Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 


Forgot your password?
Close
typodupeerror
×

Self Contained Water Cooled Radeon X1900, Retail 129

Posted by CowboyNeal from the not-just-for-case-modders-anymore dept.
Spinnerbait writes "Graphics cards are all the rage in the Enthusiast Computing community, where overclocking standard off-the-shelf components is commonplace. Recently innovative cooling solutions have been brought to Graphics cards in an effort to tame the thermals of their power hungry GPUs. It looks like some of the major vendors have taken it up a notch in this area, with this ATI-based Sapphire Graphics card that employs a self-contained water cooling system. Not only does the card have potential for serious overclocking but it should do so relatively quietly as well."
This discussion has been archived. No new comments can be posted.

Self Contained Water Cooled Radeon X1900, Retail More

Self Contained Water Cooled Radeon X1900, Retail

Comments Filter:

  • More Power! (Score:4, Interesting)

    by cosmotron ( 900510 ) on Saturday March 18, 2006 @11:50AM (#14948165) Journal
    I wonder how much power this uses if you have your own water cooling for everything else, then a seperate system just for your video card?

    • Remember that spate of stories we were getting about the new 1000W = 1KW power supply units [PSUs]?

      Here are a few of them:

      New 1 Kilowatt PSU - Too Much Power?
      Monday, August 22, 2005
      http://hardware.slashdot.org/article.pl?sid=05/08/ 22/2157244 [slashdot.org]

      Thirty Four PSUs Tested - Is Biggest Best?
      Monday, September 26, 2005
      http://hardware.slashdot.org/article.pl?sid=05/09/ 26/1052233 [slashdot.org]

      A Kilowatt of Power
      Wednesday December 28, 2005
      http://hardware.slashdot.org/article.pl?sid=05/12/ 28/0353252 [slashdot.org]

      Now consider,

      • Why do you assume they will be putting electrolyte in the water?
    • I use a Zalman reservator1 for the watercooling in my system. My processor and videocard are cooled in the same water circuit, with pleanty of extra cooling capability to spare considering I have both chips (CPU and VC) heavily overclocked. All of this is cooled via the standard 5W pump; which is basically a fish tank pump. The truth is that I am using fewer total watts then before considering I have removed 3 case fans and the stock fans on the processor and video card. The whole PC is now much quieter
    • Check out this picture [hothardware.com]. It looks like it just dumps the heat inside your case... great...
      • Look again. It dumps half the heat in the case, and blows the other half out. That's the sacrifice for a near-silent, 2-slot water cooled solution. If the cooler card took up 2 slots by itself I'm sure the fan could blow the air out the back. But then that's 3 slots total. The 120mm case fan(s) should handle this card just fine.
        • Look again. It dumps half the heat in the case, and blows the other half out.
          I don't even think it does that. That looks like a standard laptop CPU fan, which sucks air in at both sides, and sends it out in a disc from the blades, absent any directing plastic bits. It looks like a standard squirrelcage fan, which will blow all the air out the back of the slot. As a bonus, it gets some hot air out of your PC case, too.
      • Some jackass modded me redundant, which I wasn't, so I'll post again with more info. From Sapphire's Press Release: [sapphiretech.com]

        Air is extracted from the case by an integral transparent blue LED fan, passed through the radiator and vented out of the case at the rear mounting bracket.

        As you can see from the other photos, one end of the cooler vents out of the case [hothardware.com] while the other end is closed.

  • But does it run... (Score:4, Funny)

    by Anonymous Coward on Saturday March 18, 2006 @11:50AM (#14948166)
    Wait, no, it's from ATI, nevermind.
  • Man, I've always wanted a water cooler in my PC. Maybe now if I press "Tab", I really do get a Tab?

  • What a shitty submission. (Score:5, Insightful)

    by Spazntwich ( 208070 ) on Saturday March 18, 2006 @11:50AM (#14948170)
    The "article" is a shitty little blurb with a few pictures. Not a single performance number or any bit of useful information.

    My first instinct? Check the link for the submitter's webpage. Oh, what a coincidence.

    Look, I'm not one to normally complain about poor stories and worthless submissions, but this one takes the cake. It's the most obvious grab for clicks and advertisement revenue that's been posted on /. in a while.

    For shame, CowboyNeal.
    • Aha, but his site's been slashdotted!
    • Please...

      Just because its his site doesn't make it any less news.
      • I wouldn't complain if he waited until they ran benchmarks, tried overclocking it, or had some actual news to report.

