Slashdot Mirror

O'Reilly News recently interviewed Arjan van de Ven about his efforts to improve Linux performance and reduce power consumption. Arjan works for Intel in the Open Source Technology Center. This interview is approximately 30 minutes.

One of the projects you're probably most known for in the past couple of years is the PowerTOP utility, which I found very fascinating. Looking at some of the gains you've made over the past 18 months, it seems like Linux-based devices are saving a lot more power than they used to. What do you consider the big successes in the past year and a half?

To be honest we fixed effectively the entire Linux desktop space. It's not--PowerTOP is more--it's not just what we fixed with PowerTOP is not individual pieces. We fixed everything. For me that was a success.

Is that everything in terms of not just desktop but servers as well?

Yeah; we fixed not just Evolution. We fixed Firefox; the thing with Firefox was that it wasn't one thing that was broken. Everything had problems and we had to fix all of it. So for me the success was how quickly everything got fixed; it was just amazing.

In this context you consider fixed--everything is no longer broken in the same way or--?

Everything is no longer keeping the CPU out of idle basically.

Do you have a reference machine? I guess I'm asking what's your benchmark for this, a particular software configuration stack or particular type of machine, or are you willing to say it's pretty much every Linux based machine out there?

I'm looking at several machines--my own laptop but to be honest, what runs on my own laptop is what I care about most. At least that's where I got more battery life, this is where I see the changes. I tend to run a quite rich environment on my laptop but I also look at service. We look at all kinds of machines and we see the same trend everywhere in that all the various pieces of it--never polling or keeping the CPU up. They all got fixed.

In fixing this, is there a component of education, for example, saying "Instead of doing a busy wait on a select loop or continually polling you should set a kernel timer and wait for that to call you"?

That's part of it but the biggest thing is that you had no visibility. Just two days ago at IDF I spoke with a developer of the GNOME desktop and he said, yeah; when I saw it happen I fixed it in 10 minutes, but you don't know it's there until you see it from PowerTOP. Adding the visibility turns out to be enough for people to start fixing it. They know how to fix--how to not poll most of the time.

You can't fix something you can't measure.

If you don't see that it happens you don't know it happens and you can't fix it.

Are you getting the same sort of results from other projects you run into?

GNOME was there but it's almost everybody goes oh yeah; we should have not done that; either they fix it themselves or some--a lot of people give them the fix and in general it's like oh yeah; we shouldn't have done that. Unless you see what's happening you don't know what to fix, so the biggest thing that PowerTOP did was add visibility. We can see under the hood what's going on and then we can fix it. And quite often the fix is very simple.

It sounds then, maybe I should be able to say that just about everybody is happy to see this. Is that the case?

Yes; people--all the developers I've worked with--and that's quite a few--they all go oh yeah. Thank you for the fix; we should have no problems in the first place. We didn't know this; it's fixed now. In the beginning I did most of the fixing when PowerTOP was very new and now days the people do it themselves. The developers learn

I have a tilt as well, and can definitely vouch for this. Despite being a -huge- supporter of FLOSS, I have (and enjoy) my tilt. I really wanted to find a Linux-based smartphone within my price range, with the features that I wanted. The OpenMoko was too costly.

It came down to these things:
* I needed (actually, really wanted) a touch screen and wi-fi. This phone does both.
* I wanted to ability to install/develop apps for my phone without any policing. This scrapped the iPhone in my plans (Yes, I know about jailbreaking. I have several reasons about why I'm morally opposed to this, which I won't get into).
* I wanted expandable storage. The tilt goes up to 32GB, and it's removable. Try doing that with an iPhone.

The integrated GPS, slide-out keyboard, GPRS/EDGE, 3 megapixel camera, and charge-via-USB were just additional perks, but quite appreciated.

Linux and other OSS-based phones will eventually start making a dent in the US market, which will bring the cost down. I'm not crazy or proud about usinga Windows Mobile-based phone. It's not Free, but parts of it can be deemed as "open".

