internet Explorer through a scripted series of business productivity tasks. These include the creation of a compound document with multiple sections; a supporting spreadsheet with multiple chart objects; and a presentation with multiple slides.

The script makes use of a variety of common application functions, including:

1. Formatting/scrolling, copy/paste, search/replace and print preview/printing (Word)

2. Scrolling/selecting and chart generation/printing (Excel)

3. Creation/formatting/slide style application, sorting and printing (PowerPoint).

The script also makes use of IE to simulate access to an online knowledge base-type site. At the end of the script you have a completed document, supporting spreadsheet and slide show suitable for conducting an internal meeting or similar business case scenario.

Note: You can learn a lot more about OfficeBench by simply downloading. It's part of the DMS Clarity Studio testing framework and is available for free through our web site: www.xpnet.com.

RS

Anonymous said...

I think it is alright that Vista is running "slow" now.

Remember the days with Windows 98SE? It runs really fast. It took me 2-3 years to fully switch to Windows XP since it first came out.

Vista is slow. Think about Windows 2k and XP, how many drivers does these OS contain and how many in W98 and Vista? Can you compare that? An updated OS needs to recognise some old hardware for compatibility, also, include NEW drivers for new hardwares!

It is pain in the S to find USB, 1394, etc, drivers on Windows 98!

But well, it's better with W2k, then a bit more better in XP, and I think will be better in Vista. All these "recognition" and "make use of" the new hardware certain takes time and resources.

There is no free lunch, including the features you have with a new OS. New hardware will develop in time, although not exactly this moment.

Remember back in year 2001 how many MB of memory you have, as typical user? 128Mb? 256Mb? 512Mb? How do you feel at that time your experience with XP? I think it sucked. But now, with better hardware, faster HDD and more RAM?

Vista will make use of these, in time, in the future. Until then you will find it good to use. My 2 cents.

DARK said...

I`dont know why is this noise all about!?Vista is slow!No you are not agreed !? Ok Vista is fast your machine is slow !? Is that better ? Now Try to run 3DS MAX on XP and on Vista to render.Result on XP less time.Conclusion the machine is enough fast Vista is slow!I can rotate the things a million times and Vista still will be behind XP in performance!On an ordinary NoteBook it is real pain in the ass!Try this take your brand new NoteBook run some test write somewhere the results Install XP (this will require format i think) Install Windows XP (on some new notebooks you will need a special patch of microsoft drivers for the sound) and corect drivers run the tests again and compare the results.Before someone start doing that i want to tell him that there is no point in it because XP again will perform faster. Have a nice day and you the folks of exo.blog don`t try to convice mad man that he is mad.There is no way for him to understand that he is mad not the others (that didn`t see how fast is Vista).

Anonymous said...

I'm convinced since you've retested. But does MS acknowledge speed optimizations or are they coincidental?

Anonymous said...

Hmm, that's very interesting then. Have you performed tests to isolate the slow point of Vista? With so many testing factors, it would be instructive to discover if there's a(re) particular bottleneck(s) in Vista.

Anonymous said...

Many of the arguments I have read over the past couple posts on this topic seem to display confusion. One person points out one thing, another points to a different and slightly off the commented topic. The process continues exponentially. This is where a judge would pound his gavel to get some order. Honestly I think most of the people here are right just in different senses.

I should point out two main scenario's that OS performance testing usually target. One is for the home/small business consumer while the other is for large corporations or government entities. They aim to help the target determine if upgrading is reasonable and take into account all of the externalities affected. Because the two main customer types are extremely different, the results can also be. Not to mention the fact that there are tests which take place in the middle or that test certain aspects directly.

When I hear the comments about memory being cheap… Well, yes this is very true, and in fact if this article was about home/small business consumer OS advice you might indeed be right. Again though, this depends on multiple scenario's. I have had a better XP experience on my HP Athlon X64 3000+ w/ an Nvidia (330 maybe) card and 2gb than with Vista. Now on another laptop, tablet actually, 2.something Core Duo w/ a compliant Intel graphics card, I felt Vista run noticeably better. One of my desktops that is almost 2 years old is an Athlon 64 X2 Dual Core 4400+ w/ a 7800GT and 2gb ram, XP couldn't compare to my experience. Perhaps this is because I have maybe applications open at once with multi tasking taking place in every corner. When I try the same things in XP the system would be unresponsive until one application had finished its operation(s).