        You know, the kinds of details that usually make it into a news article. This is "Hey, look. It's a videocard with a waterblock on it. Betcha never seen one of them before!" and the fact that it's shameless self-promotion only makes it worse.
    • ROFLMAO! Its on a shared web hosting server to boot it seems. Seems MySQL isn't available .. but the banners seem to load just fine .. hmmm.

      (mental note to add no self slashdotting to AUP on my hosting sites).

      List to do today :

      1 - Shoot self in foot
      2 - Self Slashdot
      3 - Find new web host (looks like he got suspended? Just re-checked and getting a blank)
      4 - Make static pages

      The card itself is a great idea (Not a huge fan of ATI but I'll wait for the specs), idea itself sounds solid and kinda neat. Too bad a f
    • Isn't this called Slashertising or something like that? Shouldn't we charge a fee, boil him in oil, DRDoS his server, or tar and feather the boy?
      • Since the /. article blurb is accompanied by the "tagging beta" that says "slashvertisement", who really deserves blame here? Not the submitter, but either /., or the readers who clicked anyway and then complained.
    • Yeah, but the site is called HOThardware, and they're reviewing (if you want to call it that) a water COOLed graphics card.
      Besides, all the fancy graphics hardware in the world can't prevent the Slashdot Effect!

  • nice teaser (Score:3, Insightful)

    by gEvil (beta) ( 945888 ) on Saturday March 18, 2006 @11:51AM (#14948171)
    Since the article is really only a teaser, there isn't much to be said about it. However, assuming this thing works well, it's a nice direction to see the high-end GPU market head towards. Now if only I could afford one...
    • I like how the slashdot tag already says 'slashvertisement'. That just gets it over with, if you don't like these articles, you can skip it!

      Anyway, I thought it interesting enough to look at it, there are several things I don't understand:

      - There's still a fan in there! Why does this thing help reduce noise? Doesn't it just places all the heat to another place, using a just as noisy fan there to get rid of it? Is there anyone who can say something useful about this

      - This is a very silly trend of course

  • Thats all great but... (Score:5, Insightful)

    by Monkeys!!! ( 831558 ) on Saturday March 18, 2006 @11:52AM (#14948174) Homepage
    Overclock all you want and you are still playing the same game as everyone else.

    I'm not trolling, I'm just bitter that everyone is focused on pushing the graphical boundaries of games and leaving the game play for later. I remember a time when it was about hours of game play not frames per second.

    *goes off to play Deus Ex*
    • Glad to see i'm not the only one who still plays that game!
    • Overclock all you want and you are still playing the same game as everyone else.

      Rendering power is overrated when talking about game graphics. If you look at World of Warcraft [photobucket.com], the game looks incredible. But it looks incredible on lower-end systems too... the art direction is just spot on across the board. Half-Life 2 [hardwired.hu] uses some neat graphical tricks, but in general the game looks amazing because the artists had a clear vision of what they were creating and ensured that every pixel that went on the screen

    • Once you get past a certain number of frames per second, it doesn't freaking matter. It's not going to one bit more for the quality of the experience.
      • Not exactly true. Anyone who's ever played a shooter can tell you that your FPS mileage may vary depending on what's going on around you. If you're running at 40 FPS when there's two guys on screen, but 20 FPS when they start shooting, the game ends up being choppy during fire-fights and that could get you killed.

        Case-in-point, I saw an Anandtech review of some new hotness video cards and they tested F.E.A.R. on the highest settings, and they averaged about 50 FPS, until some explosions went off and their f
    • Overclock all you want and you are still playing the same game as everyone else.
      True, but I think most of us overclock for the sheer fun of it. I agree, though, these days there is hardly the use for it that there once was; if there ever was.
    • Ah, but getting 210,000 frames per hour is so much better than only getting 205,000 frames per hour!
    • People associate overclocking with fancy cooling systems, but if this runs at normal operating paramaters hot enough to require a water cooling system, it's not exactly ideal for overclocking... it's already up toward the limit of heat dissipation at the default retail speed.

      --
      Evan

    • Overclock all you want and you are still playing the same game as everyone else.

      If I can see you but you can't see me, are we really playing the same game?

      If I can turn and point faster than you can, are we still playing the same game?

      That's the whole point behind the video card race, and why I went to console gaming. Becuase at least there, more people are indeed playing the same game (not as true now that some people play in HD and some not).

  • I'll Pass... (Score:5, Interesting)

    by evilviper ( 135110 ) on Saturday March 18, 2006 @11:53AM (#14948176) Journal
    I really don't want graphics cards with better cooling systems. I want them to RUN COOLER in the first place. Water-cooling a device just allows you to push the problem back a little further, before it really starts causing problems. Pretty soon you'll have to upgrade your power supply and home airconditioner to use a shinny new GPU.

