Slashdot Mirror

... at least Windows 2003 does on HPs: http://docs.hp.com/en/A5201-96043-en/apcs03.html I don't know about Windows XP 32-bit, but since I already have a licensed, legal copy of Windows 2003 Enterprise, I'm good to go!

Also, you can use the Boehm garbage collector as a leak detector.

Use a Wide format printer. it prints at 300 dpi on 42" by 50ft paper. 24,000 sq in printable area per sheet. So excuse my math but that would be 2,160,000,000 bits per page per side right?

The documentation is dated February 1997.

There is one (1) operating system with only one (1) local vulnerability (in older releases) and only one (1) denial of service (all releases): VMS . Certainly outstanding! But, I bet the media will not notice.

It is good for a language to have threads "built in". As mentioned in this paper, "Threads Cannot Be Implemented as a Library": http://www.hpl.hp.com/techreports/2004/HPL-2004-20 9.pdf

if you do threads in a library, you run into problems with semantics or performance. Semantic problems == compiler breaks your multithreaded program. Performance problmes == compiler does naive translation of program, terrible performance.

Rather than your average pseudo-commercial list of branded devices, it's a list of improvements

Scroll down a bit.....

Pogue is talking about TVs that have all their connectors in the front, not just a set of A/V inputs, like your Sony and mine. He describes the HP TV in TFA.

In fact, the HP does appear to have some interesting ideas - the cables all slide under the TV and connect to a hidden, lighted panel at the front of the "set".

As has been posted, the summary doesn't really make it clear.

The "innovation" is to have all the connectors on the front. Not just one of the sets of A/V inputs.

An example is at:
http://h10058.www1.hp.com/digital/entertainment/us /en/img/theater/mdtv_guide/connect.jpg

Whoops, found one:

http://h10058.www1.hp.com/digital/entertainment/us /en/theater/tvs/mdtv_guide.html

Scroll down a bit..... not a bad idea, it seems to work

What you are describing is the Principle Of Least Authority. PLASH (Principle Of Least Authority SHell) is a nifty project to tackle this at the application level for Linux http://plash.beasts.org/. HP Labs has a project called Polaris which does this for windows http://www.hpl.hp.com/research/mmsl/projects/adv/p olaris.html.

Of course, one of the most commonly used commands is not a verb. See if you can remember which one. (I'm not counting the commands for invoking compilers, none of which are verbs either.)

The "alias" comes in two varieties. The symbol can hold any arbitrary content and can be used to abbreviate commonly used commands together with calling parameters if you didn't like the default ones, as you describe. They function as variables in DCL. The logical name is used for abbreviating or abstracting devices or pathnames.

Granted this an extremely recent phenomenon, but options are appearing in the marketplace:

http://www.sun.com/desktop/index.jsp

http://h10018.www1.hp.com/wwsolutions/linux/certif ications.html

http://linux.dell.com/desktops.shtml

I'm cheap, but even I'd spend the extra $200 to have a screen twice the size, 4x faster processor, CD-RW/DVD, etc etc etc. Even if I was given this for free I would use it very infrequently, even before I had the laptop I'd probably just wait until I got to a desktop.

Might be nice for people who have no access to a PC and really can't afford a few more dollars to buy a real laptop or who have little access to electricity like soldiers on a battlefield.

It's all well and good for TI to benefit from the open source community. But TI still refuses to publish their WiFi information for open source driver developers.

In 2001, TI (Texas Instruments) decided to make a big push on the 802.11 market. ... From the start, TI has refused to give any help towards a Linux driver and have decided to totally ignore the Linux community.

Sure it's all great to see some more uptake of Linux, but beware that TI has not shown itself to be a great friend in the past.

Nice to know drivers available for HP and Epson!

I'm interested on any one of the following printers:
HP Photosmart 8700
Epson Stylus Photo R1800
Epson Stylus Photo R800

When I goto HP, this is what I see:
http://h10025.www1.hp.com/ewfrf/wc/softwareCategor y?product=426109&lc=en&cc=us&dlc=en&submit.y=7&sub mit.x=5&lang=en&cc=us

When I goto Epson, this is what I see:
http://www.epson.com/cgi-bin/Store/support/supDeta il.jsp?BV_UseBVCookie=yes&oid=14121&prodoid=535409 19&infoType=Downloads&platform=All

Do you see anything different than I see?

