Multi-brand GPU mayhem likelyATI + Nvidia = LucidlogixBy Paul Taylor: Wednesday, 16 July 2008, 12:44 PM ISRAELI OUTFIT LUCIDLOGIX Technologies is touting its brand-spanking-new Hydra Engine technology that will allow cross-branded GPUs to operate hand-in-hand. Sounds grand, doesn’t it?This market-shaking feat is performed with the use of a dedicated SoC, dubbed the Hydra Engine, which distributes processing tasks to different GPUs and optimises data flow to and from each one, maximising (in theory) the potential performance per GPU. The firm uses some nifty data compression/decompression algorithms to simplify things, it says, in a cost-effective manner (although, as we all know, GPU economics reside in the Twilight Zone).The claim is quite grand, as it states that this will allow you to throw a bunch of different graphics cards into the mix and get “near-linear to above-linear performance” across different GPUs. The Hydra Engine analyses and adapts on-the-fly to feed each GPU, creating parallel processing engines under any CPU, chipset and GPU combo.A simple graph on the site shows several graphics cards wearing green and red colours (nudge nudge wink wink) fed by the Hydra Engine and outputting to a display. It’s pretty vague, if you ask us, especially since they don’t mention if you can combine the processing power of the GPUs.As the technology is described as a “real time distributed processing engine”, and if their idea of distribution is anywhere close to CPU load balancing, it’s likely to be just distributing processes to the idlest GPU available. We don’t expect Hydra to SLI/Crossfire ATI and Nvidia GPUs into a single graphics processing behemoth, but rather keep things nice and tidy within the data path.We called them up to get some clarification, but the prez was already on the phone with another inquisitive hack and was getting back to us “as soon as possible”. We’re waiting for that call.In the meantime you can see that parallel processing is becoming big, both on CPUs and GPUs, and that a device of this nature could do a lot more than just improve “gaming” and “office computing”. We’re sure render farms would be more than happy to tap into this technology... and we’re quite sure there’s even a place for the IP within Intel’s own Larrabee or AMD’s Fusion.Getting back to the details: the Hydra Engine can be soldered onto the motherboard or slotted as an add-in board, depending on what each partner will want to do. The technology will support OpenGL and DirectX and should be available through a number of said partners in early 2009.If the firm isn’t gobbled up in the meantime, that is. Guess which chip maker has a stake through the company right now. Yeah, the one beginning with I.
Lucid makes multi-GPUs easyIDF San Francisco The best so farBy Charlie Demerjian: Wednesday, 20 August 2008, 9:00 AMClick here to find out more!EVERY ONCE IN a while, a company comes out of nowhere with a really good idea, and Lucid Logix's Hydra 100 chip is one of those. It basically allows for seamless multi-GPU load balancing between disparate GPUs.The Hydra Engine is a little chip that sits on the motherboard taking up only about 5W and requiring no heatsink. It lies on the PCIe bus physically, and logically between the DX or OpenGL software and the PCIe Bus driver. The software intercepts graphics calls and shunts them off to the Hydra 100 for 'magic'.The Hydra 100 chipWhat happens really is magic. You can take up to four GPUs from the same manufacturer and put them together in a system. Got an old 2600 series ATI card on the shelf that you replaced with a 4850 recently? Why not plug it in and add a bit to performance, with the Hydra Engine, it just works.The way the magic happens is the chip will dynamically read GPU time used, GPU memory used, textures left in GPU memory, pixel shader bandwidth and a host of other things in real time. It also dynamically figures out the capacity of each GPU in the system.It knows if it needs to draw one million pixels a frame, and GPU1 has 3x the power of GPU2, that 750K pixels go to GPU1 and 250K go to GPU2. It can do the same for geometry and all the other functions on a sub-frame basis. Each frame, it reevaluates the mix, so if the scene changed from geometry heavy to shader heavy, it can deal with it in real time.Lucid rendering picAs you can see above in a debug mode demo, half of the frame is rendered on one GPU, the other half on the other. Unlike Crossfire or SLI, you don't get alternate frames, and you don't get split screens, although in some modes Hydra can do split screens. Instead you get per object, or sometimes sub-object breakdowns of the render.Although it is a little hard to see here, the screen on the left is rendering the floor and some of the windows while the right renders the windows that the left does not, and skips the floor. Basically, it renders things in an uneven way on both GPUs. This changes on a per-frame basis, so if you run around, it turns into a psychotic lightshow as the chip parses things out