    • Re:I'll Pass... (Score:3, Insightful)

      by iamplupp ( 728943 )
      I agree. And all it takes is lowering the frequency by maybe 20% and you will be able to run it on much lower voltage and thus less power. And who needs that extra 20% anyway, really?

  • External PCI-X connector (Score:4, Interesting)

    by Twillerror ( 536681 ) on Saturday March 18, 2006 @11:54AM (#14948181) Homepage Journal
    I think it is about time to start thinking about the GPU as a device, like a cd-rom or harrdrive instead of a card.

    They already require an external power supply in some cases, and with SLI are using a special type of connector.

    I'd like to see a GPU that comes in the form of a 5.25 bay expansion, with a pci-x card that connects it via a cable to the mobo.

    Then I think the industry could come up with a standard cable for all cards. Or not, given that nvidia or whoever could come up with just about anything. I could see a card that interfaces through the memory slots ( if your motherboard had enough realastate). A GPU directly connected to Hypertransport anyone?

    • As cool as that would be, the frequencies, data rates and low latencies required for a video card would cause the cable to be prohibitively expensive in order for the error/corruption rate to be comparable to a plug in card.

    • while you have an interesting idea, i don't think anything like that will be made in the near future for amyn reasons. first off, it would go against an industry form factor (PCI type expansion slots) that has been in place for a lomg time, and companies like nvidia and ATI would not want to risk making a venture outside such a long standing form.

      another problem with changing the form for graphics units is comaptibility, as i for one do not have a free 5.25" bay in which to put something (i already have

      • Frankly, I'd like to see entirely modular computers. Implement it like legos: bricks stuck to each other. The interconnects would be awful, but if it could be done, and made hot-swappable, it would be way cool --- and would allow nearly eternal upgrades. Why replace the CPU each year when you can just slap a newer one on and have them all going? And so forth.

        • well CPU's are modular, you can put a new one in so long as its compatable. i have a socket 939 motherboard and have many upgrade options as my needs change. sure socket 939 will be phased out when a better format comes along, but by the time i run out of 939 options, i'll want something new to begin with. computers are very modular and upgradable, its just that new upgrade options are coming out so frequently. i still use my 166MHz thinkpad laptop everyday because it works great for what i use it for (

    • Phhbb... pretty soon your not going to hook your vid card to your pc, you'll hook your pc to your vidcard.

  • Water? (Score:3, Interesting)

    by nurb432 ( 527695 ) on Saturday March 18, 2006 @11:58AM (#14948196) Homepage Journal
    It could be someting other then water, like antifreeze.. Woudl be able to use a peltier block and run it even colder then you can with water.

    Or even sodium. So what if it breaks and the user dies ;)

  • Why water? (Score:4, Interesting)

    by bogaboga ( 793279 ) on Saturday March 18, 2006 @12:02PM (#14948216)
    I have always wondered:

    Does it have to be water to cool these GPUs? Is it water because of its relative hight Specific Heat Capacity, or because it's cheap and readily available?

    I can see slashdotters increasing the capacity of the "tank" that stores water on these GPUs to make sure the GPU stays cool.

    • Re:Why water? (Score:1, Interesting)

      by Anonymous Coward
      It's because water is cheap and readily available. It's also non-toxic, an important consideration for overclocking types who accidentally trip over the hoses they use to run their machine 0.2% faster than stock. As for increasing the water capacity of the system, unless you're planning on increasing it to infinity it won't make any difference - at some point the water will start coming back hot unless you've got a method of cooling it in the loop. Hint: water cooling isn't magic, it's just heat transfer ov
      • Ah man, now I really want to put some watercooling on my old P4 rig. I mean, the 30% overclock is nothing compared to a 0.2% increase...

        P4 1.6a that has been running at a 30% overclock on aircooling for 3 years. It gets maybe a few hours off per year and has so far outlived a motherboard and PSU and is still running strong. Thanks for playing...
    • Like the above posters said, water is readily available and non-toxic, but the high specific heat capacity certainly does help. I think the only liquid with a higher specific heat capacity is ammonia.

  • Too embarassing (Score:4, Insightful)

    by Expert Determination ( 950523 ) on Saturday March 18, 2006 @12:02PM (#14948217)
    Remember when that company introduced cryogenically cooled PCs? I got one at work. It was amazing - it ran at 1GHz and I had the fastest machine in the building. But a few months later it was no longer the fastest in the building but it was definitely the dumbest machine in the building - especially with the 5 minute wait for it to cool down before booting. I won't make the same mistake with water cooled graphics cards.
    • Watercooling doesn't require you to wait for anything to cool down.
      • Watercooling doesn't require you to wait for anything to cool down.