Long story short, the tilt is a very robust phone. It does have some very important shortcomings; mostly due to HTC's reluctance to release any of their specs, and the fact that only runs Windows Mobile. That being said, the fact that you can do all of the things mentioned above with Windows Mobile, along with being able to develop and distribute your own apps for it (look up PuTTY and OpenVPN. There are Windows Mobile versions for both) is pretty cool. Intel even a released a BSD-Licensed SDK for mobile apps (MPSDK).

Oh yeah, I forgot to mention that my laptop runs Ubuntu. The tilt integrates quite well with it, when using USB or wi-fi. Bluetooth doesn't quite work yet, but that's due to the unfinished subsystem packaged along with Ubuntu 8.04.

Yes, I have and the free cooling experiment / paper is mine. Your point is well taken and aligns with why I did this experiment. There are a lot of reasons people run DCâ(TM)s very cold but most of them can run much warmer if you control the air⦠people over cool to deal with the hot spots. If you have questions about the experiment, blog on my site at http://communities.intel.com/openport/blogs/datacenter/2008/09/19/free-cooling-for-data-centers-video-and-whitepaper

The increase failure rates were almost 95% hard drives. Iâ(TM)m the author of the paper and if you have more questions, let me know. I also started a blog at: http://communities.intel.com/openport/blogs/datacenter/2008/09/19/free-cooling-for-data-centers-video-and-whitepaper

What a joke of an article. Every semiconductor manufacturer has several generations of process in various states in the lab. Woo IBM's showing sneak peaks at 22nm!

I met with an Intel VP for an interview a while back and talked about where things are going. He had some nice lab-pr0n of what the photos claimed were 11nm transistors. I believe it was said that was "about 15 years out", and meant to offer reassurance that Moore's Law still had a bit more time left to go.

Actually here, let me go dig up my transcript so I can get a proper quote:

You're going to see that platforms are going to continue to evolve. We're moving to a faster cadence. The processor cadence is about a two year cadence, in terms of process technologies. By the way this is interesting. We know how to do Moore's Law for about another fifteen years which we've never had that kind of length of projection before. ...it sort of takes 3D transistors and all that, but we know how to do these things. It's all using standard silicon, it's CMOS it's extraordinarily well charictarized right? But we've got transistors running at 11 nanometers, I can show you photographs of them. We have the leakage issues but we've got a very good plan.

That was 2 years ago, early October 2006. Who leapfrogged what now?

is a very interesting read

I admire your gusto - nobody should take away from this thread "ideas are worthless". Keep em coming. Unfortunately I will now tell you why your idea won't work. If you were a practising programmer, you would know this stuff already.

For example, why don't we have a root/user distinction on email? you could set it up so the user account could read the mail but not reply or delete it and the root account had full "regular" control - then if you wanted to view mail using an unsecured computer that would be fine; even if someone did steal your password they could at best be an annoyance to you (so long as you don't have loads of passwords stored there). It would make it so much easier to check email whilst you were staying with family who think an unsecured copy of XP is "good enough".

This won't work because:

• If you can read somebody elses email, you can reply to it convincingly. Remember that authentication in SMTP is very weak - if the users provider has set up SPF/DomainKeys then a fraudulently sent mail might get a phishing warning or spam filtered by the receivers software. Or it might not. But you can't rely on that. Insecured read-only access to email would still let anyone send mail that appears to come from you, but now including details of all the previous conversations you had with them. So the only thing it'd protect against is "delete".
• Most websites assume your email account is secure and will mail password reset links to it. You only need to have read-only access to use them.
• I guess a lot of people who use email care about replying to it - any conversation that involves humans usually requires it :) So even if you could enforce your no-reply rule, it'd dramatically cut the usefulness of the system.

There are better ways to achieve what you want (secure email checking for your parents house). For instance you could just use a mobile phone instead of a computer. Or you could take a laptop you trust. Or if you want a technological solution, you could build a solution on top of trusted computing. The hardware for this is only starting to ship now, so it's a long way from being in your parents place, but in theory it allows you to go from a system in an arbitrary state (rootkitted, ridden with malware etc) to running provably secure software. The technology is very complicated but it'd provide what you want, without needing to compromise on emails features or making people think they are secure when they are not.