Could you post hear the location of the drivers for the above printers?

OK, I understand this reasoning but I just don't agree.

the browser is becoming the app

And has been becoming "the" app for 20 years. Now its great for many things, but replacement for everything??? Don't think so. The browser "taking over" is like the flying car. Its always just around the corner ;-)

Couple this with the growth of thin clients

You should have been around in the 70s and 80s ;-) Almost everything was thin client (terminals). I assume you weren't around then so trust me. There has been an AMAZING shrinkage in thin clients, not growth.

a full blown $2000 desktop computer

Completely agree, but I think you need to shop around a bit more ;-) No they don't need a kick-ass gaming machine, but why spend $300-$500 for some impotent little thin client when for basically the same price you can get a mahcine that can do everything your thin client can plus actually have some balls (just in case needed)?

for 90% a great webmail client is going to be all ppl

At home... maybe. In a corporate environment... no.

Think server-side, think thin clients

Think 1970, think 1980


OK, all that said I do think there are cases where what you suggest may work fine. However, to think this is a wide-spread solution for "normal" cases is fooling yourself. I like the idea of it and I think ideally centralized computing has many advantages. At the same time, I also think a flying car should be doable and would have lots of advantages. However, I'm not betting on seeing either in wide use anytime soon ;-)

OK, I understand this reasoning but I just don't agree.

the browser is becoming the app

And has been becoming "the" app for 20 years. Now its great for many things, but replacement for everything??? Don't think so. The browser "taking over" is like the flying car. Its always just around the corner ;-)

Couple this with the growth of thin clients

You should have been around in the 70s and 80s ;-) Almost everything was thin client (terminals). I assume you weren't around then so trust me. There has been an AMAZING shrinkage in thin clients, not growth.

a full blown $2000 desktop computer

Completely agree, but I think you need to shop around a bit more ;-) No they don't need a kick-ass gaming machine, but why spend $300-$500 for some impotent little thin client when for basically the same price you can get a mahcine that can do everything your thin client can plus actually have some balls (just in case needed)?

for 90% a great webmail client is going to be all ppl

At home... maybe. In a corporate environment... no.

Think server-side, think thin clients

Think 1970, think 1980


OK, all that said I do think there are cases where what you suggest may work fine. However, to think this is a wide-spread solution for "normal" cases is fooling yourself. I like the idea of it and I think ideally centralized computing has many advantages. At the same time, I also think a flying car should be doable and would have lots of advantages. However, I'm not betting on seeing either in wide use anytime soon ;-)

Sun

HP

IBM

Dell

The first 3 I know are for real. I don't know much about Dell, but much more prominenet like that you can't have it.

And altough other manufacturers (shame on them) are less helpful (Toshiba...) they may be weaking up to the reality of the marketplace that Linux is becoming. You would need to hack far less since many hardware manufacturers are realising Linux is here to stay. You did not have a resource like Toshiba's a few years ago.

The old pitiful excuse not to use Linux argueing there is no hardware officially certified to run it, should be soundly ignored.

...Who is this Martin Fink I'm hearing so much about? http://www.hp.com/hpbooks/authors/fink.html Yes, I hate his ads, too.

He's a snappy dresser, that's all.

...Who is this Martin Fink I'm hearing so much about?

http://www.hp.com/hpbooks/authors/fink.html

Yes, I hate his ads, too.

My former employer DC'ed some Tandem (Now HP Non-stop) servers, called Himalaya K1000's.

VERY expensive machines when new. Nice black roll around case about 5 feet high, 2 wide, with a swing open front door. I took it home, gutted out all the internals, and now it contains not only my PC internals, but also an HP/UX 133MHZ system, and a beer cooler in the bottom.

And a nice Intel Inside sticker!

