Oil-Immersion Cooled PC Goes To Retail

notthatwillsmith writes "Everyone's seen mods where someone super-cools a PC by submersing it in a non-conductive oil. It's a neat idea, but most components aren't designed to withstand a hot oil bath; after prolonged exposure materials break down and components begin to fail. Maximum PC has an exclusive hands-on, first look at the new Hardcore Computer Reactor, the first oil-cooled PC available for sale. Hardcore engineered the Reactor to withstand the oil, using space-age materials and proprietary oil. The Reactor's custom-manufactured motherboard, videocards, memory, and SSD drives are submersed in the oil, while the dry components sit outside the bulletproof tank. The motherboard lifts out of the oil bath on rails, giving you relatively easy access to components, and the overall design is simply jaw-dropping. Of course, we'd expect nothing less for a machine with a base price of $4000 that goes all the way up to $11k for a fully maxed out config."

Comments

Hmm.

[Anonymous Coward]

There is a joke involving Natalie Portman and hot oil here, but I just can't quite find it :(

Re:Hmm.

[Shaitan Apistos]

Too close to 4:33, try again at 10:04.

Re:Hmm.

[plover]

All I can think of is "This sounds slick!"

Thanks, I'll be here all the week. Try the veal.

Re:Hmm.

[Anonymous Coward]

Great,

So now I have to change the oil in my car AND in my computer? What, every 5,000 petaflops?

Re:Hmm.

[inKubus]

That's the SOFTCORE Reactor you're thinking of. And it's the only computer cooled with 100% grits.

Thanks God

[MyLongNickName]

When I am web surfing and playing solitaire, I can't afford to have my CPU or graphics card overheat. And don't even get me started on email.

Re:Thanks God

[frosty_tsm]

When I am web surfing and playing solitaire, I can't afford to have my CPU or graphics card overheat. And don't even get me started on email.

You're using Vista, aren't you.

Re:Thanks God

[blair1q]

We can slashdot that for you wholesale.

Bulletproof?

[derfy]

Bulletproof? Seriously?

Re:Bulletproof?

[gruvmeister]

Hey, if I'm paying 11 grand for a machine, it damn well better be bulletproof.

Re:Bulletproof?

[CaptainPatent]

They're trying to appeal to two growing demographics: hardcore gaming mobsters and gangsta autocad designers

Re:Bulletproof?

[afidel]

They probably mean bullet resistant aka Lexan.

Re:Bulletproof?

[mathx314]

Both of us will be very happy.

Re:Bulletproof?

[frosty_tsm]

Exactly, what happens when someone comes busting in and puts a cap in your gaming rig?

We all know what happened to the XBox when it got shot... the DVD was knocked off it's track.

(sorry, I looked but couldn't find a link)

Re:Bulletproof?

[Anonymous Coward]

Fo shizzle.

Re:Bulletproof?

[MentlFlos]

Bulletproof? Seriously?

Yes, but only for very small values of bullet

Re:Bulletproof?

[Dogtanian]

They're trying to appeal to two growing demographics: hardcore gaming mobsters and gangsta autocad designers

They're trying to appeal to the Scots- we'll deep-fry anything!

Re:Bulletproof?

[Anonymous Coward]

Here is how to test it (its old news really):
http://uk.youtube.com/watch?v=mAuKwTDGnCg

The hilarious part is at the end..

Re:Bulletproof?

[Anonymous Coward]

Yo bitch, this be Killahertz and you is creepin on mah terryterry!

I will put a motherfucking slug in yo motherfucking rig, you got me byatch?

Re:Bulletproof?

[Martin Blank]

Isn't most of your cuisine based on dares anyway?

Re:Bulletproof?

[Miseph]

'Dats right, best watch yoself befoh I blast those weak-ass technical diagrams sucka!

Re:Bulletproof?

[g0dsp33d]

Mock all you want, but you'll be glad for it when you experience the realism of GTA 6.

Re:Bulletproof?

[MadMidnightBomber]

I spent a year at University of Edinburgh, and I can tell you the parent should be modded informative, not funny.

Re:Bulletproof?

[theaveng]

$11,000?!?!? Ha! Not for me. I'd rather pay somewhere around $20 (cost of shipping) and use a slightly-older machine (like a 700 megahertz Pentium with XP).

I don't really need the latest hardware to watch Heroes at nbc.com.

Re:Bulletproof?

[Fred_A]

I concur, Scotland has to be the only place where they would even think of making deep fried risotto (although I couldn't find the elusive deep fried Mars bar despite a long search).
And yes I tried it.

Regarding the PC, I prefer not to do my cooking under my desk, so I'll stick with my current rig.

Re:Bulletproof?

[ElectricTurtle]

TFA said the starting price was $4k. I could see doing that, I spent $2500 or so on my last system.

Re:Bulletproof?

[Bomarrow1]

Noo! The slug will dissolve in the oil and make it conductive.
Also can you imagine all those horrible trails over your shiny ICs?

Space age materials?

[Burning1]

Seriously? Space age materials?

Did anyone inform them that the space age was the 1960s?

Re:Space age materials?

[jandrese]

Yeah, but compare that to air cooling. Air is HOW old?

Re:Space age materials?

[KDEWolf]

Yeah, but compare that to air cooling. Air is HOW old?

Mod parent up! =P

Re:Space age materials?

[The Orange Mage]

Yeah, but compare that to air cooling. Air is HOW old?

Older than dirt...I mean John McCain?

(On that note, do HTML strikethrough tags not work? I couldn't get the deprecated or working, or the proper either.)

Re:Space age materials?

[value_added]

Yeah, but compare that to air cooling. Air is HOW old?

About the same

Re:Space age materials?

[DrData99]

About 6000 years if you listen to some people...

Re:Space age materials?

[Anonymous Coward]

> Yeah, but compare that to air cooling. Air is HOW old?

I dunno. McCain won't return my calls so I don't have anyone who knows first hand. But Palin seems to think it's about 6,000 years old, give or take.

Re:Space age materials?

[ari_j]

This is actually a step backward. The computers we had when I was your age were cooled by a proprietary oil. Of course, back then "proprietary" meant "whale" and "computer" meant "steam-powered abacus," but this isn't one lick different. Those who don't study history are doomed to repeat the whale-oil-cooled, steam-powered abacus, it seems.

Re:Space age materials?

[garaged]

and oil is way cheaper !!

so much for quick repair

[Brigadier]

I'm not sure but like most extreme PC users i know. They are in the case at least once a day trouble shooting, or fixing something. (dead fan, dead drive, upgrades etc.) This pretty much kills that idea. If My PC seizes and I need to reset the bios, pull ram etc it's a no go.

I'm also curious how they address hardware which need to have breather holes like hard drives.

Re:so much for quick repair

[Anonymous Coward]

Did you even glance at the summary before you started typing?

"Everyone's seen mods where someone super-cools a PC by submersing it in a non-conductive oil. It's a neat idea, but most components aren't designed to withstand a hot oil bath; after prolonged exposure materials break down and components begin to fail. Maximum PC has an exclusive hands-on, first look at the new Hardcore Computer Reactor, the first oil-cooled PC available for sale. Hardcore engineered the Reactor to withstand the oil, using space-age materials and proprietary oil. The Reactor's custom-manufactured motherboard, videocards, memory, and SSD drives are submersed in the oil, while the dry components sit outside the bulletproof tank. The motherboard lifts out of the oil bath on rails, giving you relatively easy access to components, and the overall design is simply jaw-dropping. Of course, we'd expect nothing less for a machine with a base price of $4000 that goes all the way up to $11k for a fully maxed out config."

Re:so much for quick repair

[Gat0r30y]

RTFA - Drives are in separate bays for easy access. HDD's would not fare very well in oil, even proprietary super secret oil. Quite the impressive piece of hardware. - Your not on trouble shooting does raise a good point though. Seriously, what do you do when you have a problem with this thing? And redundant PSU's? They must know that these will be the first to go, and a tremendous pain to replace.

Re:so much for quick repair

[Brigadier]

I don't know about you, but on my system drive arrays generate the most heat, and require quite a bit of cooling.

I don't see the practicality of having all this case hardware to support what looks like a few liters of oil if the hard drives still have to be cooled by fans.

This feature would only be useful to me in creating a fanless ultra quiet system.

Re:so much for quick repair

[Hockney Twang]

The storage provided by the manufacturer is SSD (three of them, in RAID) and is submerged, but they have bays for removable drives you might want to add on your own. Also, in response to your comment about resetting the CMOS, they have a button that does that on the case (behind a little plastic door so you don't accidentally hit it).

Re:so much for quick repair

[CodeBuster]

RTFA - Drives are in separate bays for easy access. HDD's would not fare very well in oil, even proprietary super secret oil.

Which is as it should be. Most of the noise from your drives (DVD/CD, HDD, etc...) is going to be from vibration transferred to the chassis and not from airflow. This can be easily mitigated by rubber standoffs so that there is no metal bolted directly to metal (the Antec QuietPC cases have come standard with this feature for some time now).