Although, the topic being blurred with XP now involved I'm not sure what the post would try to answer? "Consumers - Upgrade to Vista or stay with XP now that Vista SP1 is released". Would you assume the user would be using Office 2007 regardless of their OS? I read a few people requesting Office 2003 be tested on Vista. Would it not be fair to assume that it would be just as reasonable, if not more so, to test Office 2007 on the XP setups as well? Office 2007 absolutely uses more resources than 2003. Perhaps another good test might be to run either Office 2007 or Office 2003 on Vista and XP while both are doing Virus scans or burning DVDs.

On the other side of the customers, you have a corporation who has to do an implementation of 100's or even 1,000's of computers, the $50 memory price tag increases just a tad. Then you have to incorporate the actual implementation and training costs occurred. I did read one or two people commenting on the performance gains (I have read) in which Microsoft has stated SP1 brings to the table: IO gains, network transfer fixes, etc. Those fixes are something that a large corporation would most defiantly take into account. Although they still have to decide on an Office upgrade as well as an OS upgrade.

These corporations usually take many more things into account though. They look at their help desk logs and try to identify problems in which Vista (and SP1) would fix. They estimate the time of their help desk as well as the user down time. And the opposites as I mentioned before, training costs, etc that would determine if the upgrade is worth wild. That is just a broad overview.

When you are reading any of these testing articles, the first thing you need to be aware of is the target of the testing. Does it apply to you directly, indirectly or does it even apply to you at all? If you are a XAML .Net 3.0 programmer and work directly with XAML animations than you need to look into different Vista performance examples. If you are working in 3D Studio max, well you just might want to stay on XP for now. But the advances in best utilizing multithreading will not be present within XP. Besides someone in that industry usually tries to use the highest notched hardware and is probably already on Vista with a quad core. I think this was just some user who doodles (I could be wrong). Graphic design, well if you have a high end pc (2 years ago) Photoshop CS2 smokes on my Vista RTM and it does fine on XP although with Vista, the quantity and size of the open images is increased.

Ultimately, I believe there is a bigger picture here with where they are taking the OS. Major advances in programming frameworks intertwined with graphics and all of the vast information available forced Microsoft to head in a direction. They do need the critical feedback so that the issues can be addressed. Mainly, what works for you might not work for someone else. If you voice your story, great! List your hardware configuration as well as your machine setup and you could possibly help someone make a decision. Otherwise, saying yours works great and the other option sucks is just plain ignorant ranting. Criticism and constructive criticism are two different things, try to stay constructive.


Zac B.

Michael said...

I am curious as to which version of Vista was tested.

I have heard that this test was testing Vista Ultimate and that running Vista Home Basic does not show the performance lag. I tried finding the information but must have missed it.

Thanks.

James said...

Did anyone else actually run the test? On my Vostro 1700 (1.6 C2D, 4GB), XP and Vista x64 both take 58 seconds.
So much for "XP outshines Vista"...

Research Staff said...

James,

We're skeptical of your results. After over a year of testing various Vista flavors on a variety of hardware platforms, we have *never* encountered an instance where performance was on par with Windows XP.

We'd be very curious to see your raw data. If you've uploaded it the Portal site, please let us know your account number so we can examine the OfficeBench test output. If not, then perhaps you can save the Clarity Studio Session and send us the .cfw files. You can email them to support@claritysuite.com.

Anonymous said...

And how is the improved performance for the 64 bit versions of XP and Vista???

For those of us with 4GB or more.

I even remember the day's when we had 64kB and Bill'G(e) said that he couldnt foresee the use of more ram than that ;-)

Research Staff said...

Actually, it was 640KB - some of us were there when he said it... :-)

RS

fjarlq said...