Here is the pic!

http://h50146.www5.hp.com/products/servers/nonstop /casestudy/nttdata/images/08.gif

AC

Since I am making the assumption that you are a professional photographer, you can easily write this stuff off as a business expense, depreciate it appropriately, etc. What I would not do is follow anyone's advice on a "homebrew/homebuilt" solution. I am sure that there are plenty of people out there that *could* build something reliable as the solution I suggest, but personally, having worked at several LARGE companies in the I.T. department in various capacities, we would NEVER EVER build our own solution. Vendor support, etc. becomes increasingly important when data is critical to the business and downtime results in monetary losses

In 2003 some of the HP Labs researchers looked at the related issues and published a paper titled: "If Piracy is the Problem, Is DRM the Answer?" http://www.hpl.hp.com/techreports/2003/HPL-2003-11 0.pdf

You might find the white paper interesting if you've not read it before. This caused quite a stir when it was released, both inside and outside HP, and is still quite relevent in light of the Sony issue. This provides an counterpoint even inside HP where we try to maintain some form of management across all the issues.

The conclusion reads:

"We pointed out that unauthorized use and unauthorized acquisition are two aspects of piracy. A key concept is how licenses are bound to content. We saw that various kinds of DRM technology address these issues in very different ways, but that all of them have some kind of flaw that make it highly unlikely that they will be able to solve the problem of piracy. The real problem with piracy is that it takes only a small fraction of users who are capable of dissociating licenses from content to make managed content available to a significant fraction of users in unmanaged form.

We explored the concept of draconian DRM in which devices that handle managed content do not handle unmanaged content at all. Draconian DRM could potentially be effective at eliminating piracy if it were ubiquitously adopted, but introduces a new problem of how to handle public content.

Our conclusion is that currently proposed technical measures will not be able to completely stop the illegitimate distribution of pirated content. We believe that content producers must take steps to compete with the piracy as an alternative."

1. Ad-Aware SE
2. Spybot Search & Destroy
3. SpywareBlaster
4. Microsoft Anti-Spyware
5. Some Anti-Virus Program that you like (at my work, we install Norton even though it is a resource hog, but never Norton Internet Security since it eventually always fucks a computer up)

Set your Anti-virus program to scan at least weekly, and automatically update itself, Update and sca with Ad-Aware and Spybot weekly at a minimum, and update and protect with SpywareBlaster weekly at a minimum.

It is absolutely ridiculous that a person should have to do this to keep their computer running decently. We get so many Windows machines in the shop that it isn't even funny, but thusfar, whenever we have managed to convince someone to upgrade to a MacOS X machine (Typically when their Dell, Compaq, HP, E-Machines has a motherboard failure). They have came back completely excited and astonished that they don't really have to worry about spyware and viruses so much.

My reccomendation on keeping your WIndows XP machine in top performance. Go buy a high-end Mac and run VirtualPC if it can run whatever program you NEED to run (Note: Games do not count), if you cannot run your Prorgram under VPC, buy a low-end PC and keep it off the network.

While we are on the subject of Quantum Mechanics. Check out Caltech's website on Quantum Computers.

I would also like to put you towards HP's Research on it.

The future is quantum mechanics, no matter the subject.

If you check the beginning ofthe 2.6.14 changelog you will find mention of a "ieee80211 subsystem" being merged. So hopefully all drivers will be ported to the generic in kernel 802.11 stack. This will hopefully take over WPA crypto duties from drivers. This also goes hand in hand with version 28 of Wireless Tools which will provide a basic (but standard) interface for WPA.

Yes the *BSDs are further down this track as pointed out in another comment buy it's nice to see Linux catching up and I believe HostAP and the Centrino drivers have already been converted over to using the generic 802.11 stack.

As an OpenBSD user I can tell you that it does not have the broadest support. Let me back this counterclaim up.

