VizWorld.com

Tech Report has a great roundup of multi-GPU solutions from NVidia (SLI) and ATI (CrossFire) across multiple pieces of software and multiple graphics cards, pushing the peaks of performance to their limits across a variety of uses.  To anyone who’s followed graphics cards, the results shouldn’t be a surprise:

One impression hasn’t changed since our value roundup: the latest GeForces tend to be better values at present than AMD’s 5000-series Radeons. That’s a clear reversal of fortunes since the GeForce GTX 400 series’ somewhat underwhelming debut, when the GTX 480 was really no faster than the Radeon HD 5870. The Fermi architecture still quite new, and Nvidia has extracted enough additional performance through driver tuning to put the GTX 470 on equal footing with the 5870—and the GTX 480 well ahead. That’s true across a range of games, not just those where the GeForces have an apparent advantage, like Metro 2033 and Borderlands. The addition of the cheaper and more architecturally efficient GeForce GTX 460 has further solidified Nvidia’s value leadership.

The ATI cards do seem to beat out the NVidia cards when you start looking at 3-card solutions, but that’s still rather rare.  To those who said MultiGPU’s were unlikely, here’s another step towards their inevitability.

SLI vs. CrossFireX: The DX11 generation – The Tech Report – Page 1.

Hardware ati, benchmark, multigpu, nvidia

Taiwanese company Manli has just introduced a new ATI card called the HD5970, which consists of two fully functional HD5870′s and 2G of GDDR5 memory on a single card.

Video consists of two GPUs HD5870 and is equipped with 3200 pokovymi processors and 2048 MB of memory GDDR5. It shows good results when working with 3D-graphics, fully supports Microsoft DirectX 11 with Shader 5.0 and OpenGL 3.2, and, of course, fully compatible with Microsoft Windows operating system 7. DirectX 11 uses stream processors and directly accesses the CPU and graphics core – multi-threaded applications faster and more efficiently. In addition, Manli HD5970 supports dynamic rendering ATI HDR, creates 3D effects

The translation is a bit clunky, but you get the idea.  It’s cards like this that blur the definition of “Multi-GPU”, and hence we say Multi-GPU is an inevitability.

via Technology News: Manli presented a dual-chip graphics card HD5970.

Hardware ati, manli, multigpu

[H]ard|OCP has taken delivery of the MSI Big Bang motherboard which has the Lucid Hydra. Lucid is a chip designer company with funding from Intel. The Hydra chip is an independent solution to allow multiple GPUs to render scenes in games. This means that you are no longer dependent on SLI from NVidia or Crossfire from ATI. If you are using Windows 7, you can use both NVidia and ATI in a multiple GPU configuration, although there are caveats with that. Hit the link below to read more about it.

via: MSI Big Bang’in with Lucid Hydra

Hardware ati, hydra, lucidlogic, multigpu, nvidia, video game

Some prototypes of the Lucid Hydra 200 Multi-GPU solution have been shipped to reviewers, and the results are promising for an early build unit.

After running Lucid’s test bed through an assortment of tests in several graphics configurations, we have a better idea on how well the Hydra 200 performs. Using the components available to us, we saw impressive scaling in most instances, which peaked at 85% in dual-GPU mode. The mixed ATI / NVIDIA combo ran pretty well and fell within the expected range of performance throughout testing.

It remains to be seen if Nvidia will allow something like this to exist in the market, they’ve previously stated their intent to intentionally disable it at the driver level.  If it comes to market, it shows huge potential tho.

via Lucid Hydra 200 Multi-GPU Performance Revealed – HotHardware. and PCPerspective

Hardware benchmark, hydra, lucidlogix, multigpu

Hess Corp., a global independent energy company, had a problem.  Their visualization software and their simulation software ran on separate machines and separate operating systems (Windows and Linux) and moving the massive amounts of data between them was becoming a problem.  They turned to Nvidia for a rather unorthodox solution, using Parallels to emulate one OS on top of another.  Thanks to NVidia’s new MultiOS GPU Support, it wound up being a perfect solution.

