Become a fan of Slashdot on Facebook

 



Forgot your password?
typodupeerror
×
Displays Hardware

U of CA Constructs 220 Million Pixel Display 145

eldavojohn writes "Engineers at the University of California, San Diego have built a 220 million pixel display across 55 high-resolution tiled screens. Linked via optical fiber to Calit2's building at UC Irvine, the display can deliver real-time rendered graphics simultaneously across 420 million pixels to audiences in Irvine and San Diego."
This discussion has been archived. No new comments can be posted.

U of CA Constructs 220 Million Pixel Display

Comments Filter:
  • by admactanium ( 670209 ) on Thursday August 23, 2007 @02:37AM (#20327359) Homepage
    they're an hour apart. that's a lot of people fighting over the remote.
    • Whether the screen is in Irvine or San Diego, it must be pretty big if they can also see it in the other city. Or is it on a mountain top between the two cities? And as the previous poster insightfully observed, which city is the remote in? And the really important question: Satellite or cable?
    • I actually do some work for this professor. The whole thing with UCI is that he came from UCI last year and now works at UCSD, so there's some collaboration between the two universities and they share some of the same grads.

      I asked the professor if he was going to play games on this and he laughed, but I guess he's played some FPS games in the day.
    • I tried to view the picture of this, but it won't fit on my paltry two monitors...
  • Human eyes (Score:3, Insightful)

    by biocute ( 936687 ) on Thursday August 23, 2007 @02:38AM (#20327365)
    That's all good, but are our eyes capable of viewing every single px of it?
    • Re:Human eyes (Score:5, Informative)

      by tonsofpcs ( 687961 ) <slashback AT tonsofpcs DOT com> on Thursday August 23, 2007 @02:58AM (#20327447) Homepage Journal
      No. "It's exciting," said Joerg Meyer, a professor of computer graphics and visualization who helped develop the screen's software. "This display has higher resolution than the human retina can see." http://www.latimes.com/news/printedition/californi a/la-me-highdef13aug13,1,5603082.story?coll=la-hea dlines-pe-california [latimes.com]
    • No, but yes (Score:5, Informative)

      by SpeedyDX ( 1014595 ) <speedyphoenix@gmailCOMMA.com minus punct> on Thursday August 23, 2007 @03:26AM (#20327577)
      The resolution of the human eye is relatively minute (it's usually not measured in MP, but I think the best equivelence was quoted around 15 MP at any given time). The easiest way to explain it, I think, is that your eyes are never in the same position for more than a split second. It's constantly moving and looking at any given object from a multitude of different angles. So no, it isn't able to see 220MP, but at the same time, it is (theoretically) able to see a better image with a higher pixel count, because of the fact that your eye is never stationary.

      But that doesn't take into account your brain. Your eye transfers raw data to your brain similar to a bitmap/RAW file. The way your brain processes this information, though, is more like a vector image. Our brains "see" lines and shapes much more than it sees individual points of colours. Which makes the answer even more complicated. We don't really see all the pixels, but we're able to piece together most of the pixels while our eyes move about, ALTHOUGH our brain "transforms" that information so it makes more sense to us.

      A really neat example that illustrates how the brain processes raw data: close your eyes, and get a friend (or yourself, if you can trust yourself not to cheat) to hold up something that is near the outer edge of your peripheral vision. Open your eyes, but don't move them - keep looking straight ahead so that the object is still near the edge of your peripheral vision. You can SEE the object, and can possibly even tell what it is. But what colour is the object? Even though your eyes are able to see colour even in your peripheral vision, the brain doesn't think that the information of colour is as important as the outline/shape of the object. It is only when something is near the centre of your vision (in other words, where your attention usually is) that you can tell what colour it is.
      • Re:No, but yes (Score:5, Informative)

        by wumpus188 ( 657540 ) on Thursday August 23, 2007 @04:29AM (#20327855)
        No. You cannot see the color of the peripheral object because the cone cells, which are responsible for the color vision, become sparse towards the edge of the retina.
      • Re: (Score:2, Interesting)

        by Anonymous Coward
        I got the impression from a new scientist article that the transfer to the brain is not ina "raw" format, but in the format that you describe the brain using- the processing is local (in the optical nerve) to save bandwidth.
        It described the mechanism being a second set of nerves behind the rods and cones that fired in response to certain relative changes between nearby rods and cones.
      • Re: (Score:3, Informative)

        by E++99 ( 880734 )

        The resolution of the human eye is relatively minute (it's usually not measured in MP, but I think the best equivelence was quoted around 15 MP at any given time).