"Actually, it was 640KB - some of us were there when he said it... :-)"

Ah, perfect.. I've been looking for an actual source on that quote for some time. Where/when exactly did Bill Gates say that, and what did he actually say?

Anonymous said...

I just think it’s funny how people try to compare different OS's designed for different hardware. As each new generation of an OS is released its tailored to the latest 'average' pc. You get more features, more pretty curvy interfaces but that takes more hardware requirements. The same is true for both Linux and windows. Try running NT4 on current hardware it’s stupidly fast but doesn’t have all the features you'd expect of a modern day OS. Compare a cut down Linux disty such as DSL to Fedora 8. DSL: fast, light, less features (earlier lighter kernel). F8: slower, ‘bloated’ but compatible, comprehensive (newer larger kernel). Think of NASA, the first shuttle in space was by today’s standards a microwave in terms of computing power, but it did its limited job. Today they send amazing amounts of power into space to analyze a million and one different science projects. I guess I’m really just getting at that point of its utilisation of available power. Don’t upgrade your existing hardware and of course there’s going to be a performance hit.

Anonymous said...

And yet, in many ways, on every Vista system I've touched in the last year, I've noticed slightly better performance from a subjective point of view.

However, where in any of the SP1 docs, does it say that SP1 will give better performance to office 2003? What? It doesn't? You mean this is a service pack for operating system functions, and not applications? Who knew?

Anonymous said...

Run the comparison again, only this time measure the things that a lot of us do everyday, such as file transfers & copying, and even booting up and shutting down.

horsecharles said...

I also vote for a test with any 2K version.... In turn, it/they should kick XP butt.

As well I'd be desirous of a 64bit compare... include Linux too(ubuntu please), as well Mac-- even if with a different app.

horsecharles said...

please do a 64 bit version w/ 8gb ram, quad core test with mac/ubuntu/vista/xp/2K

Anonymous said...

Vista does more than XP (feature wise and behind the scenes). Perhaps, benchmark Vista against XP by turning off Vista features such that it's feature parity with XP.


A better benchmark is Vista (a '06 OS vs OS X Leopard (a '07 OS).

James said...

I emailed my results to the developer, so we will see where that goes...

Not surprisingly, hard drive activity can make a significant difference when running the test: compare 73.7 seconds with Vista reporting 100% hd active time with a mere 56 when completely idle. XP clocked in at 57 seconds, so it appears that Vista wins(!)

I'm disappointed that more people haven't run OfficeBench for themselves ... in the meantime, tons of major news sites have already reported that "Windows XP with the beta Service Pack 3 has twice the performance of Vista" - but my results contradict that. Guys, it's free software ;)

DARK said...

The two OS are not so much different!It is not like to compare 9x core with NT core!!!to compare NT6 core with the new one sounds not so different!And what means developed for different hardware?!!?There is no way to talk that this is like comparing NT4 core with the latest OS.It is more possible to compare the NT5 core cause NT4 is tottaly different from the next one!Vista is for x86 and 64-bit platforms XP too and both of them can use all instructions sets of the porcessors!So for what difference speaked somebody !?If the two architectures are different i suppose Vista have to be more fast on the latest CPU-s and MB-s but obviously it is not it is even slower!What a shame!The difference between XP and Win 2k is not so big like the difference between XP and Vista if we talking for performance!And isn`t the performance the main thing that we whant from our hardware and software!?Because if it is not why everybody is russhing into the shop for the latest and fastest CPU-s , MB-s , VGA-s , RAM-s ?!

Anonymous said...

Hello, what is the software used for the benchmark, I will try on my 1520 Dell, 2.2 core2, 4gb, 7200RPM HDD.

XP and Vista.

Thanks

Research Staff said...

We agree with James: Everyone should sign-up ASAP and run their own test scenarios to see what they see.

As for us, we just revisited the tests (again), stripping Vista to the "bone" by removing all of the eye candy and background services. We managed to get Vista closer (~40% slower vs. ~80% slower in previous attempts) to XP, but nowhere near on par.