“Instead of running applications on different machines, and moving data back and forth over the network, SLI Multi-OS radically streamlines users’ workflows and instantly enhances productivity,” said Jeff Brown, general manager, NVIDIA Professional Solutions Group. “We believe Hess’ use of fully virtualized workstations will makes their employees more efficient, while lowering the total cost of ownership.”

Get all the details after the break.

Read more…

Science hp, multigpu, nvidia

Users of OpenInventor, the popular scene graph technology, will be happy to know that the latest version (8.1) adds in support for multi-GPU solutions like the NVidia Quadro Plex.

“This innovative package makes it possible to manipulate and access massive scenes in a practical, efficient way,” said Jean Bernard Cazeaux, CEO of VSG. “An application powered by Open Inventor 8.1 with CompleX running on a Quadro Plex system delivers a smoother, faster frame rate, and offers a much more interactive experience for engineers and geophysicists when viewing a seismic dataset of 100 million triangles. Without it, the performance would be unbearably slow.”

They’ve implemented it via NVidia’s CompleX application acceleration engine, so it’s optimized for NVidia hardware.

via NVIDIA and VSG Accelerate Oil & Gas Exploration.

Hardware, Science multigpu, nvidia, openinventor, vsg

LucidLogix’s Hydra system, the vendor-agnostic multi-GPU hardware system discussed previously, is finally coming to market in a new motherboard from MSI.

There are three versions of the Hydra 200: the LT22114, the LT22102 and the LT22114. The only difference between the chips are the number of PCIe lanes. The lowest end chip has a x8 connection to the CPU/PCIe controller and two x8 connections to GPUs. The midrange LT22102 has a x16 connection to the CPU and two x16 connections for GPUs. And the highest end solution, the one being used on the MSI board, has a x16 to the CPU and then a configurable pair of x16s to GPUs. You can operate this controller in 4 x8 mode, 1 x16 + 2 x8 or 2 x16. It’s all auto sensing and auto-configurable. The high end product will be launching in October, with the other two versions shipping into mainstream and potentially mobile systems some time later.

It’s an interestingly priced addition as well, running about $1.50 per PCIe lane, which means each x16 slot on the system adds an extra $24.  The high end Hydra chip support 48 lanes (3 PCIe x16 slots) for about $72.

via AnandTech: Lucid Hydra 200: Vendor Agnostic Multi-GPU, Available in 30 Days.

Hardware hydra, lucidlogic, msi, multigpu

Recently, Ars Technica finally found the JPR report on Multi-GPU from the beginning of this month, and smashed it as being pretty unlikely due to manufacturing constraints and consumer interest.  Well, I disagree and in this feature I’ll lay out why I think that Multi-GPU penetration of 30% is not only likely, it’s inevitable.

Read the full article and chime in with your thoughts after the break.

Read more…

Science feature, Hardware, jpr, multigpu

LucidLogix is back in the news with a new chipset called ‘Hydra’ which offers a completely new way to support multi-GPU’s in systems.  What makes it different than SLI and CrossFire you may ask?  Well, first you need to understand one of the major drawbacks of those technologies:

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.

The “Hydra” system works by dividing the scene into discrete objects, such as segmenting a scene into floor, walls, characters, weapons, particle effects, etc.  These objects can be distributed among the various GPU’s, and then re-composited by the Hydra chipset.  It’s a far more scalable solution than SLI or Crossfire, but puts additional load on the software to properly segment the scene.

Will it take-off? Who knows.  As multi-GPU becomes more commonplace as an alternative to higher-end video cards (Moore’s law in action, if you can’t make `em faster, just make more of `em) solutions like this will keep popping up.

via Multi gpu tech lucid to take on graphics giants – The Inquirer.

Hardware gpu, lucidlogic, lucidlogix, multigpu, sli, xfire