So Intel marketing would like people to forget their SSDthat is in the Acer Aspire One>? Which is notorious among the Aspire community for being a dog.

There is an Intel Google gadget for detailing current laptop watt usage. There probably is a standalone application that does the same thing from Intel.

Whiy not ? That might be their response to China's threat :)

http://hardware.slashdot.org/article.pl?sid=08/09/03/1437254

Quite possibly. Intel have put up a page on being lead free too

http://www.intel.com/technology/silicon/leadfree.htm

Do those laws apply if say the microprocessors are fabbed in Germany/Israel/Ireland, assembled/packaged in Malaysia, and then exported to China?

How about if the microprocessors are assembled/packaged in China itself?

http://www.intel.com/pressroom/kits/manufacturing/manufacturing_qa.htm

http://www.amd.com/us-en/Corporate/VirtualPressRoom/0,,51__104_543~117787,00.html

Then this, sir, is the processor for you and I both. http://www.intel.com/pressroom/archive/releases/20080325comp.htm Each of the 50watt L54x0 cpus comes in 2 retail flavors, here's what the model numbers look like:
BX80574L5420A
BX80574L5420P
The only difference is the size of the included heat sink based on the intended case size, A for 1U and P for 2U and bigger. Due to the extra room for a larger heatsink these 45nm, 12MB cache, 2.33-2.5Ghz Quad Core monsters can be passively cooled! The shit part is having to shell out $600 for a compatible motherboard. IMO, $365-425 is pretty reasonable for the CPU. I don't think we're going to have to wait 10 years for it. The mfg's are starting to realize tons of heat and energy usage just for a few more fps or opening Word .1 seconds quicker is not what the market wants or needs anymore.

Yeah my OEM Intel motherboard uses EFI too:

http://www.intel.com/design/motherbd/bx2/bx2_industryspecs.htm if you're curious.

Intel is aiming at number crunchers (note that their chip uses doubles, not floats).

That's not true. From their paper:

Larrabee gains its computational density from the 16-wide vector processing unit (VPU), which executes integer, single-precision float, and double-precision float instructions.

And it's definitely aimed largely at games: the paper gives performance studies of DirectX 9 rendering from Half Life 2, FEAR and Gears of War.

Lol, right. I mean, companies would just love spending billions of dollars to develop something and then having a rival just pick it up and copy it. That would _totally_ spur innovation, dude!

Yeah, you're right. The concept of open sharing sure hasn't spurred innovation in the past...

The Atom being used for this is a horrible thing. It's designed as a low-power processor but so much so that under real use it uses considerably more power with less capability than an athlon 64.

That's not the CPU, it's the chipset.

The board in question uses the 945GC chipset, while all mobile devices use the Mobile 945GM or better. The Mobile 945GM features a much cooler 7w TDP (see section 11.2), while the 945GC is 22w TDP. You can even get as low as 5w TDP if you pay extra.

Why the difference? The 945GC is the rejects pile. If they can't get the chip to run an the lower voltages specced for the 945GM (about 1.05v), then they attempt to bin it at the much higher clock and voltage of the 945GC.

The Intel Atom board with the 945GC is intended for ultra-cheap PCs (sub $150), and is mostly marketed in third-world countries. the board itself is NOT designed for low power, although there's nothing stopping a third party from making such a beast.

Linux is known to be more power-hungry than Windows; I noticed the same on my computers.

Windows XP works about 40min longer than openSuse11 on the same machine, using default settings.

Here is some reading material:
- http://www.lesswatts.org/projects/powertop/
- there was a white paper written by folk from Intel, I don't remember where I found it, but it could be somewhere here: http://oss.intel.com/en-us/casestudies/

You need to switch to a tickless kernel, and tinker with powertop - that should improve things.