RS

Research Staff said...

Once again, we're using DMS Clarity Studio for testing. It's available as a free download from our web site, the exo.performance.network (www.xpnet.com). Just register for a portal account, login and download the software.

RS

B-Net said...

Windows 98 is faster than XP and Vista, so let's use that,

Anonymous said...

Hi

My result is 38 (Ofice 2007)on Vista 64 bit.
My system: Windows Vista Ultimate SP1 v.275
Intel QX6700 OC to 3.2 GHz
Intel BAD AXE2
2 150GB WD Raptor Raid 0
4 GB Kingston PC 8500 RAM
ATI X1950XT
Dell display 24"
Perhaps someone has a system close to mine with XP sp3 on, so we can see how Vista is, on a faster computer.

Anonymous said...

Do these office benchmarks have to do anything at all with closing and opening menu's and sub windows ever? as animation of effects in Vista takes longer than in XP, I'd like to know the results of these office benchmarks when all the animation options etc are turned off, both with and without DWM enabled.

It would be far more interesting to compare xp sp2,sp3, vista rtm (no patches), vista rtm fully patched from win update, and vista sp1 then.

(as MS often state SP1 is no biggie as the patches are mostly on win update anyway)

Yakumo.

Brian said...

In my own experience, with my AMD Athlon XP 1600 with 512MB RAM in 2002, I had one hard drive with Win98 and one with WinXP. The performance was much greater on WinXP than Win98. There were lots of great reasons to use XP with all the improvements.

Now, with Vista, it is much slower on even new hardware and what are the benefits that really matter?

James said...

After squashing an issue that resulted in IE freezing in the middle of the test, I was able to run a "full set" of 10 iterations:

XP: 47.22
Vista: 55.44

According to this Vista takes 17.4% longer - no "stripping" was done here. To ensure fair results, I found it necessary to wait at least 10 minutes from startup before testing in either OS.

(Customer ID is 56568 for those privy to the information.)

Anonymous said...

Contrary to popular believe, MS has a lot of Vista patches that are not on Windows Update.

Here's something interesting: http://wiki.winehq.org/BenchMark-20050419

Wow, isn't XP great?

The question one needs to ask oneself is this one: would you upgrade to Vista considering the good & bad things?

For me that answer is yes, until someone can give me the same Vista interface under XP. And I do not mean Aero.......

BTW on behalf of all the people who wanted to see test-results with 2 gig of mem & different Office systems, thank you for doing what we wanted.

Bill said...

I've a P4 1.7GHz with 768MB "only", a Geforce 6200 with 256MB (Aero enabled!), Hard Drive IDE 120GB 7200rpm and Vista is faster than XP, Vista boots faster than XP, I can launch applications as the desktop appears, no slow downs, all applications runs great without performance problems on my old PC.
I have all features enabled by default.

Anonymous said...

another user wanting to see Server 2008 RC0

Anonymous said...

All these tests were performed with one motive in mind.. to slam Vista and make it look like it is a slow and resource hog.
I can do basically the same thing with the same result in reverse and make XP look bad. All depends on what you have running in the background.

Anonymous said...

Vista--the OS designed to turn a Intel Core 2 with 2Gb of RAM into an 8088. --AC

Nigel said...

ive got a dual opteron 170 @ 2.5ghz 2gb ram running vista 64bit, i get 71 out of that benchmark, you'd think there core2duo about would flog mine.. i dont think this test is really a test to say if a operating system is faster than another, there's alot more to an os than word and excel!

Anonymous said...

quote
"After over a year of testing various Vista flavors on a variety of hardware platforms, we have *never* encountered an instance where performance was on par with Windows XP."
/quote

http://www.firingsquad.com/hardware/amd_nvidia_windows_vista_driver_performance_update/

Anonymous said...

I also would like to see Windows Server 2008 RC0. I would test it myself, but I registered for the testsuite and never got my e-mail for activation.......

Research Staff said...

Regarding your email with account info, we've had some trouble with DNS recently and that caused some of our SMTP traffic to get lost.