        It's not usually measured in pixel count because pixel count is an entirely irrelevant concept to eye resolution. The angular resolution of the eye is extremely high at the center of the image, and falls off extremely rapidly in a very steep bell curve. So unlike a monitor, the number of pixels across the eye's vision does not correlate at all t

      • Re: (Score:3, Insightful)

        by $random_var ( 919061 )
        Your eye transfers raw data to your brain similar to a bitmap/RAW file.

        No. While the bulk of the signal processing is performed in what is theorized to be a 4-layer neural network, the retina is actually able to perform a substantial amount of processing on its own. For example, lateral inhibition between receptors highlights edges - at the edge of black and white, the data that is sent to the brain actually shows the black as more black and the white as more white. There are also thought to be motion de
  • Remove the seams (Score:5, Insightful)

    by QuantumG ( 50515 ) <qg@biodome.org> on Thursday August 23, 2007 @02:38AM (#20327369) Homepage Journal
    Then we'll talk.

    • by Fred_A ( 10934 ) <fred@freds[ ]e.org ['hom' in gap]> on Thursday August 23, 2007 @03:22AM (#20327559) Homepage
      I think that since you're supposed to look at it from 7km away you can't see the seams.

      The main problem is that they need 408 repeaters for the USB mouse and keyboard.
      • by QuantumG ( 50515 )
        If you're supposed to look at it 7km away then all you need is 1 pixel per screen.

        • True, but the sheer number of pixels allows for a much larger palette of colors and gradients, in only a 55-pixel grid. Imagine your favorite NES sprites squashed and rendered in more colors than your eye can see...
        • Re: (Score:2, Funny)

          by Woy ( 606550 )
          And a high chair.
    • by mwvdlee ( 775178 )
      The only thing amazing about this is that they managed to get budget to buy 55 high-resolution LCD displays just so they could stack them together.
      • Yep... what is the point? Without all the seams it would be kinda cool (for a couple of minutes), but as it is, it seems to be a totally useless novelty.
      • Re: (Score:1, Insightful)

        by Anonymous Coward
        You obviously don't know shit about graphics if you think you just stack them together and they magically output high resolution images. How large of an image do you think the graphics card in your computer can output? So if you think it's easy, then tell me, how would you solve i/o and scaling? Hint: it's not a lego set
        • Re: (Score:3, Informative)

          by MrNaz ( 730548 )
          Anything like this is easy if the Uni gives you its yearly showoff budget.
      • by mikael ( 484 )
        And that's just the prototype version - the final version will have no gaps between the monitors.
    • You are right, and it is possible. Last year at the IBC I saw two coupled very large HDTV screens, fed by two separate HDTV streams originating from two coupled cameras, displaying one soccer match. That was nice: supersharp and a display ratio of 32:9.
      • by ragnarok ( 6947 )
        Large seamless displays are not that exotic, they're used all the time for industrial control and monitoring applications among other things.
        These guys make pretty nice displays: http://www.barco.com/ [barco.com]
        • These guys make pretty nice displays: http://www.barco.com/ [barco.com]
          Nice to hear that name again! At the second IT project I ever worked on, in 1990, they had just bought a truckload of 'the best monitors they could find', from 'a Belgian company'. That was Barco. They were to be used for monitoring large natural gas networks.
          Good to see that quality does pay off.
    • VisBlock(http://www.visbox.com/visblock.html) has displays without bezels (seams). They have a "wall" setup at Virginia Tech (http://infovis.cs.vt.edu/gigapixel/facility.html) .
  • by Ohreally_factor ( 593551 ) on Thursday August 23, 2007 @02:39AM (#20327373) Journal
    University of California at San Diego.

    Can't samezenpus get the least bit of editorial right? Oh, yeah, he can't. He's samzenpus, and he's not an editor, he's an idiotor.

    I mean wtf is U of CA? I've never seen it written like that, ever.

    And to get this rant back on topic:

    Is the screen effervescent?
    • U.C. San Diego is something I've heard/read as well, but U of CA? I don't think anyone uses that.