        Yes, your post is very embarassing indeed. The parent was obviously talking about the systems with built-in phase change cooling (e.g. Prometia) which do take several minutes to get down to the right temperature.

        But because you've never heard of those, you assumed he was talking about water cooling and made a tit of yourself flaming him.

        It's pretty hard to think of anything more embarassing than that.
    • So I take it those 5 minutes to boot weren't made up during the day with your screamingly fast computer?

      • So I take it those 5 minutes to boot weren't made up during the day with your screamingly fast computer?

        In CPU time yes, in psychological time no. You can imagine how it is: you get to work and want to read your email. You have to sit there waiting fof the compressor to get going and watch the temperature on the gauge slowly drop. It's a fun novelty thing for the first week or two and after that it's annoying. On the other hand, suppose I trim 5 minutes off a 1 hour fluid simulation (which is what I was

  • Self contained and silent? (Score:5, Interesting)

    by the real darkskye ( 723822 ) on Saturday March 18, 2006 @12:04PM (#14948224) Homepage
    Is it actually self-contained if you have to refill it?
    Is it actually silent if it still has a fan?

    I think what TF[A|S] actually ment was 'pre-fitted water-cooled ATI'
    • A fan doing 18/26dBA at 2000/2500RPM is effectively silent.

      A pin drop is around 10db & a whisper is about 20db

      As long as you don't sleep with the tower right next to your head, I doubt it'll be an issue.

  • Thank GOD (Score:3, Funny)

    by zerocool^ ( 112121 ) on Saturday March 18, 2006 @12:12PM (#14948244) Homepage Journal

    The last publicity pictures I saw of this card had a couple of "Models" showing it off. The blonde one looked like they had picked her up off of W124th and Lennox at about 4AM.

    Shudder.

  • Silence or? (Score:5, Informative)

    by Bender Unit 22 ( 216955 ) on Saturday March 18, 2006 @12:17PM (#14948253) Journal
    When I looked at the pictures, my first thought was that it would not be more silent, since it still would have to move the same heat, while it might be good for overclocking. But I guess that there is a possibility of it running quietly since there are a bigger surface area for the air to move the heat than on a normal GPU cooling fan and heatsink.

    I made my own PC watercooled about 1½ year ago with the purpose to make it more silent. My idea was to cool the CPU and GPU using both passive and active cooling.
    I got a radiator from Innovatek.de for passive cooling inside and a small deep one for the inside for active cooling with a Papst fan.
    Then I got a microcontroller that can run on its own, measure the water temperature and control fans, as well a a emergency shutdown if the water gets too hot or the pump fails.
    The end result were fantastic, the PC runs almost silent when doing anything than playing games(which I don't do much anymore) but when playing games it still have the power. It manages to run mostly with passive cooling when not playing games. It is so silent that you have to look at the power LED to make sure that you have turned the PC on. When I do play games and the water starts heating up, the microcontroller starts the watercooling fan and adjusts the speed to keep the temperature down.
    On mistake that I did make was that I went into it with my usual approach of reading tons of reviews on the internet to find the best cooling CPU/GPU heads and generally getting parts from different vendors that I determined would make the best solution. Exept from the internal radiator that was deeper than any other I could find, I can now see that it didn't matter which parts I used when I was not going to do overclocking.
    It is better to stick with parts from one vendor so you don't have problems getting them to fit or work together. Like different sizes of tubes etc. Also the microcontroller from one vendor could not monitor the pump for another.
  • Anyone care to recommend a $100 non-AGP graphics card? I'm looking to upgrade from integrated graphics on a new system, and don't have the coin for a high-end choice. I'm not expecting to get 7800-level performance; just wondering what might offer good bang for the buck.

  • 7900GTX watercooled retail card. (Score:1, Informative)

    by Anonymous Coward
    The EVGA GeForce 7900GTX CO 512MB Black Pearl [webhallen.com] is water-cooled, but not self-contained.

  • Where's the bling? (Score:5, Funny)

    by Glacial Wanderer ( 962045 ) on Saturday March 18, 2006 @12:23PM (#14948274) Homepage
    This card looks very boring. My quad fan card with state of the art stealth cooling fins and racing stripes looks way better. If they added some gold and diamond dust to the water to increases this cards bling I would consider buying it.

