Slashdot Mirror

Tom's Hardware agreed to the terms of AMD's carefully-managed benchmarking sessions. Way to drink the Kool-Aid, Tom's. Anand stuck up for his own integrity as a reviewer and produced a much better review of the chip. Moral of the story: If you want a Phenom X4, wait for the B3 stepping!

Basically the reviews on Anandtech & Tom's Hardware have drawn some interesting conclusions... In terms of write performance, some are significantly worse than most notebook HDs, but all are better in terms of read performance. The idle of SATA SSD drives are significantly worse than UDMA ones (0.5w vs. 0.05w).

Basically, do your research... How much speed you'll get depends on how they bank the flash chips. More banks of lower density chips will yield a higher transfer rate--but uses more power. (Good luck finding how any one brand of SSD drive is banked...) Tom's Hardware found that the Samsung 64GB SSD offered double the transfer rate than their 32GB SSD. Anandtech found the Transcend & Super Talent SSD's to be extremely weak offerings. But then again Anandtech found the MTRON 32GB SSD far superior to most other drives they tested.

Basically SSD drives help with bootup times but in mixed tests, only the MTRON SSD drives are near Raptor speed, but I found only one retailer that even sells them--and a 32GB one for $2336.95 !!!

I hate to say this, I had a subscription to PC Mag from 1989-1993 then I learned UNIX and learned about real computers. That must have been the hey-day for them.

Does PC Mag matter anymore? I mean, isn't it just like Computer Shopper now? Full of ads for things I don't want or need with 1-2 articles of any use?

For example, they seem to have lost their way - current cover story is about smart phones? What, exactly, does a smart phone have to do with a PC? Of course, they have the gPhone link too.

Another article is GPS - what am I doing with a GPS on my desktop PC?

PC Mag has clearly lost there way. For me, Tom's Hardware http://www.tomshardware.com/ and Anandtech http://www.anandtech.com/ have taken their place.

Once the article actually loads I'd love to see what they came up with for Crysis for less than the price of my video card (of course I'm secretly hoping that it's running at 800x600 on low with a bad framerate to justify my ridiculous purchases!)

No, the difference is that the 8800 GTS is 90nm, and the 8800 GT is 65nm.

The Anantech review details things much, much better (unsurprisingly).

http://news.softpedia.com/news/Intel-039-s-X48-to-Come-in-Just-Another-5-Weeks-67604.shtml
http://www.dvhardware.net/article22289.html
It appears the X48 chipset is actually the X38 chipset without the ECC support and for DDR3 Only? Great, just when we weren't confused!

Here's another X38 review: http://www.anandtech.com/cpuchipsets/intel/showdoc.aspx?i=3120
A chipset comparison graphic: http://images.anandtech.com/reviews/chipsets/intel/x38-launch/memory-lg.png
And another review: http://www.tomshardware.com/2007/09/26/intel_x38_chipset/

http://news.softpedia.com/news/Intel-039-s-X48-to-Come-in-Just-Another-5-Weeks-67604.shtml
http://www.dvhardware.net/article22289.html
It appears the X48 chipset is actually the X38 chipset without the ECC support and for DDR3 Only? Great, just when we weren't confused!

Here's another X38 review: http://www.anandtech.com/cpuchipsets/intel/showdoc.aspx?i=3120
A chipset comparison graphic: http://images.anandtech.com/reviews/chipsets/intel/x38-launch/memory-lg.png
And another review: http://www.tomshardware.com/2007/09/26/intel_x38_chipset/

Anandtech has a nice write-up of Intel's Nehalem presentation at IDF: "Nehalem: Single die, 8-cores, 731M transistors, 16 threads, memory controller, graphics, amazing."

I agree that Intel's "two dies on one package" strategy worked great for the Core 2 architecture, but it seemed pretty lousy when they did this for the Netburst architecture (Pentium D). I guess the superior per-core performance of Core 2 overcomes the "messy hack" of putting two dies on one package.

Anandtech had a preview of Wolfdale including benchmarks back in August (here). The ironic thing is that with the limited availability of the K10 and its late arrival at most review sites, I've seen about as much real benchmarking of the unreleased Intel parts as I have of the supposedly widely-released AMD parts.

http://www.tech-hounds.com/article29/ArticlesPage1.html shows that there are niggling I/O performance issues. As well as issues with sound cards and network transfer (reminds me of all the linux issues in 2.4 before some of the fair scheduler changes in 2.6.x)

http://www.thetechlounge.com/article/411-8/ATI+Radeon+HD+2900+XT+Vista+Performance/ shows that game performance in Vista isn't any worse.

http://www.anandtech.com/showdoc.aspx?i=3060&p=1 shows high memory usage with recent fixes to ameliorate the problem somewhat.