Note that in my case, none of the powertop tricks had any impact - I was surprised to see that no matter what I did, the estimated time would always be 1h45min. This is still an experiment in progress, so don't count this feedback as 100% certain.

PC World has a decent summary of Intel Remote Wake Technology.
http://www.pcworld.com/article/149863/2008/08/.html

Then there's also the actual Intel site
http://www.intel.com/technology/chipset/remotewake.htm

We've had a lot of problems with Intel graphics software. You are correct, however, we haven't tested the latest offerings from Intel. We felt so abused by the previous chipsets that we have had no desire to test the new software.

The last video driver we tested was version 14311 for the 945 chipset. It had a LOT of problems. There was a LOT of denial by Intel that there were problems.

So, I would be very interested to know: Is the video in the 965 chipset better? Is the software trouble-free? How about rotated vertically on a 1920 x 1200 monitor?

I think this part of the computing timeline is going to be
one that is well remembered. I know I find it fascinating.

This is a classic moment when tech takes the branch that
was unexpected. GPGPU computing will soon
reach ubiquity but for right now it's the fledgling that is being
grown in the wild.

Of course I'm not earmarking this one particular project
as the start point but this year has gotten 'GPU this' and
'GPGPU that' start up events all over it. Some even said
in 2007, that it would be a buzzword in 08.

And of course there's nothing like new tech to bring out
a naysayer.

Folding@home released their second generation
GPU client in April 08. While retiring the GPU1 core in
June of this year.

I know I enjoy throwing spare GPU cycles to a distributed
cause and whenever I catch sight of the icon for the GPU
client it brings the back the nostalgia of distributed clients
of the past. [Near the bottom].

I think I was with United Devices the longest.
And the Grid.

Now we are getting a chance to see GPU supercomputing
installations from IBM and this one from MIT.
Soon those will be littering the Top 500 list.

I also look forward most to the peaceful endeavors the new
processing power will be used for... weather analysis,
drug creation, and disease studies.

Oh yes, I realize places like the infamous Sandia will be using
the GPU to rev up atom splitting. But maybe if they keep their
bombs IN the GPU it'll lessen the chances of seeing rampant
proliferation again.

Ok, well enough of my musings over a GPU.

-AI

I say it every year and this is as good a place to say it as any.

The easiest way to get the watts off your processor and out of your PC is...

not to put them in. Duh. Fortunately, somebody is listening. Finally.

You said,

"Apart from this, numerous HP employees have been discussing this subject within HP. People like myself and other individuals from the Software branch have pointed out this is a wasteful approach. And judging by the brand spanking new e-delivery option and certain other efforts within HP, I see that this is actually worked on for SoftWare."

And:

"Probably the costs that are associated with a radical change of this system are quite high, because it's likely that many changes need to be made in databases, order systems, processes and procedures."

And:

"As said, I have seen indications this is being worked on, but one has to remember we are a company the size of a small country, and that makes it a little more difficult to maneuver than a one man company."

Wow! Isn't there some mid-level manager at HP who is willing to say, "Stop sending single pieces of paper in boxes! Never do that again." ???

You made me realize the weakness of my grandparent comment. I thought I was giving an accurate picture of the misery inside HP. But I forgot to mention the most scary part of being a corporate drone: The drones don't realize they are drones. Woooo-oooo-ooo.

The situation is even worse than that! (Similar to the late-night Infomercials: But wait!! There's more!) At least, when it is only excess packaging, it is possible to just put the cardboard in the recycle bin. The real misery happens when drones become involved with technical details. I remember a conversation with an HP representative about a model of HP laser printer that costs about $1000. He told me, if I remember correctly, that it was entirely reasonable that if that model printer needed routine maintenance, the work could not be done locally, in our big city.

Nothing about this should give the impression that I think corporate drones are a particularly bad problem at HP. I have had worse experiences with Microsoft and Intel representatives.