Re:so much for quick repair

[Anonymous Coward]

Glance at the summary? You're new here, aren't you?

Re:so much for quick repair

[plover]

Dead fan -- I'm sure that's a problem with an OIL-COOLED box. I suppose the extreme PC users you know would also complain that they wouldn't be able to vacuum the dust from their heat sinks, too.

Now, a dead oil circulation impeller, that's a completely different animal.

Re:so much for quick repair

[Anonymous Coward]

Sounds like you got your idea of "extreme" from watching too much Star Wars. Technology doesn't have to be like the Millenium Falcon, you know. Some of us manage to have fairly high-performance computers that manage to stay up for a whole week without something breaking.

Re:so much for quick repair

[nog_lorp]

No, he's old here. It seems like it was within the last year when "RTFA" changed to "RTFS" (read the fucking summary) or even "RTFT" (read the fucking title).

Re:so much for quick repair

[dpiven]

Also, in response to your comment about resetting the CMOS, they have a button that does that on the case (behind a little plastic door so you don't accidentally hit it).

Yeah, wouldn't do to accidentally scram your desktop.

Re:so much for quick repair

[wild_berry]

HDD's would not fare very well in oil, even proprietary super secret oil.

They're sealed against dust. Is the seal not enough to prevent ingress of cooling fluid?

Re:so much for quick repair

[BiggerIsBetter]

Now, a dead oil circulation impeller, that's a completely different animal.

Actually it's pretty much the same animal, just with a different fluid to move.

Re:so much for quick repair

[BOFHelsinki]

I glanced through the MaximumPC article but that SSD RAID still doesn't make sense. The by far best advantage of SSD is the orders of magnitude lower seek time, and RAIDing simply kills that. There is room for a couple of hot-swap 3.5" HDs but no special cooling for them. Why not, say, an oil-tight aluminum 6-drive HD cage for some 15K SAS RAID-5 goodness, submersed in the oil, then a couple of SSDs firmly in JBOD? Meh. Idiot specs.

And at $4k you'd expect a totally fanless and silent solution. This had a radiator with four apparently 120mm fans. It's good to have lots of fan area, lower rpm means lower noise, but silent this case is not. Stovepipe effect cooling or a similar passive solution would be trivially easy to integrate to an already difficult to access design like this...

BTW, looking at how the SLI interconnect bends sideways in that special staggered placement... I'm not sure how good that is for long term.

But I have to admit that the case is very good looking, even with the obligatory blue LED bling-bling.

Overall, cannot help thinking that they got those millions of VC just as a reaction to the Alienware/Dell deal. Hop on the bandwagon...

Re:so much for quick repair

[Molochi]

HDDs have a little hole on the plate that covers the platter. It says do not cover. It's a vent that regulates air pressure in the drive IIRC.I don't think submerging them would be a fantastic idea anyway. On the other hand I don't see the point of better cooling on an SSD either. So I think they're just doing it because they can charge more for the system since this way you can't opt for raptors.

Re:so much for quick repair

[Molochi]

The first time the drives warmed up, air would exit the vent. when they cooled down the air left n the drive contracts. If oil doesn't force its way through the filter. It would probably tear the filter or compress the plate covering the platter.

I let you experiment with it, though. I don't see any need to provide special cooling for my HDDs.

Re:so much for quick repair

[wild_berry]

I think that I'll keep my drives cool with good airflow. I worry that the process of expansion and cooling with changes in temperature might reduce their life.

Re:so much for quick repair

[Molochi]

Yep, that's what I do too. Nothing special. I don't usually second guess the engineers when there isn't problem. That's what 3 year warranties are for.

Eww

[Hatta]

FTA: every centimeter of the machine oozes custom computing.

I hope that's all it's oozing.

Re:Eww

[manXxon]

FTA: every centimeter of the machine oozes custom computing.

I hope that's all it's oozing.

Best thing... when you run it, it smells like your local fast food restaurant :)

Re:Eww

[X0563511]

It's also oozing with potential patent abuse:

U.S. Patent No. 7,403,392: A portable, self-contained liquid submersion cooling system that is suitable for cooling a number of electronic devices, including cooling heat-generating components in computer systems and other systems that use electronic, heat-generating components.

How delightfully generic and self evident. Nobody has ever thought to immerse components in liquid to cool them.

Re:Eww

[PitaBred]

It's not even like people have built any oil-cooled electronic devices and sold them, either.

Re:Eww

[mollymoo]

*sigh*

You can still build an oil-cooled PC, but you might get a call from Hardcore if, for example, you include "a hard drive mechanism disposed in the interior space and submerged in the dielectric cooling liquid, and a snorkel connected to the hard drive mechanism and in communication with the exterior of the interior space to achieve pressure equilibrium between the hard drive and outside air pressure".

Re:Eww

[makomk]

Given that they don't include "a hard drive mechanism disposed in the interior space and submerged in the dielectric cooling liquid, and a snorkel connected to the hard drive mechanism and in communication with the exterior of the interior space to achieve pressure equilibrium between the hard drive and outside air pressure", I don't think you have to do that to get a call from them. That's probably an entirely speculative claim, in fact.

Remember that you only have to infringe one claim to be done for patent infringment. In this case, that means just sticking your heat-generating equipment in a tank of "dielectric cooling liquid" together with a pump that sends the liquid through a heat exchanger with fans on bolted to the outside. Hardly rocket science. (Actually, they don't even commit to fans until a later claim - any air-moving device will do).

Re:Eww

[Muad'Dave]

Cray was doing this in the early 80's with Fluorinert. Unisys has two patents that seem duplicates.

Re:Eww

[X0563511]

Great. It's bad enough to have poor/obvious patents, but we have duplicate and contradictory patents.

press release,much?

[fermion]

I try to stay objective, understanding that we will have adverts and press releases on the front page, and slashdot is ad supported, and that is becoming harder, and where would we be if slashdot were not here. But sometimes the press releases are so lame. I mean, don't people get paid good money to write a press release that the intended audience won't laugh at?

For instance, who outside infomercials aimed at the homebound uses the term 'space age materials'. People have been in space, of and on, for 50 years, two generations. Who does not use space age materials? And what are these materials anyway, plastic? Sometimes these adverts just makes me screams incompetent design, and I wonder who in their right mind would buy from these people. Probably the same people who buy the magic stock market accounts.

Re:press release,much?

[GWRedDragon]

And what are these materials anyway, plastic?

Don't forget aluminum!

Cool

[santix]

or at least it should be.

Not oil or not PCs?

[Anonymous Coward]

Speaking of oozing, weren't there some G5 Macs that leaked coolant a year or so ago? Not oil or not PCs?

Re:Not oil or not PCs?

[Starayo]

They're liquid cooled, these are immersed in the oil.

Patent for Liquid Submersion?

[RobertB-DC]

From TFA: "The U.S. Patent Office does indeed show Klum, CTO Chad Attlesey and CEO Al Berning with a patent for liquid submersion."

You gotta be kidding me. I've seen "PC in a tank of oil" on Slashdot half a dozen times. Or is it not a real technology company if they don't have at least one bogus patent on an obvious process?

rushes off to patent "Method for legitimizing a company and attracting venture capital by means of a transparently invalid patent application"

Re:Patent for Liquid Submersion?

[Anonymous Coward]

I Am Not A Patent Lawyer.. not even an Anonymous Patent Lawyer.

But here's the quip I see on here from those who are:
Read all the claims. The patent is not defined by its title, but by its claims. If the claims are too vague, open for interpretation and as a(n indirect) result covers existing inventions and/or obvious implementations - that's when you complain about a patent.

Re:Patent for Liquid Submersion?

[Kjella]

You gotta be kidding me. I've seen "PC in a tank of oil" on Slashdot half a dozen times. Or is it not a real technology company if they don't have at least one bogus patent on an obvious process?

rushes off to patent "Method for legitimizing a company and attracting venture capital by means of a transparently invalid patent application"

See, now you're getting it.
2. Attract VCs
3. Profit
4. ???

Re:Patent for Liquid Submersion?

[RobertB-DC]

Read all the claims. The patent is not defined by its title, but by its claims. If the claims are too vague, open for interpretation and as a(n indirect) result covers existing inventions and/or obvious implementations - that's when you complain about a patent.

No, silly. If I'm on Slashdot, and I have an excuse to mention patents - that's when I complain about a patent.

Re:Patent for Liquid Submersion?

[Anonymous Coward]

this is a patent accusation on the grand parent.

Re:Patent for Liquid Submersion?

[Anonymous Coward]

So, in simpler terms, you're just a karma whore.

Not everyone...

[Anonymous Coward]

Everyone's seen mods where someone super-cools a PC

Not everyone... I'm blind, you insensitive clod!

Reminds my of Kryotech.

[Chris Burke]

So my first worry is upgrade path. Which my scanning of the article's many usages of the words "custom" and "proprietary" leads me to believe there really isn't one.