  • A little more info (Score:5, Informative)

    by Daath ( 225404 ) <(kd.redoc) (ta) (pl)> on Saturday March 18, 2006 @12:25PM (#14948278) Homepage Journal
    On Sapphire Technologies [sapphiretech.com] there is a little more info, but not much.

  • Most obvious question: (Score:2, Informative)

    by WillerZ ( 814133 )
    What keeps the graphics RAM chips cool? Usually air is blown down onto the processor whence it goes across the RAM chip coolers.

    In the photos, the RAM chips still have cooling fins, and they're still aligned radially around the core; however, since there is no airflow there they're surely going to overheat...
    • My best guess, would be to say, Memory wont need the cooling devices that CPU/GPUs require. As for this situation, Because the GPU doesn't have a fan to blow the hot air towards the memory, the memory themselves won't be in trouble. Most cooling fans, usually blow air towards the heat sinks, then that air gets blown to the sides where the memory resides. When all that hot air to blown towards to memory chips, then they would be in the need for the cooling element.

      Since watercooling takes the hea
    • Surely you're right and every time the company tests their prototypes crashes are a huge problem because they didn't think of this.
    • In the photos, the RAM chips still have cooling fins, and they're still aligned radially around the core; however, since there is no airflow there they're surely going to overheat...

      All I can say is that my watercooled 7800GT has the same sort of cooling fins on the RAM, low air circulation inside the case, and so far no troubles. Remember that since the GPU is not heating up the air inside the case, the air is a lot cooler.

      Main problem I can see with this card is that many case designs would have the PSU s
  • One day, graphics cards will be able to do soo much on their own, they won't even want me to play the game with them anymore.

    They grow up soo fast. =P
  • Why do ATI give Mobile chipsets the finger? Acer is one OEM that is not supported in their mobility, why can NV provide Unified driver sets and ATI cannot, is it too technical for ATI? ATI as usual, let down their end users by drivers in favour of OEM's. Money from OEM's, Finger to the end user. My next laptop is a dual core AMD but most likely they will be ATI mobility chipsets and as such if it is Acer im screwed out of the box.
  • ...onto more efficient chips rather than just bolting on cooling to the problem. That goes for the entire processor industry too.

  • Seems an odd design? (Score:5, Interesting)

    by SmallFurryCreature ( 593017 ) on Saturday March 18, 2006 @01:25PM (#14948506) Journal
    I always thought the problem with cooling a PC was in getting the heat OUT of the PC case. Air can be a good cooler but the problem is that it is so hard to direct the cold air in and the hot air out. Escpecially in a PC case wich is usually cramped and filled with great big walls wich block the airflow.

    CUE watercooling wich takes the heat via the water outside of the case where you can have a slow unobstructed fan get rid of the heat.

    Almost every design I seen always gets the hot water out of the case to be cooled down by large unobstructed fans.

    Yet this setup seems to pump the water from the hot graphics card to a spare PCI slot in your PC where the fan will be blowing the heat away right inside your computer.

    If you unlucky right back onto your gpu.

    To be efficient the cooler would have to be outside your case, with the water cables coming out of the back of the gpu and the cooler not having a PCI mounting but something that is easily attached to your case.

    It is not the first time I seen this mistake, people put the nicest fans inside their case but never spend any time considering that all that does is blow the hot air around if you do not somehow setup a flow to carry it out. Oh and another to get cool air back in.

    • That is how this works. It also has the advantage that it doesn't have the radiator outside the case. The radiator sits in a spare pci slot cooling down the water and blowing the hot air right out the case. The biggest problem I have had with cooling video cards is that cases were not designed to move air through the video card and thus the fan just blows air into either the powersupply or another card and then sucking it in agian. It creates hots spots in the case to the point that I can have a 20C tem

  • Is this some sort of lame pun [hothardware.com]?

  • That was my first reaction. This thing eats up a lot of room in a case, more than an SLI rig I would think (which strikes me as the competitor). Are you better off with this or with 2 X1600s in CrossFire?
  • I would like to remind everyone that the reason why water or whatever fluid is inside it helps cool down the card is because of its high heat capacity. This also means that its slower to cool it down when it's already very hot.
    • Poor conclusion. The high heat capacity allows for lots of heat transer, in either direction. Without enough information about the copper waterblock and radiator on the cooling end you cannot make an assuption that it will cool slower. Simply this all depends on the design performance.

Slashdot Top Deals

I tell them to turn to the study of mathematics, for it is only there that they might escape the lusts of the flesh. -- Thomas Mann, "The Magic Mountain"

Close