(But wait!! There's more!) It's even scarier when entire departments become drones. I was on an Intel marketing email list. I got many emails suggesting that I would be motivated to buy Intel processors by the fact that Intel would give me a free bunny suit doll.

Hah! Are there people who don't believe that Intel was using dolls to sell processors? Believe! It's not my photo; I just found it with a Google search. I would never jump through the hoops necessary to get an Intel Bunny Suit Doll.

How did the department at Intel scarily called "Marketing" first arrive at the idea that making customers jump through hoops is doing something good for the company?

Then, later, the entire idea that "Intel Marketing" should do something good for Intel was completely abandoned. That happened years ago, so long ago that no one who is there now can remember when Intel Marketing was good for the company, or even cared about being good for the company.

Want a recent example? The new Intel 45 nanometer processors, which are an extremely impressive engineering achievement, I think, are called Centrino 2. Before they were "Core 2 Duo". Believe! (More Infomercial talk: That's not 1! Not 2!! But 3 uses of the concept two!!! The second person in the infomercial says, "John, that's Amazing!!!!")

You think that monster attacking Sigourney Weaver in the movie Alien was scary? "Intel Marketing" is even scarier than that. At least Sigourney Weaver realized she was being attacked by a monster. The really scary thing is when someone has become the monster, and doesn't realize it.

Maybe Intel top management thinks that Intel Marketing is like prostate cancer. Sometimes, if a tumor isn't growing, it is considered better to let it stay in the body.

The one on that list that surprises me is Intel. They make very little software

http://www.intel.com/cd/software/products/asmo-na/eng/index.htm?iid=siteindex+prod_software

They produce C++ and Fortran compilers, debuggers, performance analysers, "cluster tools".

I'm not sure about their market share, but that's a non negligible amount of software IMHO.

Moore himself (you're probably aware of this, but I'm sure many people are not, and I just learned a lot of this today) couched his observations in terms of cost. He was extremely prescient! A presentation by Moore himself has some plots showing several factors (lithography cost, shrinking sizes, etc.) vs. time. http://download.intel.com/research/silicon/Gordon_Moore_ISSCC_021003.pdf

Roughly parabolic on a log-log scale.

    http://www.intel.com/technology/mooreslaw/pix/originalgraph.gif

CPU != GPU

He was probably talking about the Intel GPU.

That way, there is no incentive for the vendor to solve the problem since they don't even know about it.

Except that the problems are documented here: developer.intel.com. Intel knows about them. Whether they know about the security implications is another matter.

Then get to coding. Intel is based in the US so they HAVE TO obey US law.

Intel was founded in the US, but has a legal presence in over 50 countries. Saying they have to obey US law is meaningless -- they have to obey the laws of other countries as well.
 

They've been making features smaller than the wavelength of light for almost a decade now using interference lithography, it's nothing new, but it did scare people that Moore's Law would be over sooner than they thought (after all, how would they write features using extreme UV with the materials that have?).

What's interesting is that their interference lithography mask allows them to reach a minimum feature size limit of 25nm for silicon.

Hurray, Moore's Law should continue for another 10 or so years :) (At least, in terms of using silicon... But we're already using other materials (specifically hafnium) to keep going.)

That's a manual for Intel 64, their version of x64 (aka IA-32e or EM64T) which they rushed out in response to AMD64.

You can see they also have manuals for Itanium (aka IA-64), which is vastly different. Early Itaniums actually had some IA-32 emulation hardware, but it was never good enough to run an IA-32 operating system, and it was so slow they dropped it in favor of their "IA-32 EL" software emulator everyone was using.

Having worked in the Windows group at Microsoft, I assure you that we had to build separate x64 and IA-64 binaries for everything. They definitely aren't compatible at all, and as far as I know they don't even have an emulator that can run x64 code on an Itanium.

-- and of course nothing whatever can go wrong with a *nix based platform used in the same environment.