OpenBSD supports the following chipsets (as taken from the OpenBSD i386 hardware compatibility page on 19th November 2005:

ADMtek, Aironet, Atheros, Atmel, Centrino (2100, 2200), Prism 2.5/3, Ralink (2500), Raytheon and Realtek

Now by comparison a Linux distro (e.g. Ubuntu) can have support for the following chipsets (list taken from Linux WLAN Howto cross referenced against Ubuntu) on 19th November 2005:

ADMtek, Aironet, Atheros, Atmel, Centrino (2100, 2200), Prism 2.5/3, Ralink (2400, 2500), Prism GT, Raytheon, Texas Instruments ACX100/110, Wavelan

A pretty similar list wouldn't you say (OpenBSD has Realtek which Ubuntu doesn't, Ubuntu has Ralink 2400, Prism GT, ACX100/110 and Wavelan which OpenBSD doesn't)? In fact, Linux has Realtek drivers too but as they aren't shipped in Ubuntu I left them off the list. Linux also has ndiswrapper and Linuxant Driverloader allowing the use of Win32 drivers but I'm discounting non native drivers.

Chipsets that aren't supported by either OS that are significant:
Broadcom - There's a heck of a lot of Broadcom stuff out there and no sign of open drivers ever. That's their perogative but this stuff is all over the place...
Marvel - a new 802.11g player as far as I can tell. No open source drivers that I know of so far.

For the meantime, chipsets like Broadcom mean that open source OSes will always have a more troublesome than Windows with random wireless drivers. Choose carefully and don't reward vendors with non free drivers where possible.

I'm not sure exactly why this is classed as a printer of 2005 (it was released in 2004), see here for a good review on it.

Looks like they finally released a 10.4 driver for the PSC series in September.

Not only that, but the compiler can reorder memory accesses during optimization. This paper goes into details about why multithreading can not be implemented correctly without special compiler support. (And though the paper's abstract does not make it clear, pthreads does have the support necessary for correctness.)

1. FCC regulations change.
2. Radio transmissions are tricky things. A Wifi producer might release a patch that provides greater range by constantly modulating the power levels, yet remaining within the FCC's guidelines.
3. It's about a lot more than power. You can accidently generate a massive amount of broadband radio noise if you're not careful. Basically, the software is playing HAM operator with the Wifi card in order to tune the best performance out of it without running afoul of regulations. That's a good use for software.

1. FCC regulations don't change that often.
2/3. You are arguing for a flexibility in the cards that is unneeded/unused. Vendors still want to release new cards to follow changing technology. Speaking of which, where were your simple firmware updates when WPA came out, and now WPA2? Why did I have to buy a new card to support it instead of just updating my driver? Despite the advantages of having a software driven card, vendors still sell hardware. They don't want to sell software, so they are never going to design their cards to be entirely software dependent. The issue now is the stuff that they chose to make software dependent. Your card pretty much operates at a constant power and frequency through its lifetime. Any code to modulate and regulate the transmitter is fairly constant and doesn't need to be in software. For bug fixes, do the same thing that the router people and controller card people have been doing for years: store the code in flashable firmware *on the card* and then post firmware updates on your website.

You're
not
listening

I know what you are saying. The problem I have is with this: Thus they licensed NOT OpenGL, but all kinds of bits and pieces of 3D trickery, microcode, and hardware design thats patented by someone else, produced by someone else. Most of the original patents on "3D trickery" were owned by SGI and purchased by NVidia. Along the way NVidia has improved the technology and filed for its own patents. NVidia owns all of the technology used to drive their cards. If somebody else was doing the innovating they wouldn't be at the top of their game. And, by the way, patented technology does NOT mean you can't show people the code. In fact it means quite the opposite. A patent enables you to release the design into the public so that everybody can see how it works without stealing your idea. Licensing code under an NDA only happens when a patent is not involved because that is the only way to protect it. If that is the situation NVidia is in, that seems pretty silly. Anything that is not patented NVidia is free to implement themselves, something they are quite capable of (it's just math, like you say). Considering how competitive the market is, it would be stupid for NVidia to rely on a 3rd party for something as important as the code to drive their card. Oh, and did you notice that NVidia calls their driver a "unified driver architecture." That is because they have managed to abstract all of their 3D trickery so that they can just slap a new card in with instructions for how to drive it and have it work without rewriting the entire driver. So if NVidia can't release the 3D code, why can't they release the specs for the card? You still haven't addressed that.