Kryo's problem was that while you could buy a bad-ass refrigerated system for a mere 2x the cost of a top-end system that got a good 30% more performance -- they broke the 1 GHz barrier when air-cooled athlons were still running at around 600-700 MHz -- but then six months to a year later that system was merely "top of the line", and then of course soon after that "sub-optimal". Air-cooled athlons hit 1 GHz, and of course Kryotech came out with even faster systems, but it was obvious that the advantage you were getting was temporally speaking not worth the price.

Now with a slide-out motherboard and all it seems that upgrading this thing is at least -possible-, so perhaps if the company stays in business, you could at least purchase a compatible upgrade from them. Assuming there isn't a huge premium for the upgrade parts, that could be reasonable. The main thing is to have the re-usable oil cooling system. If they could make it so it can use off-the-shelf parts, and just sold the case itself, then that would be the ultimate to me.

Re:Reminds my of Kryotech.

[Neoprofin]

Price? Upgradability? Cost/Benifit?

You are the last person they're trying to sell to. No one who needs 3x top of the line videocards running in parallel is asking about price. This is $4000 for the base line hardware!

Re:Reminds my of Kryotech.

[Chris Burke]

You are the last person they're trying to sell to. No one who needs 3x top of the line videocards running in parallel is asking about price. This is $4000 for the base line hardware!

True, but this didn't work out too well for Kryotech in the end, and it's likely to not work too well for these folks either. Price may not be a consideration, but how long the bragging rights will last is. Even the most cost-immune geek figured out that instead of buying a Kryotech system and feel awesome for six months but then it becomes a ridiculous refrigerating paper weight, they could buy a top of the line Alienware system, and once the hardware was obsolete they could go buy a new slew of top-of-the-line hardware and still brag about their awesome case.

Re:Reminds my of Kryotech.

[Dogtanian]

Air-cooled athlons hit 1 GHz, and of course Kryotech came out with even faster systems, but it was obvious that the advantage you were getting was temporally speaking not worth the price.

Yes, but you could make the same argument for a lot of ultra-high-end computer hardware; I don't know if this is still the case, but a few years back people would pay through the nose for state-of-the-art graphics cards etc. that were sure to have plummeted in price and gone mainstream within months. Didn't stop them.

Re:Reminds my of Kryotech.

[Glonoinha]

I've been building / hacking together machines since shortly after the words IBM and AT were used together in one breath. What I've seen, over generations of building machines with the intent of upgrading them was that if you build it using a good upgradable motherboard and power supply, you might get lucky and the machine will be good for a single generation of upgrade after 12-18 months.

- Meaning maybe roughly 12 months later it's cost effective to upgrade :
- - to a same generation CPU that is quite a bit faster (think Pentium 33MHz to Pentium 133MHz, or PII-300 to PIII-900MHz)
- - the amount of memory can generally be doubled (although given the current cost of RAM and the limit of 32 bit OS, it's almost cost effective to max it out at purchase - but with 64 bit OSs I think this will still hold)
- - a second hard drive that is twice as large as the one you bought a year earlier (effectively tripling the space)
- - maybe adding a video card that is twice as fast as the first one you had in it.

About a year after you build your ultra-upgradable machine, the architecture changes (chipsets, video card platform, memory sockets, memory type, CPU sockets, hard drive interfaces) so the upgrade options taper off fast. Six months after that you can still upgrade to faster hardware at a premium price, but in very short order it's cheaper to replace the machine w/ current generation architecture than to upgrade 2 or more sub-systems. About a year after that, the machine is relegated to door-stop duty.

The problem is - the year over year increases are so steep that this happens no matter where on the curve you are when you buy. Think back - the premium for the DX2-66 was about $400 over the DX2-50, the premium for 16M of 72-pin SIMM memory was about $800 over 4M of 72-pin SIMM memory - but two years later it really didn't matter because with the P2-300MHz machines coming with 64M of PC100 memory - your two year old box wasn't fast enough regardless of whether you payed the $1,200 premium to get those two bumps or not. You could always pay another $1,200 premium for that next machine and get the P2-500MHz w/ 192M of memory, but two years later when the P4 based machines running 1.8GHz and 512M of memory - it really didn't matter whether you splurged for the extra umph or not - the box needed to be upgraded.

Is $11k out of line for a machine that's going to be on-par with the next generation of hardware, and obsolete in three years regardless? Probably. Unless you have a business reason for it, I'd say yes. I said the same thing about LCDs when IBM was selling their 16" LCD for $1,600, and SSDs were $100 per Gig. Today - both are reasonably priced, and maybe these guys will pioneer the path towards the next revolution in hardware platforms. Have to admit this much - of the $11k, how much is actual material cost and how much is 'OMFG 1337 haxor' premium? That's the margin that these machines will come down in about three years, making the platform affordable (or not.)

In my opinion there are two places that this machine makes sense - high end CAD where a company is paying $30k per seat to license the software and $125k per year for the guy behind the keyboard (25% faster machine = 1/4 fewer seats) or working on hard-duration projects (finish designing the next Space Shuttle by June 1st and get a $5M bonus.) Other than that - and the obvious rich gamer - I'd say a given day's task set would be better served by a couple three or four desktops all coming through to a single multi-LCD machine that rdesktop's to all the others, allowing the user 1 machine per LCD and the ability to mouse from machine to machine and control all of them from a single keyboard / mouse. RAIC - redundant array of inexpensive computers - it gets normal multi-tasking 'work' done faster.

Re:Reminds my of Kryotech.

[mr_mischief]

Where the hell did you find a Pentium 33Mhz?

Re:Reminds my of Kryotech.

[Glonoinha]

Sorry - brain no worky today.
486DX-33MHz vs 486DX/2-66MHz.
The Pentium example was going to be the 60MHz vs 66MHz (and yea, those extra 6MHz were pretty damn expensive on the first generation boxes.)
Merge the two, add a glass of scotch and what I wrote is what you read.

Re:Reminds my of Kryotech.

[Neoprofin]

That's why this is purely for the "I have too much money and I want you to know it" crowd, it's a small market but they've got deep pockets.

The problem is it only work if people continue to buy them, which means they have to keep updating their hardware. Hopefully outfitting the blocks for cards isn't too different from revision to revision or they'll destroy their margins just designing and fabricating a new set of parts every few months.

Re:Reminds my of Kryotech.

[Whiteox]

you might get lucky and the machine will be good for a single generation of upgrade after 12-18 months.

On one hand I agree with you, but boxes built around 2004 or even earlier, I'm still maintaining some good machines that are still powerful enough to use today for 99% of apps.
A 2.6 - 3.2GHZ P4 with 400MHZ ram works fine. A 2.6GHZ machine with 2 gig DDR ram I built in 2003 with a hyperthread capability is still a great workhorse.
Now earlier than that, you are absolutely right. Every 12-18 months there was something bigger and better. When Intel decided to go with their D class Pentiums, I thought that was absolute shit. Now with bog standard dual core machines @ 1.6GHZ up to a whopping 3.2, I can't see any radical improvement unless the software can use multicore processing.
It's the FSB and ram speed that is pushing the generation envelope and not so much the processors. The change from 32 bit to 64 bit OS will be pushed (I reckon Windows 7) and you'll see a move there.
In the meantime, ex-lease P4s for around $100 each is pretty sweet! At least for another 12 months!!

Propietary oil

[gmuslera]

Snakes(oil) on a Mobo. What would say Samuel L. Jackson about it?

Re:Propietary oil

[greyhueofdoubt]

Get this mothafuckin' snake oil off this mothafuckin' mothaboard!

-b

Forget how much it costs.

[MaxwellEdison]

All I want to know is can I mate with it.

Okay just kidding...kind of. Every other glowing clear-sided tower I've ever repaired I have been repulsed by. They always remind me too much of those arse-ugly Honda street tuners. This styling is to a traditional 'custom' rig what a Lotus Elise is to those Hondas.

But...on the subject of repairing them...the propietary cards worry me. Of course at $4k-$11k I'd expect a little more than Dell's horrible warranty service.

P.S. If you play Crysis on it at max settings, can you make french fries?!?!?!?!?!

Re:Forget how much it costs.

[Anonymous Coward]

Lotus Elise is now elegant? It's the most confused car design I've seen in a long time. Fugly, if you ask me. (So don't, heh.)

Not just proprietary video cards, but the entire motherboard is custom (by Tyan apprently) as well. And pretty frigging expensive for a puny single-socket system...

Why bother with such "high end" stuff?

[diamondsw]

System is worth $11k today, maybe $2k by 2010. Super high-end systems that are not designed for professionals (or servers for datacenters) just have never made sense to me. The depreciation is just too great on a computer.

Not to mention it will be worth $0 when the oil containment fails.

Re:Why bother with such "high end" stuff?

[linzeal]

This would make a terrible workstation the cards are absolutely not made for that.

Re:Why bother with such "high end" stuff?