There is also the small natter of FDA approval:

The 8-pound in-home gadget connects caregivers and patients outside of hospitals or clinic settings. It manages vital-sign collection, patient reminders, educational content, and motivational messages. The device has a 40GB hard drive. Information collected by the device is sent to the health care professional, and from there, physician and doctor can engage in video conferencing to discuss health issues. Doctors monitor and remotely care for their patients via an online interface using software called the Intel Health Care Management Suite. It currently runs on Windows XP only.
With the ability to hook up to wired and wireless monitors, such as glucose or blood pressure gauges, a caregiver can schedule times to remotely measure vital signs, or patients can check their own. The encrypted information is sent to a remote database, as long as the device connected to the Internet via broadband.
The Intel Health Guide PHS6000 received FDA clearance to enter the market after years of development and research, including pilot studies in the United Kingdom and the U.S. Intel said it expects the product to be commercially available from health care providers by late 2008 or early 2009 Intel's in-home health device gets FDA nod [July 10. 2008], Intel Health News

The purpose of the device is to support home care for the chronically ill. Home care is cheaper. Patients tend to remain more active, engaged and independent.

clicky

One more thing... Intel has a new 4-drive (SATA2) NAS server which is only $430 on Amazon, and comes with EMC software, or you can put FreeNAS, OpenFiler, or WHS on it.
Model # is SS4200E and SS4200EHW

The EHW doesn't come with the EMC software, and is supposed to be $100 less, but the actual price difference right now is less than $30.

They also have 2 eSATA ports.

Or, you can get a Chenbro NAS case with 4 hot-swap SATAII bays. It's about $230 with the power supply, and holds an Mini-ITX motherboard.
The new Intel Atom boards are really cheap, although the VIA C7's currently use less power.

I suspect that if WHS gets more popular, a lot more cases like this will start popping up, hopefully cheaper... $220 is awfully high for a mini-ITX case.

Theres alot of arguments in the replies about things unrelated (like the symantics of raid levels.

My personal option at home was (for some time) a very cheap amd mb with a gig connection and 6 sata disks 3 medium sized fast canavores striped and 3 larger (slower) disks (used as raw volumes for backup with bacula. Some of this storage was on shares, others were used as iscsi volumes and after trying a few of the bits of software out there (freenas/openfiler) i finally decided just putting fedora (any linux based os will do, ubuntu, suse, whatever floats your boat) and webwin was perfectly satisfactory (later it all got replaced with a very speedy xeon machine work gave me and so now i have no real nas as such.

There are a number of devices that i have used (the linksys nslu2 for example) which are perfectly good (only 100mb ethernet though). And i've heard good things about the d-link dns-323, both exceptionally low power devices.

But also the other day i saw these:
http://www.intel.com/products/motherboard/d945gclf/ and http://www.intel.com/products/motherboard/dg45fc/ which both look like fantastic little products to build off. Ultimately, there are quite a number of good products out there that do it "out of the box" like the thecus things, but I personally prefer something a little more flexible then they prefer.

Theres alot of arguments in the replies about things unrelated (like the symantics of raid levels.

My personal option at home was (for some time) a very cheap amd mb with a gig connection and 6 sata disks 3 medium sized fast canavores striped and 3 larger (slower) disks (used as raw volumes for backup with bacula. Some of this storage was on shares, others were used as iscsi volumes and after trying a few of the bits of software out there (freenas/openfiler) i finally decided just putting fedora (any linux based os will do, ubuntu, suse, whatever floats your boat) and webwin was perfectly satisfactory (later it all got replaced with a very speedy xeon machine work gave me and so now i have no real nas as such.

There are a number of devices that i have used (the linksys nslu2 for example) which are perfectly good (only 100mb ethernet though). And i've heard good things about the d-link dns-323, both exceptionally low power devices.

But also the other day i saw these:
http://www.intel.com/products/motherboard/d945gclf/ and http://www.intel.com/products/motherboard/dg45fc/ which both look like fantastic little products to build off. Ultimately, there are quite a number of good products out there that do it "out of the box" like the thecus things, but I personally prefer something a little more flexible then they prefer.