      With the University of California system there is no at in the title. University letterhead generally includes the names of all campuses. When a specific campus is sending mail, the entire title of that campus will be specified at top. So for UCSD, it would say "University of California, San Diego". Under the seal is the address, which would be: "San Diego, California 9????-????" (I don't know the zip fo

      • >U.C. San Diego is something I've heard/read as well, but U of CA? I don't think anyone uses that.

        My first thought was University of Canada, except I don't think there is one. Googling UCA gets you University of Central Arkansas. For San Diego it's either UCSD or UC San Diego. The title UC Surf is disputed with UC Santa Cruz.
    • Re: (Score:3, Informative)

      THANK you. I made a similar comment [slashdot.org] a while back, and got hit with a -1, offtopic. I sincerely hope you fare better. U of CA should be written as UC, end of story. It's not offtopic this time though, look at the freaking page title!
    • San Diego University?
       
    • by mchanaud ( 1087413 ) on Thursday August 23, 2007 @05:26AM (#20328069)
      as a non american, U of CA is much more comprehensible than UCSD. But, wait, who really matters about non american readers?
      • Re: (Score:3, Funny)

        by cdrudge ( 68377 )
        I would have thought that non-Americans would have interpreted U of CA as University of Canada...
        • by MrNaz ( 730548 ) on Thursday August 23, 2007 @08:25AM (#20329279) Homepage
          We're also aware that /. is run by Americans, and that Americans aren't aware that there are any Universities in Canada.

          Mods: That's a joke. That's "haha" as distinctly opposed to "die troll scumbag take this -1 and then we'll see who has the last laugh!".
          • The best way to present a joke which could be taken the "wrong way" is like so:

            "We're also aware that /. is run by Americans, and that Americans aren't aware that there are any Universities in Canada. BTW I just trashed all my wma, aac and MP3 files and am going exclusively Ogg from now on. I'm also a Linux coder and just bought Biowulf Clusters for Dummies."
          • by e2d2 ( 115622 )
            Complete nonsense. They have Greater Eskimo College (GEC). I looked into a major in seal pup culling with a minor in ice/snow structures. Turned out they wanted me to come to Nunavut, which wouldn't have been a problem... had they roads!
      • U of CA may be more comprehensible, but its NAME is UCSD. You call something by its name, not its descriptor. People don't call you "Human slashdot poster from outside America," or "the letter m followed by chanaud," they call you mchanaud. I have no idea what an mchanaud is, but it's incumbent upon me to learn what one is if I'm interested in a story about you.

        The University of California is a public trust defined by the California Constitution (article 9, section 9, http://www.leginfo.ca.gov/cgi-bin/wa [ca.gov]
      • by rm999 ( 775449 )
        That's beside the point - U of CA would refer to Berkeley, not San Diego. They shouldn't give incorrect information to (try to) make it more clear for non-Americans.

        University of California at San Diego is the optimal compromise.
  • by Joebert ( 946227 ) on Thursday August 23, 2007 @02:43AM (#20327401) Homepage
    Just in time for my $420 million webpage.
  • Eye strain might be something now... :S
  • . . . they could have found a better solution that didn't have the bevels.
  • by Anonymous Coward
    It's not a display, it's a tanning salon for geeks.

    The only thing is, they'll risk overexposing themselves to pr0n and who knows what kind of a mark that'll leave on virgin skin.
    • by ZOMFF ( 1011277 ) *
      Frankly I would be horrified at the sight of pr0n that size. Somethings just don't need that amount of detail (nor size)
  • Now we can watch our undefeated football team in all its glory.
  • pac-man anyone?
  • But seeing as I lack the budget or space for such a display or it's roomful of wiring and rendering nodes, I'm stuck with my trusty old GDM-W900 crt. Hehe.
    • Re: (Score:2, Informative)

      by Anonymous Coward
      Its tiled displays. This is really more about something along the lines of technology(think video card) capable of outputting ridiculous resolutions.
      • I know. Bigass tiled displays like this have been done a few times before. Last one I recall had one "real" computer which had an OpenGL that sent rendering information over a GigE network to SGI Onyx boxes that drove two or four displays each. I'll never forget the article's picture of the highest resolution FPS fragging ever...
    • Re: (Score:2, Insightful)

      by AnimeDTA ( 963237 )
      Its 50x 30" displays, its not about the display so much as the cluster of 18 computers/video cards working together. That sets each individual computer at rendering just under 3 displays each. An easy feat in and of itself. Now getting it to work syncronously with 17 other computers... thats neat.
      • by tucuxi ( 1146347 )
        Still, the seams are no fun. With that resolution, if it were feasible to use tiled back-projection, the display would have a much higher resolution than most IMAX theaters - which use about 70 M pixels per frame, according to Wikipedia. Representing maps or textual information on a tiled display makes it much, much harder to read.