[afidel]

That explanation is SO stupid. The Quadro line is the same chips just binned and jumpered differently which is why the hard and soft-mods work, it's the same freaking hardware with a different driver. If you have a marginal card you might get some problems when exercising units that games don't normally push, but anything past a first rev of a new chip is likely to be just as good as the Quadro binning of the first rev.

Re:Why bother with such "high end" stuff?

[linzeal]

If you are using a video card for CAD, game design, modeling or stereo vision the likelihood of you softmodding or hardmodding the video card is directly proportional to how much you are actually doing because your time should be worth much more than paying the premium for the better supported FireGl and Quadro lines. Yeah, students might do that but no one in the industry would; it doesn't make sense, as most consumer cards have half of the memory of the workstation cards and if you have ever worked with assemblies of 1000's of parts in Inventor or ran stereo vision to view architectural models with gigs upon gigs of textures.

The low end workstation cards come with 1 gig nowadays and came with 512 megs 2 years ago and are much more reasonably priced if you compare them to ultra high end consumer cards set in SLI mode as CAD still does not work for shit in SLI as it is windowed. Even if you went out and bought the latest and greatest video card and softmodded it doesn't mean it will work with all programs or even will work with the same program you have been using it for if you update the driver unless you are savvy enough to be able to wade through the layers of obfuscated calls to the driver libraries that compromise a softmod. In other words if you are not playing pretend at being a mechanical engineer or game designer you pry aren't going to fret about spending 1000 dollars on a decent, well-supported video card.

Crisco oil party

[snspdaarf]

Ignoring for the moment what will happen when you tell someone your plans for the weekend are to oil up your new computer.....

I know that air cooling has its limitations, but there is something nice about knowing your computer is not going to spooge all over your desk or floor. And working on this thing will be a real nasty mess.

But if you are an alternate fuels kind of person, then your diesel Smart and your PC can both smell like french fries.

Seriously, if they have to ship this thing empty, how to you return it for warranty work? I promise you that UPS or FedEx will not be happy if your package begins oozing oil in their truck. This may be the best thing to hit town since, oh, Orgasm Queen of the SS (Godwin and porn in one post!), but I will wait a few years before I will buy an oil cooled PC.

Re:Crisco oil party

[frank_adrian314159]

Orgasm Queen of the SS

I am interested in your ideas and would like to subscribe to your newsletter...

Re:Crisco oil party

[Anonymous Coward]

My water cooled system whizzed itself to death, so I went back to (mostly) passive aircooling. It's cheaper, simpler, robust and at about as quiet. Oh, and the temperatures under load are only a couple of degree's higher (with the right heat sinks, obviously).

Remind me again why people want to screw around with liquid cooling...?

Re:Crisco oil party

[greyhueofdoubt]

>> Orgasm Queen of the SS

The PC is apparently submerged in Kroil- The oil that creeps (you out)

-b

Re:Crisco oil party

[GWRedDragon]

Seriously, if they have to ship this thing empty, how to you return it for warranty work? I promise you that UPS or FedEx will not be happy if your package begins oozing oil in their truck.

Um. Empty it first?

Re:Crisco oil party

[snspdaarf]

Mod you funny. It's oil. If you empty the case, what's left is not going to dry. There is going to be oil clinging to everything. What do you do then, take it to a car wash? Run it through the dishwasher? I don't know if the anti-static bags we use to ship motherboards are oil resistant, but I know they are not liquid tight. This is going to be a nasty mess to ship after the first fill.

Re:Crisco oil party

[I.M.O.G.]

I am interested in your ideas and would like to subscribe to your newsletter...

You muffed the meme. It's a simpsons quote parroted so much on slashdot that few recognize where the meme originates. Here is the correct form:

"Your ideas are intriguing to me and I wish to subscribe to your newsletter."

Source

Re:Crisco oil party

[GWRedDragon]

That's true, but I bet if you leave it to drain for a day or so it will drain sufficiently to allow shipping without oil coming out of the box all over the place. Pack it with some paper towels! :D

Re:Crisco oil party

[Whiteox]

Orgasm Queen of the SS

Have you got a copy? Can you torrent it for me pls???

Re:Crisco oil party

[Anonymous Coward]

Sigh. It's so pathetic to watch you geeks nutting over an oil-filled PC, how to clean it, and what not. Did it ever occur to you that this idea has been in use by some sectors longer than you have been alive? A simple spraying of brake cleaner on the oil-soaked board, followed by a hot-water rinse and then blowing dry with compressed air will clean things up nicely.

Re:Crisco oil party

[hesiod]

My PC is submerged in Krelm, with the miracle ingredient 345-4900 (347-3816 after 6pm).

if you buy an oil immersion pc

[circletimessquare]

your money goes to support vladimir putin, hugo chavez and saudi wahabbism

do the patriotic thing and buy a domestic american wind immersion pc

and if you aren't in an area where a wind immersion pc is practical, follow t boone pickens' lead and get yourself a natural gas immersed pc

eh... on second thought, maybe not such a good idea, a natural gas immersed pc, heh

Re:if you buy an oil immersion pc

[greyhueofdoubt]

An LP tank would be a great place to submerge a pc as long as you let off enough pressure to keep everything cool. However, your gas bill would be outrageous.

For extra credit, design a closed system that keeps the pc in the cold (expansion) side of a freon/refrigerant air conditioner. No condensation, no corrosion. Perfect.

-b

Re:if you buy an oil immersion pc

[oldhack]

do the patriotic thing and buy a domestic american wind immersion pc

Don't fall for that. That's a Taiwanese astroturfer. There is no domestic american wind immersion pc.

Re:if you buy an oil immersion pc

[TD-Linux]

Sssh! These are manufactured in my home town of Rochester, Minnesota - I don't want you blabbing all our secrets across the face of the internet :)

Re:if you buy an oil immersion pc

[Anonymous Coward]

naa I am all for recycling, any body knows if pee is conductive?

Space-Age???

[mamono]

I've always wondered why this term is ever supposed to mean anything. I mean, wasn't the "Space Age" in the 50's and 60's? Are they trying to tell you that their product is using 5o year old technology?

Re:Space-Age???

[rossdee]

The Space Age started on Oct 4, 1957

It hasn't ended yet. We (USA, Russia) still have people in orbit, and other countries (China, India) are getting involved.

And..... why?

[Ancient_Hacker]

I read TFA and was waiting for all the reasons this was a good idea. Like why all Buicks have three holes in each side. Or why glad bags are so much better with yellow and blue zippers. I expected to read that the machine was like, totally silent. Or that things ran, well, slicker. Maybe I missed a whole page of pluses? There gotta be a whole lot of pluses for a $4K box that you can't change the motherboard.

Here's their patent claim

[Animats]

Here's the main claim from the patent:
7,414,845 Attlesey, et al. August 19, 2008
Circuit board assembly for a liquid submersion cooled electronic device

1. A liquid submersion cooled computer, comprising:

• a case having a liquid-tight interior space;
• a lid removably connected to the case for closing a top of the interior space,
• the lid including at least one pass-through connector;
• a motherboard disposed in the interior space and attached to the lid,
• the motherboard having a top end with electrical contacts engaged with the pass-through connector that permits inputs/outputs and/or power to be passed to the motherboard;
• a plurality of components mounted on the motherboard, including one or more of a plurality of processors,
• a plurality of memory cards,
• a plurality of graphics cards, and a plurality of power supplies;
• a dielectric cooling liquid within the interior space and submerging at least one of the components on the motherboard so as to be in direct contact therewith;
• and an impingement cooling system that includes a plurality of tubes for directing a flow of the dielectric cooling liquid directly onto two or more of the plurality of components.

The only novelty here seems to be in putting the connectors in the removable lid.

Incidentally, the cooling liquid isn't an "oil" at all. It's one of 3M's Novec engineered fluids, probably HFE-7500, which is 3-ethoxy-1,1,1,2,3,4,4,5,5,6,6,6-dodecafluoro-2-trifluoromethyl-hexane. It's usable for cooling up to 150C, nonflammable, does not irritate skin, does not contribute to global warming, ozone depletion, or smog, and the MSDS even says "Ingestion: no health effects are expected". 3M developed it as a replacement for PCBs and perfluorocarbons like Fluorinert. So it can be used safely by the idiots who overclock.

Re:Here's their patent claim

[NormalVisual]

So it can be used safely by the idiots who overclock.

So could deionized water, for less money. I totally agree that there's really nothing new here, but I guess everyone's happy as long as the investors don't know that. :-)

Re:Here's their patent claim

[Animats]

So could deionized water, for less money.

Zinc, copper, and mild steel, all of which are likely to be found on a circuit board, will all corrode slowly in deionized water. You don't get mineral deposits from deionized water, and it's nonconductive, but it's not noncorrosive.

Re:Here's their patent claim

[Anonymous Coward]

Deionized water would work right up until the point that some sort of salt is introduced to the tank. It doesn't take much salt to make a 4 gallon tank of deionized water conduct electricity. There's probably enough residue on the motherboard to short out the whole thing.