More credible benchmarks from Anandtech: http://www.anandtech.com/IT/showdoc.aspx?i=3099&p=1 Harpertown is the clear leader in performance. Barcelona leads the performance/watt bencmarks.

WE DECIDED to ask Paul Otellini whether Intel would ever contemplate creating three cores on one die.
So he said: "We see a distinctive advantage in having all the cores on one die work."

WE DECIDED to ask Paul Otellini whether Intel would ever contemplate creating three cores on one die.
So he said: "We see a distinctive advantage in having all the cores on one die work."

WE DECIDED to ask Paul Otellini whether Intel would ever contemplate creating three cores on one die.
So he said: "We see a distinctive advantage in having all the cores on one die work."

WE DECIDED to ask Paul Otellini whether Intel would ever contemplate creating three cores on one die.
So he said: "We see a distinctive advantage in having all the cores on one die work."

"Throw more core and L2 cache at it. "

Supposedly, Intel's next generation architeture, Nehalem, will have much less L2 cache.

http://www.anandtech.com/cpuchipsets/intel/showdoc.aspx?i=3102&p=2

As default luser said, the difference would be marginal. I probably wouldn't bother upgrading your RAM unless you were also upgrading your CPU to a Core 2 Duo (which doesn't have any motherboards that use regular DDR, IIRC). Here's a benchmark. I have a E6600 which is stable and quiet when overclocked to 3.474 GHz with a Scythe Infinity...and 31 degrees C.

My video card is a Geforce 7900 GTX (512mb) and my CPU is an AMD Athlon X2 4600+. With that setup, would my RAM be holding me back?

No, not really. The difference in performance is typically %5 or less. Games are not usually memory-limited.

DDR3 at 1333 MHz is essentially the same performance as DDR2 800. It just has twice the bandwidth(which nothing takes advantage of at the moment).

now, some real magic begins to happen when you step up to the 1800 MHz DDR3 that is already available based on Microns Z9 chips. OCZ already has some real doosies out, check out this review:

http://www.anandtech.com/memory/showdoc.aspx?i=3053

or heres the comparison numbers:

http://www.anandtech.com/memory/showdoc.aspx?i=3053&p=4

DDR2 held it's own until the buss speeds were cranked up. the new 1800MHz DDR3 went all the way up to 2000MHz achieving a 38.28sec superpi score compared to the 45.20 sec best that DDR2 could throw out.

yeah, yeah, i can hear all you snots saying that it doesn't have anything to do with the price of rice in china. Think about it this way, if it can shave almost ten seconds off of a 5 minute long calculation, imagine what it can do to compile times. might have to cut your coffee break a little shorter!!

most times progress comes in little drops instead of floods.

DDR3 at 1333 MHz is essentially the same performance as DDR2 800. It just has twice the bandwidth(which nothing takes advantage of at the moment).

now, some real magic begins to happen when you step up to the 1800 MHz DDR3 that is already available based on Microns Z9 chips. OCZ already has some real doosies out, check out this review:

http://www.anandtech.com/memory/showdoc.aspx?i=3053

or heres the comparison numbers:

http://www.anandtech.com/memory/showdoc.aspx?i=3053&p=4

DDR2 held it's own until the buss speeds were cranked up. the new 1800MHz DDR3 went all the way up to 2000MHz achieving a 38.28sec superpi score compared to the 45.20 sec best that DDR2 could throw out.

yeah, yeah, i can hear all you snots saying that it doesn't have anything to do with the price of rice in china. Think about it this way, if it can shave almost ten seconds off of a 5 minute long calculation, imagine what it can do to compile times. might have to cut your coffee break a little shorter!!

most times progress comes in little drops instead of floods.

specfp rate was running faster on pre-barcelona dual core Opterons than on Intel's dual core Woodcrest. The reason is no big secret: specfp is memory bandwidth limited and specfp_rate is specfp's running in parallel. Here is a good anandtech article on the subject.

We already know that AMD has superior memory performance. If you are doing bandwidth-limited floating point, Barcelona is the clear winner.

If you're making a general statement about floating point performance, you're wrong.