        I also wonder if there are real applications that need that huge amount of detail *everywhere*. I would think that having a few high-resolution areas for detail work (basically wh
      • by dwater ( 72834 )
        ..neat, but it's been done before. This might be better than done previously, but I don't really see how I can know for sure.

        I wonder how many pixels the new LED display is at "The Place" in Beijing. The thing is massive (2,296' by 88') - and it's supposed to be second to one in Las Vegas somewhere (I read that's five XGA-equivalent displays working as one single display - what does that make it? Probably not a lot).
  • What's the point? (Score:1, Informative)

    by akkarin ( 1117245 )
    I mean, it's not a single display: it's a hundred LCDs stitched together.
    When they create a 220 million LCD screen, then great.
  • More like U to the C to the A fo' shizzle!
  • IR4 (Score:4, Interesting)

    by dwater ( 72834 ) on Thursday August 23, 2007 @03:48AM (#20327691)
    Silicon Graphics' Onyx IR4 could drive this many pixels, couldn't it?

    IIRC, it was 16 pipes, 8 displays per pipe, 1920x1200 per display - I make that almost 300M (pixels, not dollars - it'd be *many* more dollars) - probably not remembering correctly, but still. ... and OpenGL Performer could make it all work nicely for visualisation too. I wonder what's happened to OpenGL Performer.
    • by Verte ( 1053342 )
      In 48 bit colour :) It's difficult finding pre-origin-3000 information on high-end sgi systems.

      IR4 is a funny thing. IIRC, both the Onyx 2 and Onyx 3000 supported it, and with the same performance, however, the Onyx 3000 took up twice the floor space :) You would think, with modern technology, we'd almost have an entire IR4 pipe on a couple of cards, and SGI back in the graphics market.

      By the way, if you think the pixel count of Onyx IR4 was impressive, do you recall how many polygons it could render per se
      • by dwater ( 72834 )
        ...no...(I don't remember)

        Futuretech was one of my favourite SGI sites, and it has this on Onyx2 RealityMonster

        Also found this [sgi.com] which says each pipe is 8.3M pixels. So 16 x 8.3M = only 132.8 M pixels. I believe the limit is per pipe, since the DG basically split up the output of the RMs for each display.

        However, I'll bet there were some machines with more than 16 pipes out there....probably secret or something, but I bet they're there somewhere.

        They don't seem to quote polygons per second; I'd guess because
        • by dwater ( 72834 )
          missed the link to futuretech [blinkenlights.nl].

        • by Verte ( 1053342 )
          Those are some serious specs! Still, the system in TFA has a peak theoretical floating point performance higher than that of the largest SGI system ever built, in graphics hardware alone.
          • by dwater ( 72834 )
            To be fair, the biggest SGI system ever built :

            1) may not have yet been built yet (ie SGI is still around and ever isn't over yet), and
            2) the biggest one built to date probably also had higher theoretical floating point performance in *it's* graphics hardware too.
            • by Verte ( 1053342 )
              I actually meant this [top500.org], but that is actually 50% more powerful. Still, it's on the same order.
  • by ejito ( 700826 ) on Thursday August 23, 2007 @03:56AM (#20327733)
    People who are dismissing as just a wall of monitors are mistaken. It takes dozens of computers to run that resolution, which is no trivial task. This is not a theater system, so complaining about seams misses the point entirely. If they were just looking for a semi-large seamless screen, any shmuck could just use a single projector.

    This system allows groups of researchers to review large amounts of visual data in both macro and micro scale. If you want to see the micro scale, you simply walk up to an individual monitor. Review can be done simultaneously among many people.

    For a seamless, 100 million pixel projection screen (this is also not trivial, as removing seams requires real time brightness and color correction along edges) can be viewed here [uci.edu]. In comparison, an IMAX [wikipedia.org] theater uses a very large single projector unit weighing nearly 2 tons.

    The sister screen at UCI can be viewed at here [uci.edu].
    • Re: (Score:3, Interesting)

      This is not a theater system, so complaining about seams misses the point entirely.