differently each frame.After it is all rendered, the raw pixels are sent back across the PCIe bus for recombining on one GPU, and sent to the monitor. This has a really big bandwidth advantage over SLI and Crossfire because it means that a frame is sent out once to the Hydra 100, not once per GPU. That more than makes up for the pixel bandwidth used in recombining.The initial Hydra 100 supports up to 32 PCIe 2.0 lanes, and you can split it up to 2x16 or 4x8, hence the 4 GPU support. It could possibly do 8 GPUs in 4870x2 configurations, but that will have to wait for future drivers.If you think this sounds neat, Lucid can do a lot of tricks. Multi-monitor acceleration is the obvious one, but that is obvious. Imagine you are watching a movie on one screen taking up essentially no GPU time, and on the other you have a game going. The Hydra 100 can use the excess GPU power to accelerate the game on one monitor, and give only enough to draw the screen on the second. If you can think it up, the Hydra can probably accomplish it.What Lucid came up with, if it ends up working as advertised, is the holy grail of multi-GPU acceleration. Pick any four GPUs from a given manufacturer, and throw them at a machine. It simply makes them work together correctly, automagically. If you buy a faster card a year later, no problem, add it in. I can see this being a really nice selling point for laptops and high end systems. It should just work, and comes with a lot of bells and whistles as well.The Hydra 100 should be out around the end of the year with no price set yet. Lucid won't sell them, they will be on the motherboard, and sold by the various board makers. If you don't game, it will be transparent. When you do, it will be magic.
Multi GPU tech Lucid to take on graphics giantsSLI and Crossfire in its sightsBy Sylvie BarakFriday, 26 June 2009, 14:39AFTER A QUIET SPELL multi-GPU magician Lucid is making a noise again, this time bigged up by a report from Jon Peddie Research (JPR).The firm's impressive Hydra offering was shown off for the first time on the INQ at last year's IDF.Lack of news since then has resulted in some dismissing the firm and its products as vapourware: nice in theory, but nowhere in practice. Now Lucid, which counts Intel capital as an investor (and protector from AMD/Nvidia lawsuits), stands a good chance of scoring a big hit soon, according to JPR.Peddie's report reminds us that "the Holy Grail of computer graphics is to... create an environment that is so immersive, rich, and compelling that participants stop looking at the image and start feeling as if they are actually part of it." Easier said than done.This is where the idea of multi GPUs comes in. But throwing more AIBs (add-in graphics boards) into a PC doesn't result in commensurate increase in performance.Adding in a second AIB usually only ups performance by around 50 per cent (or slightly more on selected apps). Furthermore, load balancing remains the greatest challenge and makes scaling improvements complex.ATI and Nvidia have come up with their own ways of getting the most out of multiple graphics boards: the Green Goblin's SLI (Scalable Link Interface) concept and ATI's Crossfire method.LucidLogix, however, has gone with a front-end approach to multiple AIBs. It has designed a different load balancing technique for scaling, and can use more than two AIBs. That is truly innovative. According to JPR, Hydra's scaling isn't quite as good as Crossfire's or SLI's for VGAs yet, but that scarcely matters with its purported ability to scale far beyond just two GPUs.Nvidia's SLI and ATI's Crossfire run in alternative frame rendering (AFR) mode where each card renders one frame. Problem is, there are inter-frame dependencies, and for each GPU one always renders more than one frame ahead - often two - to make the GPU more efficient.So when users run triple or quad SLIs (or Crossfires), they are actually rendering at least eight if not more threads ahead. If a user is running a game at 30fps, eight frames is a big deal: he would experience either an eight-frame delay or skipped frames and lowered performance.However, Lucid's Hydra splits each frame into different objects - say the game character, weapon, floor, walls - to be distributed among the various GPUs. Result - no lags. In theory, it's possible, using Lucid's technology, to split up a scene and have it rendered more or less efficiently by hundreds of GPUs.Furthermore, Hydra is designed to be generic, so the number of game engines available won't affect its ability to support games as they release. Thus it can be used for any GPU supplier and doesn't need game developers poking around with it or tuning it to get results. This saves developers time so they can churn out games quicker.Here's hoping Lucid actually does make it off the drawing board and onto motherboards soon. µ
LOL...here we go again.Lame and boring is just fine when you don't have as much disposable income.At least when something breaks you'll know it wasn't because of you overstressing it.