I heard a lot of good from friends of mine about the Synology Cube Station CS407, and that's the one I have on order now. I like the fact it's expandable, I'm e.g. planning to run a Squeezebox server on it. It has good support, and a large user community.

Others I heard about: Intel SS4200-E (Helena Island). It exists in two versions, one with an embedded OS on a flash and one without any soft. The one with software included has not that much possibilities and is not expandable, it's in the category "it just works." For the other version, I heard installing Linux or Windows Home Server on it is a PITA...

The ReadyNAS by Infrant (recently bought by Netgear) also gets good comments.

An Intel SS4200-EHW. They're brand new - only out for a few months. They support 4 SATA hard drives up to 1 TB each, RAID 1+0 & 5. We've set up a couple and they pretty well rock.

They also have an "E" version that has a pre-installed OS.

Erm, when did CPUs stop showing exponential growth in performance? Was that a memo that nobody sent to Intel?

Although clockspeeds are stuck because it is no longer economical to raise them, performance and transistor density are still scaling at the same rate. If anything we are in a period of performance increases that is slightly above trend, because now that the horrific NetBurst ISA has been killed off the Core2 replacement is rather lovely. Clock-for-clock it runs twice as fast as the old ISA because of shorter pipeline stages that have reduced instruction latency, and so far Intel have doubled the number of cores every 18 months. Given that they are ready to scale up to new fabs that can handle 2B transistors I would assume that they can continue to do so for the near future.

It would be a seismic shift for the industry if processor performance flatlined but I don't see that happening for a long time. What we are seeing with the introduction of the Eee Pc et al is actually a trend that has been going on for decades. Roughly every ten years a new form factor is introduced at the bottom of the market, with the same performance, but with the price halving each time.

So although your analysis of what changes are happening is way off, your final paragraph is quite accurate about what it means. The amount of performance that people actually require for most day-to-day tasks was exceeded when processors passed the Ghz mark. Now we are seeing cheaper and cheaper devices that deliver that (roughly) constant power. The effect on Microsoft is likely to be as you predict.

Especially since there is code available... At least for the machine vision part. OpenCV anyone? There are plenty of videos on youtube.

It's amazing what you can do with a few of these (a hundred bucks at fry's) and some these, a few of these and some creative sheet metal work on one of these.

You'll need a few other bits too. If you get carried away it would look something like this. If you keep your wits nobody would know it from a typical filing cabinet except that instead of storing files it renders frames with 32 cores running at 2.6GHz or launches your precious VMs.

And you can still remote to it with your mini notebook from the beach.

No, GCC only killed the market for C and C++ compilers

Yeah, and among others, they surely drove Intel out of business...

Oh wait, they are selling a
Intel® Compiler Suite Professional Edition for Linux $1,299

From the Terms and Conditions (http://tryhavok.intel.com/terms.php), it seems as though:

You can distribute a Havok-enabled game, as long as Havok cannot be separated from it by the end user.

You can distribute game middleware/game engines/game tools as long as Havok is not included in them at all (I guess the end user will have to get their own license)

Where game mods fit into this I am not sure.





I'm not a lawyer, blah blah blah

The above sentence is self-contradicting)

The chipset in the platform integrates both the north and southbridge and consumes a whopping 2.3W.

Not sure what that picture is supposed to be, since you didn't link any context, but it's certainly not of their mobile offering, which comes in at under five watts chipset inclusive.

Also check out the specification update for the X38 chipset page 8.

Also check out the specification update for the X38 chipset page 8.

Newer processes have always yielded faster, smaller, and cooler chips. Not anymore. 60nm didn't make chips use less power and 45nm doesn't help either.

Newer processes have always yielded faster, smaller, and cooler chips. Not anymore. 60nm didn't make chips use less power and 45nm doesn't help either.

According to the Intel datasheets, the 945GC northbridge has a TDP of 22.2W - nearly six times that of the processor!

http://www.intel.com/design/chipsets/designex/307504.htm

Perhaps Intel should concentrate on cutting the power of all the components and not just the processor.