Re:Here's their patent claim

[NormalVisual]

True - I hadn't considered the corrosive nature of the water. Where I've used it (immersive cooling for lasers), it's not a problem, but you're right that there are some metals on a circuit board that might not be happy in it over the long term.

Re:Here's their patent claim

[dpiven]

Incidentally, the cooling liquid isn't an "oil" at all. It's one of 3M's Novec engineered fluids, probably HFE-7500, which is 3-ethoxy-1,1,1,2,3,4,4,5,5,6,6,6-dodecafluoro-2-trifluoromethyl-hexane. It's usable for cooling up to 150C, nonflammable, does not irritate skin, does not contribute to global warming, ozone depletion, or smog, and the MSDS even says "Ingestion: no health effects are expected". 3M developed it as a replacement for PCBs and perfluorocarbons like Fluorinert. So it can be used safely by the idiots who overclock.

Well, what the hell good is it if you can't fry doughnuts (today's captcha: wholly) in the fluid reservoir? Not only are you not getting any high-calorie, high-transfat chocolate-frostable ambrosia, it also implies that you can't overclock your machine worth a damn.

Re:Here's their patent claim

[NormalVisual]

That's why in any DI-chilled system, you run it for a bit to let the filter pull the crap out of the water before applying any power anywhere with an interlock that kills power if the water gets too conductive.

Re:Here's their patent claim

[greyhueofdoubt]

People who have tried using DI water found that after only a few minutes of water, the DI was no long- Metal ions were carried into the water very quickly (especially off charged surfaces) and then the resistance of the water dropped geometrically.

This was total immersion, not just a water-cooled cpu block. Hats off to them for risking their system. I looked but could not find the original source.

I did find this, and since this is slashdot I'll just go ahead and post the link:

THE COOLING POWER OF PIGEON LEGS
http://jeb.biologists.org/cgi/reprint/136/1/193.pdf

-b

Re:Here's their patent claim

[greyhueofdoubt]

>>probably HFE-7500

I don't know. I thought the HFE family ate through the kinds of plastics one would likely find on a computer. Also, it evaporates relatively quickly (quicker than hydrocarbon oils, at least).

I suppose HFE would explain the better part of the price tag.

-b

Re:Here's their patent claim

[NormalVisual]

That's why DI chiller systems are always recirculated and have a filter to pull anything out that might make the water conductive.

Re:Here's their patent claim

[ari_j]

Yeah, but what happens when they try to huff it?

Re:Here's their patent claim

[gworkhorse]

No, it's not 3M, but it is synthetically engineered. The MSDS and technical documentation will be available soon on the website. A Hydro-carbon base similar to mineral oil used in tranformers (not vegetable oil) so it won't be smelling like french fries, unfortunate for those fish&chip fans from Scotland & England! It's actually quite a great fluid, adheres to stringent IEC/UL/CSA IT equipment flame points (>150C) with very good thermo absorbant capacities to cool completely, keeping a very stabile ambient.

Versus water cooling

[Lost Engineer]

I'm just wondering what the advantage of this thing is versus non-submersed liquid cooling.

1) Noise? They said it's quiet "for so much hardware." Yeah well considering what a normal three GPU system sounds like that's not saying a whole lot. A good pump-and-block cooled setup can run around 25 dBa which is something like a whisper at 6 feet. In both cases you still need a couple of fans running, so I imagine non-cooling factors will dominate noise.

2) Performance? The article says cooling probably won't exceed the best liquid cooled setups that focus on the CPU/GPU.

It certainly is a cool idea, but I think I'd rather pay for a normal liquid cooled setup.

Re:Versus water cooling

[NormalVisual]

The advantage is that it cools all the components in the system, not just those that have a water block or heat sink on them, so I'd imagine the components will last longer. Not that it really matters, since the system will almost certainly have been replaced well before the components start failing, not to mention that for the same price premium you could probably buy a closet full of spare parts.

Re:Versus water cooling

[mr_mischief]

Or a regular computer and two motor scooters.

Re:Versus water cooling

[Lost Engineer]

This thing would be a lot cooler if it had NO (air) fans whatsoever, but I imagine the problem will be the same as I've had trying to build pump-and-tube systems with no fan: something has to cool the radiator. They mentioned something about a peltier cooler in the future. I hope they find a way to make that work.

can I have...

[Anonymous Coward]

fries with that?
thanks.

Equilibrium

[EZLeeAmused]

From the article:
the machine should never really run higher than ambient room temperature if all is well
How long after you switch it on does the ambient room temperature make it up to the operating temperature of the pc?

Re:Equilibrium

[greyhueofdoubt]

I'm pretty sure the temperature difference between the room and the computer would look like the graph of y=-logx where y is the temp diff and x is time.
You'd need to know the starting conditions.

But I've had some beer and I'm tired. Forget you read this.

-b

When they can do $1-4k without being a knockoff

[sethstorm]

Then I'll be impressed - provided that it's of the same size/similar functionality. Cut the SSD's in favor of less write-limited drives (or give plenty of eSATA ports, and cut back on SSD's). As long as said machine wouldn't end up being a crippled/significantly slower machine with cut-rate parts(read: Realtek, VIA and similar manufacturers), it would move a lot faster.

I'd hope that with that kind of costs they have already, that a good chunk of the parts(including electronic ones) were US made.

Re:When they can do $1-4k without being a knockoff

[Eskarel]

It's been discussed before, but SSD's are not particularly write limited. My recollection is that writing the entire volume of the drive per day you'd still end up with a drive lasting 30 years(presuming even usage), and since realistically the components won't last that long you could get substantially higher writes with little to no problems(presuming the usually best case 5 years for consumer hard drives that's 6 times the volume of the disk every day which is a hell of a lot of data.

They also use SSD's because they aren't mechanical and can be immersed, allowing you to actually cool your hard drives, which is sort of the point.

I'm not sure whether the time is right for this technology, or even whether it's a particularly good idea, but there's a good reason for the SSD's.

Pointless waste of money

[Conspicuous Coward]

First off, the summary reads like a press release, as does TFA, is slashdot that desperate for cash these days? Secondly, the PC itself seems like a pretty useless gimmick.

I don't understand who is supposed to be buying this thing at $4k-$11k.
Hardcore overclockers? OK the thing has excellent cooling, but not much better than you could achieve with a decent watercooling rig at a fraction of the price. This group will be put off by the proprietry(and probably overpriced)upgrades and the difficulty of actually opening the thing, not to mention the pricetag.
Gamers? Why would they pay this much over the odds for a system that's at best 10% faster than a commodity system? Again, this group will be put off by the lack of a decent upgrade path.
Silent PC enthusiasts? This group might be interested at first, the one thing an oil filled PC might arguably be useful for is silence. But at $4000+, you've got to be joking, there are already very good solutions at a fraction of that price.

Ultimately I just don't see any need for this kind of cooling system, PC's just don't run hot enough that it's worth dealing with the hassle.

Re:Pointless waste of money

[toiletsalmon]

Thanks. That was perfect:

First off, the summary reads like a press release, as does TFA, is Slashdot that desperate for cash these days? Secondly, the $ITEM itself seems like a pretty useless gimmick.

I don't understand who is supposed to be buying this thing at $LOWPRICE-$HIPRICE.
Hardcore $HOBBY? OK the thing has excellent $FEATURE, but not much better than you could achieve with a decent $COMPETE_TECH $ITEM at a fraction of the price. This group will be put off by the proprietry(and probably overpriced)upgrades and the difficulty of actually $MAINTENANCE the thing, not to mention the pricetag.
$HOBBY2? Why would they pay this much over the odds for a $ITEM that's at best $PERCENT_BETTER% $VERB than a commodity $ITEM? Again, this group will be put off by the lack of a decent $ALTERNATIVE.
$HOBBY3? This group might be interested at first, the one thing an $ITEM might arguably be useful for is $FEATURE. But at $LOWPRICE+, you've got to be joking, there are already very good solutions at a fraction of that price.

Ultimately I just don't see any need for this kind of $ITEM, $ITEM's just don't $BEHAVIOR that it's worth dealing with the hassle. ;)

Re:Pointless waste of money

[iteyoidar]

Gamers? Why would they pay this much over the odds for a system that's at best 10% faster than a commodity system? Again, this group will be put off by the lack of a decent upgrade path.


PC Gamers will buy a lot of stupid things. The people who will buy this thing are the sames ones who are right now buying Alienwares with liquid cooled Nvidia 8600's in SLI configuration (which is even more bizarre than an oil cooled PC in my opinion).

Re:Pointless waste of money

[Whiteox]

buying Alienwares with liquid cooled Nvidia 8600's in SLI configuration (which is even more bizarre ...

Exactly. Anything to get frame rates higher than 15 in Flight Sim 10!