Simply visit www.xpnet.com, click the Login link and then use the 'Forgot Password' option at the bottom of the login screen.

Or you can email us directly with your registration data and we'll verify it manually. Send the email to support@claritysuite.com and someone will get on it.

RS

Anonymous said...

Here is the thing. Somebody said that Vista is designed for different hardware or better hardware. You're totally wrong because XP is future proof OS with full support for multicore CPUs, lots of RAM, PCI Express 2.0 and what else not. Simply Vista sucks, total failure as a project.
I claim and i will still claim that superfetch slows down Vista big time...

dishwara said...

I got the result 172.2 by running clarity studio.

My pc configuration.
windows xp (sp2) 32 bit & uptodate, office 2003.
Pentium 3- 550Mhz, 192MB Ram, 4 MB sis video card, asus P3B-F motherboard.

I hope this will be useful to someone.

Anonymous said...

I suspect Vista really is a bit slower than XP, but what concerns me the most about these numbers is the complete lack of details of these results. Go visit a gaming or hardware site to view a graphics benchmark article, there's excruciating detail about the machines, driver versions, hardware setup, build number, system configuration. This allows the general public to look into the numbers and make their own conclusion. How about sharing that kind of detail for these results?

And my 2 cents, benchmark software is just that, a benchmark, it's rarely indicative of real user experience. It's helpful in getting a handle on specific performance metrics, but they almost never tell the whole story.

Anonymous said...

I haven't done and side by side testing of the two OS's yet, but I ahve to admit my experience so far with Vista have not been positive.

I'm still not sure what the true advantage is to upgrading yet. Performance issues aside the one major improvement was supposed to be a new file structure, but they coudln't make that work so they dumped it. I'm not all the big on window dressing so the aero interface doesn't really appeal to me. The reviews on the supposed security improvements are mixed at best.

Given a couple of service packs and another generation of hardware immprovements will definitely help Vista I am sure, but why upgrade now?

Anonymous said...

Thank you for the benchmarks. The overall experience is certainly important, but for an OS that's being marketed as next generation, this is a whopping speed difference. I'd really like to see you compare 64-bit systems with 4GB to see if the difference is still so vast when you have time.

Anonymous said...

Thanks, I can only agree with the test. I installed office 2007 once on a vista 64bit and it totaly slowed down the computer. Without office Vista seems much faster.
I did a lot benchmarks for Video and 3D Software and my expirence is that Vista is about 30% slower then XP, only on rendering it was 10% faster. Also on 64bit systems it has some extras that a worth using it.

Krishna said...

Windows Server 2008 seems to be a major upgrade internally. RC1 is out and is available for public testing. It starts up pretty faster (but shuts down slower) and runs applications faster. Also the annoyingly slow File Copy operations (in Vista) are faster in RC1 - as much as XP. The best thing is improved application compatibility. All the apps that failed to start in Vista runs well with Server 2008 (from RC1 onwards). IE 7 seems to be more stable and WMP11 performs well - without eating much resources.
I run 3 instances of MS Visual Studio 2008, 7 IE 7 Windows (each with so many tabs), IIS 7.0 web server with SMTP enabled, MSN and Yahoo messengers, FTP Service, SNMP Service, SQL Server 2005, MySQL, WMP playing my favorite collections, etc.., AERO with all effects turned on. There is no sidebar - ofcourse. But the system shows no sign of slowing down.

I run this on my HP Pavillion laptop with Intel Centrino Duo 1.73 Ghz, 1 GB RAM ("only" 1 GB), 80 GB HDD, nVidia Geforce 7400 with 256 MB RAM (128MB dedicated RAM (another 128 MB is shared from 1GB, so Im left with 768MB actually)).

I have worked with XP, various flavors of Linux, Vista and Server 2008.. XP and Server 2008 are by far the best!

I will update my test results soon
...

Anonymous said...

The thing I notice about OfficeBench is that there's one critical system that could be a bottleneck. OLE automation.

