|  | | Lucid Makes Multi-GPUs Easy |
Posted by Chaos
on August 20th, 2008, 11:58 AM
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'.
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.
You can read the entire article at The Inquirer.
Last edited by Regeneration; August 20th, 2008 at 06:04 PM..
| | | | 21 Comments | | | Cooool. Nvidia needs to buy this company - quickly! | | | | That's the only purpose of Crossfire/SLI IMHO. | | | | Quote:
Originally Posted by Syncroneyes  Cooool. Nvidia needs to buy this company - quickly! |
Yep its realy sound like super idea , multi gpu dream come tru. | | | | screw nvidia, what they need to do is keep their dirty paws off of this one | | | | Quote:
Originally Posted by Syncroneyes Cooool. Nvidia needs to buy this company - quickly! | Intel have already invested in this one....cool larrabee in crossfire/sli or whatever. | | | | I'm going to buy the add in card real quick when it's released cos you can bet one of the big players will buy them up and then only make it available with their chipset. | | | | I hope motherboards with this chip arent very expensive , any idea when will this be available to end users ??? | | | | Have read previously 2009....think it said early 2009....that was for release anyway....whether that will be the chip just for mobo manufactures or if that includes the add in card I don't know. |
Last edited by technogiant; August 21st, 2008 at 06:38 AM..
| Quote | | | | | "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."
So no add in card i think , only mobos with it. | | | | Quote:
Originally Posted by nCaine "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."
So no add in card i think , only mobos with it. | Your right I've just been re reading their info...it will be available on the mobo or on a graphics card that has several (up to 4) gpu's on it not as an add in card itself...damn it. | | | | The hydra engine is a system on a chip (SOC) seem's strange to put a SOC in an already existing system ( the PC)....why can't what it does be done on the PC? I guess it would be too slow running in software? |
Last edited by technogiant; August 22nd, 2008 at 10:31 AM..
| Quote | | | | | this must be more limited than they say, if you're playing a DX10 game, how the hell is your dx9 card gonna even understand the part of the rendering it's given? | | | | Also wondering if this is going to work with physx with nvidia cards....the hydra engine allocates the various graphics tasks going along the pcie bus to the graphics cards...but what about the physx tasks? They say it can work with gpgpu type tasks but will it do both at the same time? | | | | I was wondering the same thing that you 2 on my night shift . | | | | Ok here is some more detiled info`s. on Hydra . What about the possibility of combining an AMD and NVIDIA GPU to work together to render one image? This was my biggest hope after hearing the initial introduction. Apparently it won't be happening though as the operating system prevents multiple graphics drivers from running 3D applications at the same time. Since a 6800 and 9800 use the same driver from NVIDIA, both may operate in 3D mode without a hassle but combining AMD and NVIDIA just won't work. Let's hope for the future... How will tasks be distributed in greatly exaggerated card matchups? For example, in that 6800 and 9800 combination mentioned above? With the HYDRA Engine it is apparently very easy - the software and chip recognize the potential for each and distribute the appropriate number of tasks to each card. Maybe 5 tasks to 1 or something like that; the results are then combined by the HYDRA chip and sent to a single GPU for output. What DirectX versions are supported or will be supported and what about OpenGL? Right now, only DX9 is working though DX10.1 will be ready by the end of the year. With DX10 and DX11's implementations of multi-GPU data improving and adding to the HYDRA Engine technology will only get easier for team compared to the work they had to do on DX9. OpenGL is supported by the HYDRA Engine as well. Could this technology be applied to GPGPU work as well? Yes, though that is still far into the future. One area the team did say would be easily taken advantage of by their technology is ray tracing with its incredibly task-oriented workflow. Can HYDRA really be used with ANY graphics technology? Yep, Lucid said you could even setup some VIA S3 Chrome cards if you really wanted too...
Any below very intrestiong article about it. http://www.pcper.com/article.php?aid=607 |
Last edited by nCaine; August 22nd, 2008 at 06:06 PM..
| Quote | | | | | yeah it's all interesting stuff...was watching a video clip ealier "The Inquirers" summary of I think it was day 2 of IDF and the correspondent hinted strongly that hydra chips were going to be found on the X58 nehalem mobos....should mean both crossfire and sli without the later needing that pcie bridge chip. | | | | Something else that made me look twice in that article, the hydra chip takes in x16 pcie 2.0 lanes and outputs either two x16 2.0 lanes or 4 x8 pcie 2.0 lanes....has somehow doubled the pcie bandwidth?
quote
"From a purely hardware perspective, the HYDRA chip takes in a single PCIe x16 connection and outputs two full PCIe 2.0 x16 connections. Depending on the partner's implementation method, that could connect to two GPUs or split into four x8 PCIe 2.0 connections for four GPUs."
What does that mean for Nehalem? Hasn't the bloomfield got 32 pcie 2.0 lanes allocated to graphics? So could have two Hydra's? they would have to be linked somehow?....all mind bogling stuff and an interesting next 6 months. |
Last edited by technogiant; August 22nd, 2008 at 06:30 PM..
| Quote | | | | | Well considering the crossroads with current video rendering,lightstage, raytracing,Nehalem,Larrabee,physx,and havok......i would not expect to take anything for granted Techno....however i will say that i am gonna put a hold on a mobo upgrade until i see the results of these boards... |
Last edited by n3omatrix; August 25th, 2008 at 06:55 AM..
| Quote | | | | | I think that's a good idea n3o...there is such alot in the melting pot at the moment....its a time to just stand back and observe with anticipation. | | | | Be prepared to see the Lucid brand name dissappear quickly, people. | | | | Well if anyone buy's this company it will probably be Intel as they have already invested heavily in it.
I'd rather see them buy it than Nvidia or AMD/ATi at least then it won't be restricted to one graphics card as with physx....unless Intel restrict it to larrabee | |
|