Here's some benchmarking done by Anandtech.

And a performance preview for Barcelona desktop as well.

Here's some benchmarking done by Anandtech.

And a performance preview for Barcelona desktop as well.

But it's far from clear that this is the product that will help right AMD's ship."

Lol. Enjoy your overpriced, bloated, largely closed, incredibly slow excuse for an OS. The only things OSX has going for it are that it looks pretty and is designed for use by the mentally challenged, just like Vista. Being an OSX zealot does not make you cool or indie.

"Get yourself a real OS" indeed.

Lol. Enjoy your overpriced, bloated, largely closed, incredibly slow excuse for an OS. The only things OSX has going for it are that it looks pretty and is designed for use by the mentally challenged, just like Vista. Being an OSX zealot does not make you cool or indie.

"Get yourself a real OS" indeed.

Oh.. and after more searching... try this AnandTech article.. http://www.anandtech.com/video/showdoc.aspx?i=2886 &p=4

I would just like Linux support for the Theater 550 series of chips. For the money, they are definitely the best cheap non-HD tuner out there. They trounce any cheapo Hauppauge card out there (sorry, can't deal with the extreme blurriness), and it's the only thing barring me from trying MythTV.

http://www.anandtech.com/video/showdoc.aspx?i=2393

-S

If you read the fine print, AMD is actually not implementing all of SSE4 on the Bulldozer chip which will be the first to include SSE5. This is disastrous - the SSE "brand" has always implied backwards compatibility: SSE1 contains MMX, SSE2 contains SSE1 & MMX, etc. etc. Now AMD is breaking this, since SSE5 chips will not include all of SSE4. AMD shouldn't have named these new extensions SSE5. As it is, they are forking the x86 instruction set, which is a bad thing for all of us.

Here's some more information: http://www.anandtech.com/cpuchipsets/showdoc.aspx? i=3073

Buffer cache does not show up in usage. What is showing up is prefetch, which apparently some drooling morons are incapable of turning off. Vista does in fact use too much memory -- Microsoft recently released a patch that dramatically reduces memory usage for many apps, and almost brings it down to XP's usage -- but this isn't anything I expect intelligent discussion about on slashdot. You have to go elsewhere for journalism.

> You're kidding, right? My last Seagate ('Cuda.10 - 320GB) was made in China.

NOOOOO!!!!!!!!

Seriously, bummer. The good news is The Samsung Corporation have entered the HDD market to good reviews. I have two of their drives, and yes, they're made in Korea: http://www.anandtech.com/storage/showdoc.aspx?i=30 31

Noticed since HP switched PC manufacture to China they've been getting shoddier quality too. Bad News, so I reach for my security blanket: http://news.top100.biz/shopping/Made-in-China-blan kets-withdrawn-in-New-Zealand-Australia

I get a chuckle out of people who pay to beta test products.

The truth is that there are a host of issues.

The best experiences I have heard from anyone is that it is almost as good as XP.

Reviews that tested gaming performance, show Vista slower across the board.
http://www.hardocp.com/article.html?art=MTMzNCw2LC xoZW50aHVzaWFzdA==

Testing also revealed that Vista had changed it's driver model to virtualizing Vram into limited user address space, leading to minor things like crashing out in the midst of heated gaming sessions.
http://www.anandtech.com/cpuchipsets/showdoc.aspx? i=3044

Vista brings networking to a crawl when playing audio:
http://forums.2cpu.com/showthread.php?t=83112

So we have slower gaming, crashing from memory space exhaustion, and networking reduced to a crawl if you play an audio file, and this is just recently and major headlines.

Basically you are very lucky if it works anywhere near as well as XP. For degraded performance/lower reliability you get what? Aero?

I realize that in a few years most of use will be using Vista, but I never touch a MS OS without at least 1 service pack. Vista is MS most unnecessary upgrade since Millennium.

Anandtech recently reviewed a 32GB SSD drive from Mtron, and compared it with a Raptor HDD.
http://www.anandtech.com/storage/showdoc.aspx?i=30 64&p=6

TLDR: *big* improvement in game loading times, merely ok performance in file copying, encoding, compression. It's also silent, and saves a lot of power. $1500 though. . . should get much cheaper in the next few years.

Today only at powerpoint city.. 2 slides for the price of one! With a neat blending effect and a free press release!