Suppose OLE Automation in Vista were significantly slower than XP. It wouldn't matter how fast Office *ACTUALLY* ran, you'd get slow results because of the OLE Automation part.

Now, i'm not saying that this *IS* a problem, but that's why most reviewers do multiple benchmarks to try and give a larger picture, and so that any limitations of a single test would be noticable.

voodoo said...

Odd results.. I run the script and it gets completed in 25.88 seconds.

Windows Vista x64
MS Office 2007
Intel 6750 (2.66/333 FSB) C2D
2gb DDR2 @ 1066


Something sounds fishy...

Anonymous said...

Here is somthing for you to think of. Movie encoding - unlike office bench, video analysis in near 100% math and hard drive utilization barerly makes a difference.

In my experience averaging 18 - 20 hour jobs on Vista vs. XP - XP beats Vista by about 12%. In addition 64 bit XP is faster then 32 bit XP by about 15%. 64 bit xp and 64 bit vista - XP wins with about 17 - 20% difference.

On 64 BIT OS I used 64 bit applicaiton chain, on 32 bit os, obviously 32 bit one.

I was not running a comparison but encoding HD movies the observatoins were really a side effect. Overall, an average 1080p with 2 pass encode runs for about 20 hours.

Here are system specs:

Core 2 Duo E6600 OC to 3.2Ghz (8x400)
2GB of Corsair XMS10000 Memory Running at 1000 Mhz at 5-5-5-15 timings [note the 400x4 Mhz FSB]

All of that sitting on top of Asus P5K deluxe (P35 chipset board) with RAID 5 Drive array (3x 250GB WDC 16 MB cache drives)

I really have no favorite horse, I was just looking for a tool that would allow me to acomplish the work the fastest.

colorado coonass said...

I'm a reformed Windows Luddite - I even wrote an article for my local computer monthly mag (ComputorEdge) defiantly announcing my allegiance to Win95/98SE and denouncing XP for the kluge I thought it was (the idea being that there were many, many 100-200 MHz Pentiums out there that would run perfectly well on pre-XP operating systems) and explaining how a cash-strapped would-be Windows user could get highly acceptable results without peeling off the bucks for XP.

Since then, I found a good XP laptop at a hock shop, tweaked it and cleaned out all the spyware and other viral crap the previous user had on it. Sure enough, it WAS faster than my beloved old 98SE system, and as the system matured Microsoft distributed upgrades that made it work faster still.

Problem was that the upgrades and all the apps and data I'd accumulated were outgrowing the physical limits of my XP system (not to mention my middle-aged eyes weren't so great with the 15" screen on my XP laptop).

My wife, the one with all the compassion in the family, gifted me with a Dell Inspiron E1705 sporting Vista Home Premium. It was a reconditioned model from Dell, 1 Gb of RAM, Core Duo T2350 running at 1.86 and 1.87 GHz, respectively.

This was April of last year, and Vista was not impressive to say the least. I stuck with the process of hanging out on Microsoft's Vista "community" forums (where the sum total of Microsoft's staff support was one guy who hung out on the forums to smart-ass the users who posted requests for help getting Vista to do something besides crash) until through a tortuous process of try, try again I finally cobbled together a Vista system that worked.

Then I joined Microsoft's Consumer Experience Group and every now and then answered questions about my experience with Vista, doubtless had my system quietly interrogated and updated by Redmond, and got a nice, expensive suite of apps software from Microsoft in exchange. Not to mention a system that worked a whole lot better by January.

My point, though, is that of course the performance difference between a Vista system that's been quietly upgraded since release every time it connects to the Internet and a system just like it which has just had SP1 installed on it will be minimal.

It all depends on what you mean when you say "RTM." Released to Market without any upgrades since initial installation, or properly maintained and upgraded ever since release? There's a huge amount of difference.

Flug New York said...

I have installed SP 3 on my XP Pro. After reading your post I thinking of deinstall it and return to SP2. What do you think of it?

colorado coonass said...

"Flug New York said...

I have installed SP 3 on my XP Pro. After reading your post I thinking of deinstall it and return to SP2. What do you think of it?

