Lucid makes multi-GPUs easy
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 chip
What 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 pic
As 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. ยต
Share this:
ShareBusiness Intelligence Software
Customer Relationship Management (CRM)
Enterprise Accounting Software
Enterprise Resource Planning (ERP)
Enterprise Systems Management
"from a given manufacturer". Does that mean it cant mix AMD and nVidia cards ? Why not?
"Pick any four GPUs from a given manufacturer, and throw them at a machine."

Really?!? So Shader & DirectX versions and have nothing to do with it?

That's nice....
It is nice to realize that my neighbors did this interesting "magic" thing.

The real question is what will M/B companies do with it.
They can simply avoid using it because of GPU makers pressure.

But i definitely buy one :)


hmmmmmm, if this thing works as suggested, could it put an end to, 'mines bigger than yours' syndrome ??

errrrrrrm i dont think dammit and graphzilla will like this one little bit (time for lucid to dissapear into a black hole ??) shame ya cant use red AND green cards together O_o

lucid reckons ray tracings gonna be eazy peazy too, time to dig out those 3dfx voodoooooooooos ?? :O)
If this even comes close to how you just explained it, this is GOLD. This could do away with the pesky SLi integration altogether. Will it be out in time for a few X-58 manufacturers, who may be biding there time, to integrate? I just wished you mentioned the price point. If it does what you say it does, without breaking the bank, we have a winner.
i dont believe you ...sounds way too good to be true...just like all the SLI hype when that came out..sounds like total marketing guff

believe it when i see it
no doubt that if this works the way it is supposed to it will be the holy grail of multi GPU for more then just graphics acceletarion, this is essentially the ultimate in split load parallel processing and can be applied to more then just graphics rendering, i would reckon that this would work far more efficiently then SLI and Crossfire and no doubt a mobo with one of these in with the same configuration will have a far higher score in benchmark testing, but no doubt if it does we will end up with AMD or Nvidia Tryign to buy the company and integrate it into their Crossfire or SLI tech
I really like it when a little unheard of company comes up with an awesome idea like this. I just hope it works better than anything DAMMIT or nVidia have come up with... I swear one of the big selling points of CF originally was that it could intelligently split the frame between gpu's. I wonder what the difference is? Is it just better because it's on the mobo itself?
This is really awesome and I really hope it works as advertised. I also hope Lucid doesn't get bought out by one of the big boys.

I wonder why it took a small no name startup to come up with this tech, when nVidia and ATI have been trying to get it right for years. It's probably too early to say this but what the heck, congrats Lucid.
...they'll get bought. Probably by AMD/ATI, as they are in less trouble now than Nvidia, and Intel probably doesn't care enough.

...then again, they might be fodder for someone like Via or SiS who needs a bit of a boost to become relevant (again).

I find it highly unlikely that, if the product does what it says on the box, that this little company will be independant for long.
Well, the chipmaker who have no Multi-GPU solution right now. If you check the inverstors on the Lucid Logix website, you will see Giza, Genesis (2 well known investors in IT tech) and finally, Intel Capital. It add alot of serious to the news and the Hydra project...
Windows only allows one video driver to work at a time.
It's cute to throw around words like magic but that does not explain how the drivers are suppose to deal with this, and who makes those drivers, you can't for instance use many functions available on a dx10.1 class GPU on older dx9 or dx10 cards (or dx10 on dx9 etc.), meaning you cannot draw half the screen with one driver and the other half with another, in fact not even the same DXVA dunctionality can be counted on or used, then there's the overlay surface issues if you try to use it for video.
It might work somewhat with corporation of the makers of the graphics card (nvidia/AT) to use it to offload non-graphical stuff, but if anybody believes this will 'magically' work with any card combination he/she is a not too informed person.
Now we can finally see what crysis looks like at the highest quality settings.
About bloody time. I've always thought this is the way it should be done. It seems silly to do it any other way!
...but will it play Crysis!!!!

Just kidding. If this works the way its being hyped it'll be the defacto standard for all gaming motherboards within weeks of its release.

The idea of throwing a couple of $120 8800GT cards in a hydra board and having a GTX280 beater is outstanding.

I see this as a huge boon to Nvidia and ATI because now people won't be as hesitant to upgrade while they are waiting for "the next big thing".

It's alot easier to commit to a purchase when you know that it'll still be worth something as a booster card on your hydra motherboard down the road.

Dear lord I hope this isn't all a bunch of hype. This would be a revolution for both budget gamers and "at all costs" benchmark zealots.





Here BIG step up from Scan Line Interface, where one or other card or split top/bottom or interlaced one line then other line from other card is your choice. THIs is Monster, Doing Everything Anywhere it can find space.

Best part is, if one card can only do 25% of load, whole process does not hae to slow down, it can break job down according top cards ability. However, how variant can identical cards be?

Smoother its organized better.
drashek