That's why there are full drivers for NVidia, right? No wait, there aren't.

That's why there are full drivers for WinModems, right? No wait, there aren't.

That's why there are full drivers for WiFi cards, right? No wait, there aren't.

The NVidia driver is a difficult issue. A lot of people are content with the binary drivers for now, but if NVidia stopped releasing them tomorrow, quite a few people would step up to the plate to write a new driver.

There is at least one open source driver for a Winmodem, the IBM Mwave. There wasn't a lot of work put into these because not very many people care about them.

There are quite a few drivers for wifi cards. See this.

The purpose of calling for federal regulation is to keep costs down, not to protect privacy. Some companies are actually interested in protecting privacy because failure to do imposes costs. HP is particularly good in this regard in that it lets customers access their data. The companies you have to watch out for are the ones with business models that depend on selling personal information.

Even a dual athlon (multi-core or not) is likely to have an internal NUMA architecture. I just bought some new servers that have exactly those properties.

Just buy more Superdomes and partition away.

They need 2x 60amp feeds, but off those feeds you can run 8 hard partitions, and many virtual partitions (vPars) on top of those ...

Much more convenient and flexible than dealing with blades, or even small 2u-type servers. And the reliability can't be beat..... the only thing that outdoes the Domes for us is the mainframe, and well, the Dome is a lot cheaper. :)

Some hardware vendors will offer Debian support.

HP offers support for Debian for many of it's products.

http://h20219.www2.hp.com/services/cache/76815-0-0 -225-121.html

I think the most accurate statement you can make about sticking that government sticking with Novell is that they are probably happy with the service that Novell provided in the past and that means that they will likely like the support provided by Novell into the future.

Why bust a good thing?

Anyways. Novell has the most expertise when it comes to desktop and groupware systems. They are the ones that basicly created the first network directory system for desktops that the average admin could deal with! Microsoft basicly took all of Novell's concepts when it created AD. They made enterprise desktops what it is today.

It's just tough to compete with Microsoft 'enterprise/groupware' software when Microsoft owns the only viable desktop platform.

As far as large desktop deployments goes Novell would be my first choice.

If I needed to combine open source software technology like Lustre, GFS, and clustering file services with a large IBM NAS storage solution with support from Oracle and Veratis software.. I'd choose Redhat.

But I dont' think that 90% of Redhat's engineers haven't touched a 'real' desktop since they stopped using Windows 95.

I mean just look at 'bluecurve'!

Novell is to desktops and desktop support systems what Redhat + IBM is to servers.

Also, there's the issue of support with proprietary software on Linux. There's always support for RedHat and SLES.. never Debian.

HP offers support for Debian on their servers.

http://h20219.www2.hp.com/services/cache/76815-0-0 -225-121.html

There are many individuals and companies that support Debian. Some are listed on Debian's consultants page:

http://www.debian.org/consultants/#US

Add the fact that if they used Debian or BSD or whatever free-soft-distro-of-the-month is, they have no one to yell and scream at when something goes wrong.

There may be other companies as well, But, no suport for debian is an out right myth, no matter what type of support you are talking about.

Sure, it quite possibly *can* be done. In the same way that you could theoretically build a spaceshuttle from a T-Ford and lots of old spraycans!
When you're aiming for extreme solutions, you have to use quite extreme components. Think http://www.sgi.com/products/storage/ http://h18006.www1.hp.com/products/storageworks/xp 12000/index.html or even http://www.sun.com/storage/highend/9990/index.xml.

The only imaginable way to get petabyte on commodity hardware that I can think of is to build a seriously huge Beowulf cluster. But putting a singular FS on about ~500 seperate computers in a cluster is rather madness...
Take the good advice many fellow slashdotters has made, do NOT use a singular filesystem that spans 1 PetaByte or more...