May 19, 2008 2:36 PM"

I haven't tried SP3 on any of my XP SP2 systems, so I can't help you with any advice. One of my "I've got to do this next week projects" is to clear enough space in my work area to leave several systems running spontaneously for comparisons.

I have several XP2 laptops, so I can actually upgrade one to XP3 - after benchmarking its performance before - and compare the XP2 and XP3 systems head-to-head.

I'll let you all know what I find out when I do all the above. I suspect moderate improvement over the systems I'd been tweaking along before my Vista unit arrived.

Anonymous said...

This test is grossly retarded. I know its old as hell, but i cant pass up the chance to point out a bunch of numbnuts whos taken up Apple Marketing's job of making Vista look bad.

Deon said...

The question is not about upgrading to Vista or not. The question is: is the hardware required affordable or not? I have a 2.4Ghz quad core with 8GB ram and a 256MB ddr3 graphics card. Vista 64 bit with sp1. Sp1 was a definite improvement on the stability and usability. My system is blistering fast with as many apps open as I can use at a time. We know every OS that comes out needs more resources to function optimally.

colorado coonass said...

MY hardware is a refurbished Dell Inspiron E1705 laptop, Intel Core Duo T2350 running at 1850 MHz in both cores, 1 Gb RAM. It cost well under $700, and has a 17" screen and a comparatively sluggish ATI Mobility Radeon X1400 video board built in.

The video board is what drags my potential system performance down; but disabling the semitransparent windows and other gimmicks associated with the "Aero" screen interface removes most of the performance hit associated with the obsolete video hardware.

I've had occasion to start my old laptop (a Dell Inspiron 1150 with a Celeron processor running at about 2250 MHz) and was struck at how very slowly this system ran compared to my new Vista system.

Of course, Dell knows how to integrate a system to get the maximum performance from its components. Who knows how much of this gap in performance was due to the dual-core processor and all the other goodness Dell has added to the Inspiron line since my new laptop has come out.

My point, though, is that my system by no means qualifies as a cutting-edge system - it's fast, but nowhere near as fast as what's out there now at the top of the Dell line. You don't need a super-fast, top of the line system to get good performance from Vista.

colorado coonass said...

Hey, XP was a slug compared to Win 98 SP2 about a few years ago, as I recall. It's not a thundering indictment of Vista to say that it's a little slower than Win 98 SP3 now.

I don't think Vista is "slouch" - turning off the goofyware, like half of the settings for "Aero" which add NOTHING to the user's experience as far as I'm concerned takes care of the performance hits I've noticed with Vista. Now my Vista laptop zooms along very nicely, despite not being state-of-the-art (it's still great for what I do).

Anonymous said...

It's sad that people are still trying to use that Windows 98 to Windows XP vs. Windows XP to Windows Vista comparison.

Windows 98 and XP are completely different. Windows 98 is DOS based and retains 16-bit code. Windows XP is NT based and a 32-bit design.

Now XP and Vista are much more closely related. Both are NT based 32-bit designs, and the 64-bit versions of both are basically more along the lines of what Windows 98 was in comparison to Windows 3.11.

A more realistic comparison is Windows 2000 to XP vs XP to Vista. I find a much larger performance difference between XP and Vista than I did between 2000 and XP. On average, I find the same hardware to provide a faster experience on even a cluttered XP system than a new Vista install. Pretty sad.

I still find it funny that my old 1 GHz Celeron laptop with 384 MB RAM running XP Home runs circle's around my friend's 1.8 GHz Sempron with 512 MB of RAM running Vista Home Basic. And that's without much junk on my friend's system and plenty on mine!

Anonymous said...

i ballieve there is no reason to pay money to switch to vista especially with windowns 7 coming out in aruond a year. and really why would you i allway keep the same system through the life of my computer becouse thats what my computer was ment to run there fore it does it well. but consider this xp will be seen as a crappy system some day just like 98 (i have xp). technologis move forword and someday youll think i can't balleve we thought we would stay with xp forever.