Fallacy: oil based PCs don't break down over time

[NotInfinitumLabs]

It was thought that the mineral oil on these PCs would eat at the Rubber seals or the contacts on the motherboard and cause the PC to fail over time. This isn't true. Here's a link to the year-after report on a oil-based PC that Puget Systems built: http://www.pugetsystems.com/submerged.php#update3

The most important part:
"# There is no sign of weakening of rubber seals or PCB. We have found that prolonged exposure to mineral oil does not eat away at any components. However, you will notice in the pictures that the voltage module for the LED light has fallen down. That module was stuck in place with nothing more than a sticker -- it took 9 months for it to come down! We're amazed it stayed up that long, but definitely recommend you do not rely on stickers or tape to fasten anything. Zip ties will be more solid and long lasting."

Quality.

[sethstorm]

There are large and non-ignorable sizes of the population that want quality and do not get it. However, they've yet to figure out how to make it obtainable by mere mortals without using cut-rate brands or allowing knockoffs in the process. That is when this kind of machine would be impressive - quality without exorbitance.

With regards to quality and not price:
Cheapness is weakness

Open-source the oil!

[GFree678]

proprietary oil

Can we demand an open-source version of the oil for us hardcore geeks?

Re:And..... why?

[forceman130]

The Buick thing is easy - it tells you what size engine is inside. The V6's have three holes on each side, and the V8's have four. And yellow and blue make green, so you can see when it is sealed. I thought that was well covered by the "yellow and blue make green" ad campaign.

Oh yeah? I'm just going to rush out and patent...

[Anonymous Coward]

...air cooled computers.
Nyah, nyah na nyah nyah!

Foreign oil

[philspear]

Great! Now we're going to be dependant on foreign oil not only to run our cars, but now to run our COMPUTERS?!? I'm so disgusted, I didn't even read the summary. Our addiction has reached new heights. What's next, making PLASTIC out of oil? Sheesh.

Re: My only question...

[Linker3000]

No, it runs SL/IX and the main processor is made by Texaco Instruments.

There is a shell scripting language but the interface is a bit crude.

There will be a Mobil Computing version next year.

I'll stop now.

Re:And..... why? - glad bags

[onkelonkel]

Glad bags are awesome. Those ones with the fold over top...did you know that you can put sandwiches in them!

Re:Fallacy: oil based PCs don't break down over ti

[NormalVisual]

Or you could avoid the possibility of damaging any rubber/plastic components by immersing in DI water rather than oil. Fill the system up, bang it around a bit to dislodge any bubbles, run the pump for half an hour or so to let the DI filter get anything conductive out of the water, then you're good to go. I think I'd rather deal with a leak of a couple of gallons of DI water rather than some weird mineral oil that will likely ruin anything it touches. You'd have to change out the DI cartridge every year or so, but I think overall the system would be easier (and cheaper) to work with.

Actually interviewed here

[Anonymous Coward]

So, I actually interviewed here prior to the product launch, and got to see this (under NDA, of course). The motherboard is SUPER proprietary, a joint venture between Hardcore Computer and Tyan (which makes pretty decent equipment, IMHO). Of course, it does still take standard CPUs/RAM. But just in case there are any doubters out there, I've been to the facility, seen the board-level techs they have on staff, and played with, first-hand, one of these suckers. They've put incredible R&D into this (took them a couple of years to finally get to launch). I'm not an employee, just someone who's seen these and was pretty impressed. Also feeling vindicated, since I knew from the second I saw it that this would end up on Slashdot :-)

Re:Actually interviewed here

[BOFHelsinki]

Uh, the reception seems to focus on the proprietary hardware and the resulting disadvantage with upgrading, the foolishness of RAIDing low seek-time SSDs, and the bogus patent claim.

Are you sure you wanted it to end up on Slashdot? ;-)

(I'm sure the company is legit and sincere, but there's a hefty Fool's Gold factor in that system.)

Whoa thats cheap!!

[Anonymous Coward]

Wow, for 4000 dollars, i'll buy the PC just to dump the oil in my car

At last, a true oil rig

[Ed Avis]

Whenever you read these overclocking and gaming sites it gets really tedious that they always have to call a computer a 'rig'. But finally, we have an instance where the name is entirely appropriate.

Re:And..... why?

[lymond01]

Or that things ran, well, slicker.

I think you can assume this one at least.

Re: Where the hell ...

[Anonymous Coward]

Kansas. The Wizard of Oz retired and began selling hardware.

Great!

[reboot246]

Now when somebody tells you to shove that computer up your ass, it'll have plenty of lube to go in smoothly.

Oil is nuts, What we had was so much better.

[John Sokol]

This is from a now dead startup I did from 2002 to 2005. 100% totally silent High end PC's.

http://www.silentcomputing.com/i.html Look at the last photo.

  We had an all aluminum design as well as a water cooled design. I also came up with an advanced Carbon Fiber material with 4x the thermal conductivity of Copper that was light weight.

These systems provide much more cooling then oil could ever do.

The system was sealed, 100% total silent and easy to disassemble and re-assemble. Even easier to work on then a regular PC...

We even had the hard drives in a thermally conductive rubber allowing them to run cooler then in a normal system with fans.

The main system was sealed and designed to run with a descant and dry gas like argon inside so when overclocking using peltier thermoelectric coolers on the CPU where wouldn't be any condensation. Condensation is a major problem for overclockers that up the voltage and have to supercool the CPU.

We are still planning to open source the designs.

We never were able to raise the money to start production of these. To be honest 1/2 our problem was management wanted to court Intel,HP, SUN, and they just didn't get it. The large companies really weren't interested in something that didn't conform to what they already were doing.

We never did talk to enough small investors, and finally we ended up with a bad apple in the company that try to do a hostile take over and killed the company when he failed...

Finally the last 20 polished heat sinks I had were stolen out of my garage 2 weeks ago :( Some idiots problem going to get standard aluminum recycling prices for them too, considering each one cost me over $100 each!

I always felt doing oil was just idiotic and still do.

Too bad being an entrepreneur isn't as easy as programming.

If anyone is still interested in this tech, let me know. I have 3 years invested in it, and we were partners with NASA for much of it.

Re:Oil is nuts, What we had was so much better.

[Anonymous Coward]

Probably the other half of your failure was because you don't know the difference between 'than' and 'then'.
Investors, partners, customers, posters on Slashdot; many people worry about the correlation between literacy and ability.

Re:Oil is nuts, What we had was so much better.

[ivan256]

That looks slick.

Who would you have sold this stuff to other than Intel, HP, Sun, or Dell though? If you wanted to make any money at all (and if you were venture funded you'd have to make a *lot* to break even), those are your only choices. Once all of those guys shoot you down it's time to go out of business.

There's a slim chance that you could have sold direct to datacenters. The chance of success, though, would have been low. The datacenter would have to be designed around your product at a massive investment. Who in their right mind would invest in a datacenter that relied on technology from a startup that might not be around in a few years? The only way that's ever happening is if your product is non-critical. Cooling is critical.

Re:Oil is nuts, What we had was so much better.

[John Sokol]

Well how is this oil cooled company able to come out with their product? It can't be much cheaper then what I was doing?

Re:Oil is nuts, What we had was so much better.

[ivan256]

There's a big difference between selling a novelty item to consumers, and selling infrastructure to the enterprise.

These oil-filled PCs will be sold to individuals as a disposable novelty. Individuals are willing to take that sort of risk on a random company, for many reasons. One is that if the company goes out of business after they make their purchase they're out nothing; they still have their device. The other is that they don't expect the bulk discounts, uniformity, and support infrastructure that enterprise customers demand.

Your company almost certainly had several hurdles to overcome that this company does not. First: you made a cooling solution, but your customer base buys fully integrated systems. It is practically impossible for a startup to get traction with a product that changes multiple aspects of how the target market does business. You wanted to change how they cooled their systems, but if they bought your products independent of the rest of their systems they would have had to change how their cooling infrastructure worked *and* how they bought systems. Second: Your product would have impact on the infrastructure of your customer's datacenters. Datacenters are willing to make infrastructure changes for a product with sufficient benefit, but only if that product comes from a big company that is unlikely to go out of business before they achieve returns on that investment. That is why success would have been unlikely unless you sold to, or partnered with one of the big guys.

The biggest thing I'd point out in this context, though, is that the people making this oil cooled computer aren't necessarily going to succeed either.

Re:Oil is nuts, What we had was so much better.

[John Sokol]

Originally that was my plan, to sell this to individuals as a novelty if you will, silent computers. Since I do video, having a silent PC is important when recording audio in the same room as your PC is located.

(I was inventor & tech)
But some of the people came in as management went for enterprise too soon, and I knew it wasn't a good fit at that point, while we were pre-funded, and no revenue, and unable to put the kind of money in to making those kinds of sales.