    OK I kid, but AMD going to DDR3 in 2009 (assuming the date doesn't slip) is not exactly the biggest revolution in all of computing. I'm sure it will be a very nice memory interface, but by then the standard 'Intel is evil because of the FSB' rant will be obsolete due to CSI (now called 'Quickpath' in marketese).

      What I want instead of an AMD press release: Honest to God numbers on the K10! The thing is supposed to be 'officially' launched in 3 weeks and there is nothing from any independent source! Meanwhile over at Anandtech there is a full preview of an actual Wolfdale chip that was run, tested, overclocked, all without Intel's supervision (or apparently their permission either). We now have a whole hell of a lot more information on an Intel chip that is not due out until very late this year or next year than we do about the amazing new AMD chip that is supposed to already be at system integrators by this point. That is not a good sign for AMD.
        Oh, and I do want AMD to do well, they have some nice technology, I'm just not necessarily enamored with their more vocal fanboys. Despite what some people here think, AMD is also a multi-billion dollar company that outsources production overseas, tries to charge more for its products than it had to pay to make them, has those evil patents, supports DRM technology, and has agreements with that evil Microsoft company. Despite the propoganda they are not pure angels and Intel is not pure evil either.

Have a look at This article.
With 4GB machines being more and more common (my laptop has 4GB - and it's not a workstation class one either, just a medium high end consumer one), the need for a 64 bit addressing is no longer something in the distant future.
32 bit addressing - and particularly the 2G userland / kernel split in Windows is already causing problems for games right now.

Sun and Adobe are both dragging the chain on 64 bit browser plugins. Although, at least Adobe say they're working on it, but Sun's reluctance makes little to no sense, they've had a 64 bit JVM and JDK for years, for both Windows and Linux. If they can manage to port the hundreds of thousands (millions?) of lines of code in the JVM to AMD64, why can't they manage the plugin? Surely that's only a handful of lines? A couple of thousand at most?

Now look beyond the advertising figures and look at the actual real world application figures.
Intel/AMD beats the pants off UltraSPARC T1
http://www.anandtech.com/IT/showdoc.aspx?i=2772&p= 6

While I'm sure your 20k T1 outperformed your 100k v880. It does not show that the T1 is a better choice than an Intel/AMD system.

Sun's stuff are slower than IBM's POWER line, and they are nowhere in the same league as IBM mainframes, and IBM mainframes are not in the same league as real nonstop computing clusters.

Mainframes = very good uptimes, but you have _scheduled_ downtimes.
Stuff like OpenVMS or Tandem = uptimes of _decades_ possible, don't even need scheduled downtimes where you turn everything off, you can run while replacing the hardware. With the Tandem stuff you even have CPUs running the same thing at the same time for real redundancy. Only thing is HP seems to be burying VMS and Tandem.

Sun? They didn't even have hardware instruction retry till Fujitsu SPARC. For many years it was pretty embarassing that the really high end SPARCs were Fujitsu rather than Sun - the fastest SPARC systems till just a few years ago were all Fujitsu PRIMEPOWER (I haven't bothered checking recently, the last I recall Sun started using Fujitsu stuff for their high end systems).

Sun got where they were by making relatively cheap Unix RISC workstations and they provided servers for areas where reliability and availability didn't really matter as much as the real "high end" stuff. They caught the internet wave for "cheap" webservers etc and made a lot of money then.

The problem now with Sun is, they get blasted at the low end by x86, and at the high end they pale in comparison to IBM's stuff.

For "normal" webserver/db/internet/corporate stuff it's x86.
In the HPC arena it's x86 (for scale out), and IBM (for scale up).
So where does Sun fit?

Just go google for benchmarks of T1 vs Intel vs AMD. The T1 doesn't even do that well for performance/power consumption when compared to the Intel woodcrest CPU: http://www.anandtech.com/printarticle.aspx?i=2772

For Sun's sake, their Niagara 2 better be magnitudes much better than their T1, if not it'll be out of date even before it's released.

Don't get me wrong, I'll be happy if Sun succeeds, but they've fallen way behind.

How about 4GB for less than $100? Rebates... but I've gotten for less than $150 without rebates, just wait for sales at Fry's (online or store).

http://forums.anandtech.com/messageview.aspx?catid =40&threadid=2054141&enterthread=y

Back to original topic... Intel motherboards really are held back. Why can't I put in 16GB for about 2.5x the price of 8GB?

It seems the main selling point for Vista is DX10 exclusivity, which is aimed at gamers.