Links to MCE 2005 systems(their e-stores)
I know this takes all the fun out of it, but sometimes it really is just too much work
to get a computer to do stuff that should be easy(i.e. Suse vs. Gentoo; i use both).
Consider these a few points of reference for your plans for World Domination
("...What are we going to do tonight Brain?..." ;-)

Shuttle: Shuttle m1000
which looks like a 'normal' audio/video component, and a variety of SFF-based systems from 899$US.
the advantage of the SFF-based systems would be customizability(video cards up to 6800gt, HDDs to 400GB
(three drives in a P-series chassis=1.2TB),

HP(Hewlett-Packard): HP z500 series
also a 'component' style chassis, five models of varying performance and capacity, also customizable.

Gateway: Gateway FX400
sadly, all towers, but customizable(dual-core!)

Sony's newest vaio system: Sony VGX-XL1
a bit pricey at 2300$US but totally full featured with a dual-core P-D 820, and a 200-disk optical jukebox
NOT customizable.

(why can't i get the HP link to NOT be green? OR, better yet, why can't i get the other links to BE green? bah.)

That's not uncommon. When a company doesn't want to carry/sell a product, instead of saying no, they just price themselves out of the market. That way: a) The customer never hears a, "no", which is something to avoid. b) If someone actually does buy from you at that price, what the hell, you made a buttload.
I ran into this on my home printer. I bought an HP 2550 printer (for doing all of the printing for my wedding). It comes standard with 64MB of RAM. This is plenty until you start sending graphics to the printer. So to stop the "Out of Memory" errors, I decided to upgrade the memory. The printer would handle an extra 128MB SODIMM.
Price from HP: US$800
My response: Bullshit!
Price from Kingston: US$50
And, it only took me moments to find the right part with Kingston's website (they have a really nice memory finder). Also, Kingston offers a lifetime warranty and puts out a solid product, so no worries about a fly by night company.
So, in the end, I got what I wanted and HP got to stay out of the memory business, without ever explicitly telling me "no".

http://h10010.www1.hp.com/wwpc/us/en/sm/WF05a/3219 57-64295-89315-321838-f1-445351.html

He wants to do something that is fundamentally complex, which is edit photos. Okay, he wants to remove red eye? He's going to have to tell the program where to remove the red eye from. ... Of course it's complex. What does he expect? A miracle? Artificial intelligence?

Not to nitpick, but HP actually has a system which does precisely that:

http://www.hpl.hp.com/news/2004/apr-jun/redbot.htm l

HP researchers have developed two approaches to detecting and correcting red-eye. The first exploits an earlier algorithm that accurately detects faces in photographs or real-time video. The second is based on a detector trained specifically to find red eyes.

A Labs-based implementation of these algorithms led to HP's In-Camera Red-Eye Removal, which instantly removes red-eye from photos while they're still in the camera, without using a PC and graphics software -- an industry first.

Introduced in May in HP's new Photosmart R707 digital camera, the red-eye removal feature is one of several imaging technologies that originated in HP Labs and are now appearing in a line of new HP cameras. ...
Beginning in mid-2003, the scientists tested RedBot on about 1,700 photographs submitted by HP employees through a company intranet. The algorithms found and fixed red-eye about 90 percent of the time. "That's a very good success rate," Ulichney says, but researchers want to push the percentages even higher by running their algorithms on a much larger sample.


They have a site where you can upload photos to run the red-eye detector on, but it doesn't seem to load well right now.

He wants to do something that is fundamentally complex, which is edit photos. Okay, he wants to remove red eye? He's going to have to tell the program where to remove the red eye from. ... Of course it's complex. What does he expect? A miracle? Artificial intelligence?

Not to nitpick, but HP actually has a system which does precisely that:

http://www.hpl.hp.com/news/2004/apr-jun/redbot.htm l

HP researchers have developed two approaches to detecting and correcting red-eye. The first exploits an earlier algorithm that accurately detects faces in photographs or real-time video. The second is based on a detector trained specifically to find red eyes.

A Labs-based implementation of these algorithms led to HP's In-Camera Red-Eye Removal, which instantly removes red-eye from photos while they're still in the camera, without using a PC and graphics software -- an industry first.