True they may not succeed, but what is frustrating is that they have still made it further then we did, in terms of getting a product on the market. I guess was undermined early on when the partners I brought in didn't see money in a novelty product, but time has shown that Alien Ware and other companies with stylish cases can do quite well.
I still think it would sell although most companies have managed to quite there high end systems considerably, in 2002 the 3.2 Ghz P4 were up at the 75dB noise level, so it was really a good time for such a product, the idea being that once that market started to dry up we'd be in a much better position to create products for the enterprise market.

space age

[benicillin]

im suspicious of anything that has 'space-age materials' ....

Crysis?

[Anonymous Coward]

But can it play Crysis?

Re:And..... why?

[pipingguy]

I'm colour-blind and innumerate, you insensitive clod!

Bad SSL Cert

[Flagbrew]

What kind of company selling 11,000 dollar computers wouldn't be able to use a proper SSL certificate? The hardcorereactor.com checkout uses hardcorecomputer.com SSL certificate.

Re:Bad SSL Cert

[afidel]

HAHAHA, you're right their webmaster is an idiot, probably used IE6 to design the site and never tested in IE7 or FF.

Patents mentioned:

[nog_lorp]

Liquid submersion cooling system:
http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&u=%2Fnetahtml%2FPTO%2Fsearch-adv.htm&r=2&p=1&f=G&l=50&d=PTXT&S1=Attlesey.INNM.&OS=in/(Attlesey)&RS=IN/Attlesey

Circuit board assembly for a liquid submersion cooled electronic device:
http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&u=%2Fnetahtml%2FPTO%2Fsearch-adv.htm&r=1&p=1&f=G&l=50&d=PTXT&S1=Attlesey.INNM.&OS=in/(Attlesey)&RS=IN/Attlesey

From the first, a notable claim that relates to their "super high-tech secret oil":
8. The electronic device of claim 1, wherein the dielectric cooling liquid is a soy-based dielectric liquid.

Re:Patents mentioned:

[ymail.com]

These oil-cooled units come out of northern Minnesota, where water cooling isn't feasible ten months out of the year.

There's more than one kind of "quality"

[RustinHWright]

I know that this will disgust most folks here but afaic, the best market for these will be for receptionists and folks like that at high end customer service applications. If you're looking to build a reception desk for a boutique hotel then three or four of these in an open frame aluminum enclosure puts you well on your way to looking cool and rich for your snotty customers. In an application like that they don't even have to work very well. Back in the early nineties a company sold insanely expensive custom paint jobs for tower Macintosh CPUs and monitors and my friends and I all looked at them at the trade shows, drooled, and told each other that nobody would buy them. We were wrong. To a SoHo art gallery a thousand dollar cpu paintjob was cheap at the price and I can tell you that I kept spotting those puppies, some of them not even turned on, at various front desks at ad agencies, snotty law firms, and the like for at least seven or eight years after that.

Mark my words, these will be bought for high-profile uses at these kinds of places and almost certainly will get featured in at least one televison show or movie, probably several. All they need to do is a seed a few at Rodeo Drive stores or the offices of some casting agency and they'll be in like Flynn.

Re:There's more than one kind of "quality"

[Eivind]

Why would anyone paint a CPU ? And just what is a Macintosh-CPU anyways ?

Hint; the largish box connected to the monitor CONTAINS the cpu, and about a dozen other components...

Re:There's more than one kind of "quality"

[Anonymous Coward]

That's okay, my mother calls the entire case the hard drive. It only becomes a computer to her when there is a monitor and keyboard attached.

Re:There's more than one kind of "quality"

[hesiod]

A couple people I work with, who use computers regularly call it a modem.

Re: My only question...

[dpiven]

Not if you're running SL/IX, you won't.

Re: My only question...

[Inner_Child]

Shouldn't that have been "a shale scripting language"?

Organic Material in a Silicon World

[Anonymous Coward]

From a Biological stand point.... If I'm not mistaken, aren't there Microorganisms that Thrive on Oils!?!? 0.o

Re:Fallacy: oil based PCs don't break down over ti

[afidel]

DI water tends to not stay that way in the presence of metals, and as soon as you have re-ionized water you have a big mess on your hands. Not to mention that water+Cap's=bad idea whereas cap's+oil=no biggie.

I'm thinking bar fridge or small freezer

[gelfling]

Seriously, at this point wouldn't it make more sense to build a moisture-proof wrapper for your PC and just dump it in a bar fridge or small portable freezer? None of these systems are actually active cooling units, they're just more efficient ways to dissipate the heat that builds up. So why not actively COOL the systems down? How hard could it be to build a reasonably reliable freezer that stays at 0Deg F & that's big enough to hold a minitower? Can't possibly cost more than a $1000. Add a moisture-proof wrapper for the PC (a.k.a. heatshrink plastic sheet) and run all the cables and cords though a 1" hole drilled in the side an plugged up with foam and silicone.

Re:I'm thinking bar fridge or small freezer

[NoisySplatter]

Why bother with a moisture proof wrapper when you could just dehumidify the freezer itself and keep it sealed tight? Honestly though there are plenty of refrigerant type phase change cooling products on the market.

Oh crap...

[Terrorwrist]

I have to take my pc AND my car to Meineke to get an oil change. Im gonna be poor soon.

Re: My only question...

[beav007]

See shell. On a see sure. Or something...

Re:Fallacy: oil based PCs don't break down over ti

[greyhueofdoubt]

Second. Deionized water next to charged metal surfaces= ionized water. DI water is only good in labwork where you are using glassware.

Solid helium. That's what would work best. Except, I suppose, for the explosion. Hey, you don't call yourself hardcore unless you're willing to have a smoking, -260C crater in your mom's basement.

-b

Same thing done in the early 1900s for ...

[the_rajah]

elevator controls. I know where there's a 3 stop basement traction elevator in a private home that was installed in 1917 and still in operation with all stock components where the control relays are immersed in a tub of oil. The relays are mounted to the lid which can be raised up via a small chain hoist.

Re:Same thing done in the early 1900s for ...

[Cosmic AC]

Power transformers have been oil cooled since the 19th century.

Re:Fallacy: oil based PCs don't break down over ti

[krotkruton]

When I first read the article (admittedly, a bit late), I thought the same thing. I'm just disappointed I had to read this far down to see someone who pointed this out. Wish I had mod points.

Why the whole motherboard?

[jtgd]

It seems to me a better design would be to mount the CPU(s), and anything else real hot, on the bottom edge of the motherboard. Then you only have to immerse the bottom portion of the board in a pool of oil. The upper portion, where the PCI boards plug in, can be conventional, thus not requiring special boards.

Re:Fallacy: oil based PCs don't break down over ti

[NormalVisual]

Deionized water next to charged metal surfaces= ionized water. DI water is only good in labwork where you are using glassware.

That's why you have deionization filters in the circuit, and filtered water works wonderfully as a coolant in laser systems, even lamp-pumped systems where the electrodes are immersed and are passing 15 or so amps at better than 100 VDC. One thing that I did miss and that another poster pointed out, however, was that the corrosive nature of the water will not be particularly good for the leads and traces on the board. Mea culpa.

Mainstream??

[DarkHorseman]

Since when is a $4,000 Computer Mainstream? I consider something to be mainstream not only when it is affordable to the average consumer, but a logical and necessary choice.

I see a few problems

[PPH]

Problem 1: Fire codes. To date, nobody has come up with a non-flammable insulating oil to replace PCBs (carcinogenic, nasty stuff). Oil is flammable and, in combination with electrical equipment, a very bad idea. Generally, it is not allowed within occupied ares outside of fireproof vaults.

Problem 2: Specific heat content of oil. Its not as good as water by a long shot. So, for equipment that operates below 100C, water can transport much more heat than oil. CPU and GPU water coolers are common technology. So what good is oil?

Problem 3: This doesn't eliminate fans, pumps, heat sinks, etc. that eventually move the heat into the surrounding air. That stuff still makes noise. Apart from some custom systems that plumb coolant to remote heat exchangers where the noise isn't a problem, this thing will still have fans. So what did we gain?

Problem 4: How much power do those kewl blue LEDs consume? Here's an idea ....

Re:I see a few problems

[Anonymous Coward]

Regarding fire hazards: This is no problem, oil has been used for ages in heavy power switching equipment to quench the electric arc which is made when switching. See http://en.wikipedia.org/wiki/Switchgear for a reference.

As far as the rest of your points - agreed :-) I don't see the added value too, except for bragging rights.

Re:I see a few problems

[david in brasil]

Oh boy - a dielectric fluid thread. My specialty. The post above is not really correct. I'm Engineering Manager of a company that makes dielectric fluids, including PCB alternatives. (www.dsiventures.com) PCBs in and of themselves are not so bad from a health perspective; the bad actors are the dibenzofurans and dibenzodioxins that are created when PCBs get overheated in an electric arc or when they contact very hot metal. PCBs have been found to be bioaccumulative, however. In the 1970s and 1980s, several alternatives were developed - polydimethyl siloxane (silicone fluid) as well as High Molecular Weight Hydrocarbons. It's true that these fluids are not *non*flammable; they do have a flash and fire point when measured by ASTM Method D92. These fluids have a fire point > 300 C, however. I've performed a lot of experiments on burning and exploding transformers; in order to get a tank of these oils to burn, your building pretty much has to burn down around it. So a fluid doesn't have to be nonflammable, such as PCBs, in order to be safe; we could sit down over lunch and come up with a test method that would burn an iron bridge, but that doesn't mean that it's going to happen in real life.