But Vista is slower than XP for games and now it appears that Vista has a second problem, it memory maps the entire Video memory into user address space, wasting this precious resource:

http://www.anandtech.com/cpuchipsets/showdoc.aspx? i=3044&p=1

Also this is not done for performance reasons, but seems to be part of Microsofts efforts to tighten the DRM screws.

So vista gives gamers DX10 which is currently pointless, slows performance and steals address space leading to more crashes. Yay Vista.

Yuk.

Anandtech recently did that kind of power efficiency vs. performance test actually: (2.6Ghz vs. 2.33ghz), with AMD coming out on top: http://www.anandtech.com/IT/showdoc.aspx?i=3039

I spent three years as a senior editor at InfoWorld, and I certainly have a lot of criticisms to offer about the tech trade media industry. But I can say, with absolute certainty, that when trade media outlets like InfoWorld disappear you will all be sorry.

It depends. Tech trade media outlets tend to provide three basic services:

1. News related to products and the industry.
2. Product reviews.
3. Editorial content.

Yeah, you heard me right. Is the media industry going to shit? Corporate media is on the blame list, for sure. But first on the list is you. Have you ever written your Congressman? Probably not. But even if you have, it's probably futile to ask that you write to your favorite media outlets and ask -- even beg -- them to cover real news, and not just fluff pieces and fake stories.

I spent 15 years writing letters and comments via both paper letters and the provided online forums (I was fairly active in places like InfoWorld Electric before Sandy trashed the place, on Extreme Tech before management replaced the good forum software with a piece of crap, etc.), trying to let publications know as a Mac and OS/2 user (and more recently as an OS/2 and Linux user) that I did not appreciate the Windows-centric nature of most of their articles, and that I did not like their tendency to review inferior software from a few select large vendors instead of also including smaller (but often more capable) solutions from the freeware and shareware worlds.

I got nowhere, and I eventually got sick of it, so I now have ZERO subscriptions to printed tech media. Zero.

Technology media went to shit in the early 1990's. To hell with them.

my main question before that they answered is if the core 2 quad processors are being choked due to insufficient bandwidth. they measured the difference between 1066 and 1333fsb and performance barely increased. this brings me back to the following observations:

1. the processor is not bandwidth limited and merging the 4 cores in a single die would not yield much performance benefit. this brings back to the argument who is better the native quad core vs 2 x dual core (though in an engineering standpoint, the native quad core will be better but in real world as long as performance is good, i really don't care.)
2. intel has made a good job with their branch prediction that required data is almost always in the cache reducing any performance hit via fsb.
3. on-die memory controller might not yield great performance improvement for current core2 processors (if intel would add them now.)

the main anticipation now is if amd will beat intel in their upcoming barcelona chips. the advantage right now with amd is their ability to increase sockets through hypertransport. but that advantage might disappear soon when intel integrates the memory controller.

anyway, as of now the consumers are winning. i hope that amd will be able to keep up with intel so we can have good competition. for the mean time, i will be buying a q6600 system. quite cheap cpu coupled with cheap memory now (good time to increase memory capacity to >=4gb.)

Anandtech has a pretty good article about these releases and also about the price cuts. This is looking great for me when I build a new computer in a few months (on which I'm planning to spend $150 chip from two years ago look pathetic. Oh well.

Of course, I'll need to figure out AMD vs. Intel. I just wish Intel had a better bus design. AMD has a good bus (HT) and Intel has the best chips right now. Maybe if they merged...

http://forums.anandtech.com/messageview.aspx?catid =34&threadid=2066483

Or be in the same room! I love how they skip over the battary exploding. Guess it isn't good for advertising having your blender explode.

This and the fact he doesn't cover his mouth when opening the blender makes me think the whole thing is fake. Faking it is a lot cheaper than buying all those expensive goods to blend and a lot safer. It is not like anyone would be able to tell the difference in between the real thing and faking it. If watch slowly you would notice the bounce of the iPhone during the full speed and the slow-mo are different (so maybe they killed more than one iPhone!). If you look closely at 0:35 the internal you see is this brown-ish circlular thing. Nothing like the internals of the iPhone! Maybe they are using fake iPhones? I see nothing simillar in the iPhone (if anyone does please reply!). The only thing which is brown is this wiring unlike the Zune.

I don't get why his ebay account has a purchase of a 4GB model on the 5th for $600 but it does look like either he is using fake iPhones or faked the whole thing,

I say Blendtec are genuis as business and marketing. I wouldn't mind to have been working at such a place during my teen years. It would of given me quiet a bit of experience.