      I can't think of a large-format display usage that WOULDN'T benefit from seamlessness. I don't know why you think it's only relevant to movie theaters.

    • This system allows groups of researchers to review large amounts of visual data in both macro and micro scale. If you want to see the micro scale, you simply walk up to an individual monitor.

      Whereas I can sit on my arse all day and just use a display program with zoom function... so I guess this is a device to get lazy researchers doing some exercise? ;-)

  • by wisebabo ( 638845 ) on Thursday August 23, 2007 @04:05AM (#20327781) Journal
    It is conceivable that soon technology/engineering will make it possible to have a multi billion pixel display.

    An interesting application might be to assign a pixel to each person living. Then as they pass through the phases of life, their brightness could wax and wane. Also perhaps color could be used to identify race or geography.

    Might be an interesting display in a world's fair/expo kind of context. Being able to walk right up to it and realizing that you are just one of the billions of little dots could be pretty awe inspiring.

    Perhaps it would give new meaning to the comment "he seems kinda bright". (ba du bum ;)
  • This is what they say in public:
    "...allows us to experiment on the two campuses with distributed teams that can collaborate and share insights derived from a better understanding of complex results."

    But it private:
    "this is fucking awesome!"
  • Bad Pixels (Score:3, Funny)

    by BillGatesLoveChild ( 1046184 ) on Thursday August 23, 2007 @04:39AM (#20327895) Journal
    So how many bad pixels do I need before they'll exchange this thing? http://news.com.com/2102-1041_3-5579493.html?tag=s t.util.print [com.com]
  • My sofa is six feet from the wall.. Reckon that's too close?
  • by Anonymous Coward
    Did anyone else have to square root this to see if it was impressive or not?
  • Or, you could have a killer video update from the beach, the whole wall of your apartment could look like you were standing at the edge of the water. DO want!
  • I work at the Broad Institute on the MIT campus. In the lobby of our building there is a transected cyclinder (think half a burrito cut at a 45 degree angle) made up of seventy-six ~40" LCD screens. The video on them is contiguous, with background elements floating from one screen to another and animations running accross several adjacent screens at once. Conservatively assuming a resolution of 1366 x 768 for each screen, this is at least an 80 megapixel display. Anyone have more details?
  • by Antonov ( 29524 )
    Larger minesweeper tiles!

    "borrowed" from todays cartoon: http://www.userfriendly.org/ [userfriendly.org]
  • Basically this is just an up-to-date Reality Centre. Nothing particularly revolutionary about it these days.
  • "Engineers at the University of California, San Diego have built a 220 million pixel display across 55 high-resolution tiled screens."

    And we get to see a 300x150 picture of it.

    It's nice to see they're keeping their bandwidth for the ads instead of the actual content...

  • I believe the top end telescope is now around 3 gigapixels, by suturing a few hundred large CCD chips. The image is not meant for direct human consumption, but pre-analyzed by computer.
  • His Grandness Stevus Jobus has ordered that this new 220MP display to be the default monitor for the next release of iMac.
  • They have a "hyperwall" that sounds a bit like this in the Calit2 building here at UCI. 50 Apple Monitors stacked 5 high, 10 wide. The guy who showed me it said they use it for some sort of medical imaging or something. Put a couple thousand cat scans of peoples heads up on the display. Also displayed various protein models and the like. I asked someone else here how much it cost. Apparently you can get bulk discounts when you do this kindof stuff. My favorite part was what was behind the display. Each pair
  • Can we hack it to play Battle Field 2 on the screen? I want to be in the game!
  • tiled displays have been constructed at numerous institutes/universities/even companies... what's the novel stuffs here? only more pixels??
  • Neat, but does it run linux?
  • I find multiplying X and Y pixelcounts to get numbers like "220 million pixels" almost meaningless, because most people's intution of how exponents progress (including my own) is way off. (here's an experiment: pour some pennies onto a flat surface. Arrange them so none overlap, just a flat grouping. Guess how many there are. Then count. You probably underestimated. "10 x 10" creates a larger number than most folks' intution "expects".)

    For the same reason, I find having cameras rated in megapixels annoying
  • Why not use projectors and eliminate the LCD borders? Sure you need good curtains, but big deal!

    Projectors FTW!!!
  • Too much to see here. Move along.

Mediocrity finds safety in standardization. -- Frederick Crane

Working...