Introduced in May in HP's new Photosmart R707 digital camera, the red-eye removal feature is one of several imaging technologies that originated in HP Labs and are now appearing in a line of new HP cameras. ...
Beginning in mid-2003, the scientists tested RedBot on about 1,700 photographs submitted by HP employees through a company intranet. The algorithms found and fixed red-eye about 90 percent of the time. "That's a very good success rate," Ulichney says, but researchers want to push the percentages even higher by running their algorithms on a much larger sample.


They have a site where you can upload photos to run the red-eye detector on, but it doesn't seem to load well right now.

Then what the heck are these? Being as HP is betting its entire line on Itanium (OpenVMS, HP/UX, and Windows), failure would probably require a collapse of the entire industry (or cause it).

My HP Jornada 820 is instant-on. It's a WinCE 2.1 machine in subnotebook format. Real keyboard, VGA display, modem, USB for external mouse, trackpad, CF and PCMCIA slots (so ethernet is installed in the first and 128M storage lives in the other), OS, IE, Word, Access, Excel, Outlook Express, Powerpoint all in ROM. Ten to twelve hours battery life.

It's a really nice machine that I take on trips in preference to "real" laptops. Never understood why HP stopped making them and other manufacturers never started.

Paul

The difference is no longer an order of magnitude or two even for the SMP boxes that you are talking about. Googling a bit, I have found these reasonably priced Opteron boxes. These are variations of the same opteron v40z boxes that Sun sells. Quad Opteron boxes w/ hot swap power supplies and hot swap hard drives w/ ECC memory for reasonably cheap for the chassis.

I've never seen a terabyte of ram on a single domain but the
very high end sun boxes will allow you to use either 576 or 288 gigs of ram on a single box, so I will use that as the base. The maximum in SMP mode for the opteron boxes you can homebrew is probably one based on the iwill motherboard which supports 64 gigs, which is now grabbing the midrange line of the sun server series.

In fact HP makes the ProLiant DL585 with opteron chips. They will sell you this machine w/ up to 128 gigabytes of ram, redundant roms, redundant power supplies, hot swappable hard drives, ECC memory.

Its not quite at the high end server sun range but the opteron chip and its use in machines built by traditional "big iron" builders is not really an order or magnitude off. Its really about a factor of 5 from the really high end (aka $1m+ sunfire installations).

And as the IWILL MB points out its not even that far away from home brew or at the very least a $100k officebrew given the amount of ram one has to buy.

Also its worth pointing out the processor supports up to a terabyte, the issue is the memory sticks/mb's are not available yet.

And here we are only talking about the SMP/redundancy style computing . For distributed style computing platforms (aka everything on the top 500 super computers), the 10th and 11th fastest computers in the world are Cray machines based on opterons currently running at Sandia National Labs and Oakridge National Labs respectively.

Now perhaps in your world the top 500 super computers are not considered big iron because they are distributed but i'm w/ the joe slashdotter crowd that considers "big iron" to include the top 500 fastest computers in the world.

Here's a slightly more detailed look, genius. http://www.hp.com/products1/servers/integrity/supe rdome_high_end/specifications.html

You sir are an arrogant asshole. For me this is "commercially-available" enough.

It seems that the defect has to be potentially life-threatening for HP to react. I have an HP ZD7000 laptop. There is a known fault in something to do with the RAM controller wherein - if you have the secondary RAM slot filled - the laptop will reboot or shutdown spontaneously in instances of applications that have heavy memory usage (I'm assuming those that have requirements of memory from both slots). Generally the problems have been noticed in photoshop, but I've had them occur in GIMP or some games. Others have been experiencing the same problems. Adobe has a warning on this.

So I've contacted HP technical support about this. I've talked on the phone, and then by email. The representative from HP assured me that no such issue existed, and we back-and-forthed for awhile. Eventually, I found this article on HP's own website. When I emailed it to the HP rep, he prompted stopped answering my emails.


Maybe if my battery had exploded I would have gotten better support from HP, but it seems it has to be a big issue for them to do anything about it.

"HP fully stands behind the products it makes?" Maybe, but only when it looks like it might lose them money due to lawsuits or poses a health risk.