Re:I see a few problems

[Smidge204]

Problem 1: Silicone oils jump readily to mind as a non-flammable replacement for PCBs. If PCBs were still used in anything past the 1980's it's because industry didn't want to, not because alternatives don't exist.

Problem 2: Specific heat of oil is still better than air. Also, full submersion has the benefit of keeping the components clean, cooling *everything*, and not necessarily needing any pumping at all (Properly designed, the natural convection plus high thermal conductivity are enough to keep things within operating temps.)

Problem 3: Properly designed, it can. See above. People have dumped their PCs into fish tanks of mineral oil and it works fine without any fans or pumps, easily transferring the heat to the surface of the tank and out through natural convection.

Problem 4: Probably tenths of watts. What's the problem, exactly, besides being a pain if you want to have the room dark?
=Smidge=

Re:I see a few problems

[Muad'Dave]

Problem 1: Fire codes. To date, nobody has come up with a non-flammable insulating oil to replace PCBs (carcinogenic, nasty stuff). Oil is flammable and, in combination with electrical equipment, a very bad idea. Generally, it is not allowed within occupied ares outside of fireproof vaults.

Mineral oil is not classified as 'flammable', but 'combustible'. It must be heated to 275F (its flash point) before its vapor will support combustion.

Canola oil's flash point is extremely high - 620F, with no autoignition temperature. One of the Novec fluids has no flash point, but autoignites at 375C/707F. I'd say they're essentially the same from a fire hazard standpoint. If anything gets close to 620-700F, my computer coolant is the last thing I'll be worrying about.

Re:I see a few problems

[hughk]

PCBs are really useful when you are talking 10Kv or so but do you really need all those multi mega ohms and high breakdown for PCs?

Re:I see a few problems

[PPH]

I am very familiar with such switchgear, as I have spent a good part of my professional career designing and/or specifying it for projects. It is not allowed within occupancies other than within suitable fireproof vaults. Vacuum or air break switchgear is used indoors. Also, oil filled transformers are not allowed indoors. Dry types are used instead.

Re:I see a few problems

[PPH]

And yet, industry has avoided liquid-cooled transformers for indoor use since the EPA banned PCBs. Even when the cost of remodeling buildings to accommodate larger dry type transformers is exorbitant. I've done a few of these replacement jobs back when the PCB-filed equipment had to come out. Subsquent to that, I've spec'd quite a bit of electrical equipment for indoor use. None is oil filled. Some day, non-flamable materials may be developed and fire codes modified to allow them. But not today, economically.

Don't count out oil igniting in a structure fire. That is an important reason for fire codes' restrictions on oil-filled equipment. To reduce the quantity of fuel available in a fire. Imagine one of these things sitting under each employees desks in a building that catches fire.

Re:I see a few problems

[Anonymous Coward]

So if I understand correctly, the problem is not that the components submerged in the oil will cause the oil to catch fire, but more that if a fire were to happen on the site, the oil might fuel it, resulting in a fire of a much larger scale?

Re:I see a few problems

[PPH]

The specific heat of oil is still worse than water. CPU water cooling systems are old new by now. And in the final analysis, the only thing high specific heat buys you is a lower mass transfer rate for the same heat flow (fewer gallons per minute move the same heat to the radiator).

Free convection is limited by the case surface area, not the thermal mass of the contents. When the system reaches equilibrium, the temperature of the case will reach whatever is required to dissipate the heat generated within. All the oil does is to slow down temperature rise. That might be OK for a gaming system, which is only used a few hours a day and can cool down using convection for 24 hours.

Do a thought experiment (or heck, actually build this and test it). Take a PC case and stick 300W of electrical load inside it (light bulbs come to mind). Turn them on and allow the case to reach equilibrium. Measure its surface temperature.

What liquid (water or oil) cooling does buy you is the ability to move heat away from a few components with limited surface area to either a radiator or the surface of the case. Air cooling would require very high mass flows and the associated noisy, high velocity fans. Oil is better, water better yet. But you still have to get that heat out of the case. In the end, air may have poor heat capacity, but with forced air system, you've got thousands of cubic feet of the stuff to play with. With a radiator system, the air velocity over a large surface area can be slow, resulting in less noise. But you've still got to move the same cubic feet of air to cool things.

Immersing an entire motherboard seems somewhat bizarre. There are only a few problem areas within a PC w.r.t. heat. The CPU, GPU, power supply, and disk drives. The rest of the system can be cooled with very low air flows (internal convection may suffice).

Re:I see a few problems

[Smidge204]

It's curious how you seem to do a total 180 on your reasoning but still come up with the same conclusion by introducing new arguments.

First, you will never get away with dousing your PC in water. Even ultra-pure, deionized water will eventually damage something, and it would be a trick to keep it ultra-pure. A properly selected oil does not have that problem.

Yes, the total heat you can remove from the computer with passive cooling is limited by surface area - nobody's arguing that. Consider this, though: Nobody said the thermal path from chip to case had to be passive as well. For example: you can actively cool the oil itself to very low temps with no risk to the electronics. Compare to cold air/chilled water which runs risks of condensation.

In other words, you're right if everything is left to it's own devices... but nobody said you couldn't throw put a block of ice on top of the sealed, oil-filled case. (That's allegorical, BTW. Though that makes for a great mental image.)
=Smidge=

Tried this before, didn't work so well

[TrondS]

We made something like this some years ago. We submerged a MB in oil and put the whole thing in a freezer. It worked great, until the oil penetrated in between the pci card slots (yes, it was an old computer)and MB. This also affected the ram slots. This made the electrical connection between ramMB and pciMB pretty useless after a while, effectively shutting the thing down. We soldered the ram and pci cards to the motherboard, and this solved the problem.

Re: My only question...

[jjustus]

Looks like it's gonna be one hell of a rig.

No, you're wrong

[Anonymous Coward]

I'm the Engineering Manager for the company that makes the oil. Because of a NDA between my company and Hardcore Computers, I'm posting anonymously and I'm not at liberty to discuss the oil specifically, but I can tell you that it's a modification of another of our products; a very high grade synthetic hydrocarbon dielectric fluid that's been used for years to cool electronic components. It's highly biodegradable, nontoxic, and nonhazardous. It's not a vegetable oil or a halogenated hydrocarbon. And it's not made by 3M.

The more things change, the more they ...

[dsgrntlxmply]

... retrogress. The first computer that I programmed was the IBM 1620, designed around 1959. Its core memory was in an oil bath, but this was for heating (presumably to a temperature where it had the desired magnetic behavior) rather than cooling. When you powered this machine on, it required 10-20 minutes to warm the memory up, before it would allow you to compute.

Time for an oil change

[Drakkenmensch]

What I really want to know is where do you find a repair outlet equipped to deal with the specific problems involved with opening and repairing an oil-filled cube? I can't help it but giggle when I imagine someone calling Dell tech support and being asked to 'please open the PC and reseat the memory'...

combine your ideas

[circletimessquare]

propane is now being touted as a more environmentally friendly refrigerant than freon. major manufacturers ae looking into it. the downside is safety of course, but on the plus side, existing machines that replace freon with propane report a 10-15% efficiency increase

water not so bad

[p51d007]

And here I thought the box I've been running for 5 years with a Corsair Hydrocool was pushing the nerd knob a little high.

Only a rich douche bag would buy this thing.

[Anonymous Coward]

The kind of douche bag who is full of angst over his success, and thinks that voting for Obama will assuage his embattled psyche. Liberal White Guilt. Gotta love it!

Re:Bulletproof? Well, that oil rig will be

[davidsyes]

seriously "capped"...

Is it made by Rumley?

[Iowan41]

(The first internal combustion tractors uses oil for cooling, avoiding the need for a pressurized radiator. Rumley was the main name associated with these "oil-pulls"

Correction: not a hydrofluorocarbon

[Animats]

Correction: they're not using one of the exotic 3M fluids; what they're using is more like a specialty synthetic transformer oil.

Somebody over at Extreme Overclocking not only used HFE-7500, they cooled it down to -100F. This fluid remains a flowing liquid from -148F to +262F, so it's possible to use it in an unpressurized cooling system and without the coolant boiling off when the system isn't running.

Now that was a cooling project. They built a case from scratch, using a plasma cutter and TIG welding. The system has two compressors, 1.5HP and 2.5HP, which means the cooling system alone takes about 5 kilowatts to drive. There's plumbing, gauges, and expansion tanks. But unlike the liquid nitrogen and dry ice projects, which are more like demos, this can run continuously without adding fluids. Although, at 5KW in, MIPS/watt are rather low.