-- OU. I would of logged in but my password is random and I am on a shared computer.

Or be in the same room! I love how they skip over the battary exploding. Guess it isn't good for advertising having your blender explode.

This and the fact he doesn't cover his mouth when opening the blender makes me think the whole thing is fake. Faking it is a lot cheaper than buying all those expensive goods to blend and a lot safer. It is not like anyone would be able to tell the difference in between the real thing and faking it. If watch slowly you would notice the bounce of the iPhone during the full speed and the slow-mo are different (so maybe they killed more than one iPhone!). If you look closely at 0:35 the internal you see is this brown-ish circlular thing. Nothing like the internals of the iPhone! Maybe they are using fake iPhones? I see nothing simillar in the iPhone (if anyone does please reply!). The only thing which is brown is this wiring unlike the Zune.

I don't get why his ebay account has a purchase of a 4GB model on the 5th for $600 but it does look like either he is using fake iPhones or faked the whole thing,

I say Blendtec are genuis as business and marketing. I wouldn't mind to have been working at such a place during my teen years. It would of given me quiet a bit of experience.

-- OU. I would of logged in but my password is random and I am on a shared computer.

Read it and weep.

IBM'er says Vista's RAM sweet spot is 4GB
http://www.computerworld.com/action/article.do?com mand=viewArticleBasic&articleId=9011523

Anandtech:

http://www.anandtech.com/systems/showdoc.aspx?i=29 17&p=4

"While it's very difficult to benchmark the impact of SuperFetch well, in our usage of Vista if you have enough memory it is a tremendous ally. Honestly SuperFetch is the biggest reason, in our opinion, to move to the x64 version of Vista so you can use even more memory. Although we found that 2GB of memory is still quite passable under Vista, the new sweet spot if you happen to multitask a lot is 4GB - in no small part due to how well SuperFetch utilizes the additional memory. Do keep in mind that you'll need to make sure your motherboard has proper BIOS support for 4GB and also make sure Vista x64 has driver support for all of your peripherals before committing to the move....

How much RAM do you really need for Windows Vista? We recommend a bare minimum of 1GB of memory for all Vista users, 2GB if you're a power user but don't have a lot running at the same time, and 4GB if you hate the sound of swapping to disk...."

Tom's Hardware:
http://www.tomshardware.com/2007/01/29/xp-vs-vista /page11.html

"Conclusion: K.O. For Windows Vista?

Windows Vista clearly is not a great new performer when it comes to executing single applications at maximum speed. Although we only looked at the 32-bit version of Windows Vista Enterprise, we do not expect the 64-bit edition to be faster (at least not with 32-bit applications).

Overall, applications performed as expected, or executed slightly slower than under Windows XP. The synthetic benchmarks such as Everest, PCMark05 or Sandra 2007 show that differences are non-existent on a component level. We also found some programs that refused to work, and others that seem to cause problems at first but eventually ran properly. In any case, we recommend watching for Vista-related software upgrades from your software vendors.

There are some programs that showed deeply disappointing performance. Unreal Tournament 2004 and the professional graphics benchmarking suite SPECviewperf 9.03 suffered heavily from the lack of support for the OpenGL graphics library under Windows Vista. This is something we expected, and we clearly advise against replacing Windows XP with Windows Vista if you need to run professional graphics applications. Both ATI and Nvidia will offer OpenGL support in upcoming driver releases, but it remains to be seen if and how other graphics vendors or Microsoft may offer it.

We are disappointed that CPU-intensive applications such as video transcoding with XviD (DVD to XviD MPEG4) or the MainConcept H.264 Encoder performed 18% to nearly 24% slower in our standard benchmark scenarios. Both benchmarks finished much quicker under Windows XP. There aren't newer versions available, and we don't see immediate solutions to this issue.

There is good news as well: we did not find evidence that Windows Vista's new and fancy AeroGlass interface consumes more energy than Windows XP's 2D desktop. Although our measurements indicate a 1 W increase in power draw at the plug, this is too little of a difference to draw any conclusions. Obviously, the requirements for displaying all elements in 3D, rotating and moving them aren't enough to heat up graphics processors. This might also be a result of Windows Vista's more advanced implementation of ACPI 2.0 (and parts of 3.0), which allows the control of power of system components separately.

Our hopes that Vista might be able to speed up applications are gone. First tests with 64-bit editions result in numbers similar to our 32