Articles tagged with Nvidia

Nvidia and mental images are reaching for the Cloud to offer ray-traced rendering over the web using stacks of GPUs (Graphics Processing Units) instead of CPUs. Set for official launch at the end of November Nvidia’s RealityServer 3.0 platform will enable architects, automotive engineers and product designers to send 3D scenes up into the cloud with the rendered results streamed back over the web. The major sell for this is significantly reduced rendering times, but the tech will also be able to stream interactive 3D to any web connected device including mobile devices - though of course bandwidth will be an issue.

The platform is highly scalable, and more users can be serviced simply by adding more GPUs. Nvidia is already talking to a number cloud computing providers and expects to announce partnerships with several of them later this year, one of them being Amazon EC2 (Elastic Compute Cloud). The cost of cloud-based deployment is expected to be less than one EURO per hour.

While the Cloud computing aspect of the technology is sure to dominate the headlines, of equal interest is the fact that RealityServer 3.0 can be deployed within the confines of a firewall, not only as a GPU-based ‘render farm’ to serve up rendered scenes in double quick time, but also as a means to distribute interactive 3D graphics throughout the enterprise.

The background to this technology is Nvidia’s CUDA programming architecture that enables Nvidia GPUs to carry out computationally intensive tasks usually reserved for CPUs. CUDA was used to devise a new GPU-based rendering mode called iray, which is based on mental images’ mental ray 3.8 rendering engine. This is different to most rendering technologies which rely on CPUs to do the calculations.

On the hardware side, RealityServer consists of multiple Nvidia’s GPGPU (General Purpose GPU) Tesla cards, which are used to render out the scenes plus a few CPUs, which are really just used for housekeeping, says Nvidia.

The technology is already primed up to be exploited a number of 3D CAD companies. There are over ten major CAD applications that already use mental ray, including Autodesk (3ds Max, Inventor, Revit), SolidWorks, Dassault Systemes (CATIA), and most recently PTC (Pro/Engineer Wildfire).

The critical technology here is mental ray 3.8, which is due for release later this year and will enable GPU-accelerated mental ray rendering for the first time. Once these vendors implement mental ray 3.8 into their core products, they would have all the tools to hook up to RealityServer, says mental images, but for some CAD software, particularly the more mature products that carry a lot of ‘architectural baggage’ the implementation would not be trivial. That said, mental images told DEVELOP3D that development is already underway at many CAD companies and it expects to see applications supporting RealityServer next year.

While mental images was unable to name all the names it did confirm that all of the aforementioned CAD developers are already working on systems that would allow them to virtualise their applications or to at least have server-based collaborative solution directly connected to their applications. As a result the company is confident that this technology is well placed to take a lot of work off the CAD developers’ plate as they are essentially offering them a whole suite of tools to get started faster instead of doing everything themselves. mental images also disclosed that Autodesk showcased the technology at a conference in Munich, Germany only yesterday.

In terms of the actual rendering technology RealityServer is a progressive renderer, so users are able to get a good idea of the final render in seconds or minutes, even though the final rendering may take hours. For comparative render times between CPU and GPU-based solutions it was hard to draw mental images on exact figures. However, the company did provide an example of an architectural scene that took 45 mins to render on a four Tesla cluster system and 8-10 hours on a more traditional four core CPU-based system. That said, it was wary of comparing apples and oranges as the scenes were not identical because the GPU renderer is slightly different from the CPU renderer in terms of shading technology. The company did say that it that would be providing benchmark results from customers next month and the early results are encouraging.

While for most CAD uses the emphasis is likely to be on using Reality Server as a rendering server, mental images was keen to point out that it also provides a platform on which companies to build applications that utilise the technology in different ways. In the automotive sector, for example, it is already working with a number of manufacturers on projects to develop and enhance their in-house design / review pipelines. A dedicated car paint shader is also in development and will be released early next year.

For those that wish to set up their own facility there are three different packages. In true American style there is no small - instead just a M, L and XL. Medium is a 2U rack mounted system with 8 Tesla GPUs and is suitable for smaller architectural offices and product design teams with 10s of concurrent users. Of course, this depends on the intensity of use and some customers may need to dedicate four GPUs for a single task. The ‘Large’ package features 32 Tesla GPUs for 100s of concurrent users, while ‘XL’ features 100 Tesla GPUs for serving 1,000s of users over the web.

Nvidia is still working on overall system costs, but with a single Tesla cards costing in excess of 1,000 EUROS one may speculate that a medium system would cost around 15,000 - 20,000 EUROS just for the hardware. On the software side, however, customers should expect a one-time licensing cost of 2,000 EUROS plus 20% maintenance per Tesla card.

From complex architectural visualisations and 3D city modelling to product design and automotive styling, the CAD-centric target markets for RealityServer are huge. And with mental ray already the rendering engine of choice for most major CAD developers, one may speculate that it’s only a matter of time before RealityServer becomes a widely supported platform for CAD.

What makes this technology particularly interesting is the fact that it is designed to use GPUs in the Cloud and not CPUs, but this is also a current barrier to deployment. None of the large Cloud service providers currently offer GPUs in their facilities, but Nvidia expects this to change early next year. This coupled with the expected release of RealityServer-compatible CAD products should make 2010 a very interesting year for rendering in the Cloud.

Over the past couple of years there’s been an incredible amount of hype surrounding the use of GPUs (Graphics Processing Units) to perform computationally intensive tasks usually carried out by CPUs (Central Processing Units).

The thing is there’s been very little activity on behalf of the CAD/CAM/CAE software vendors, with most action coming from niche areas of finance, science, and oil and gas – so little action in fact, that I’d started to wonder if we’d ever see this tech appear in product development workflows.

With this in mind I was more than a little excited to hear today that Autodesk is using GPU acceleration in its Moldflow Insight 2010 application to speed up the simulation of plastic flow in injection moulded components. The technology is based on Nvidia’s Cuda parallel processing architecture, which is supported by Nvidia’s Quadro FX 4800 and Quadro FX 5800 GPUs, and the development is said to have resulted in more than a 2x performance increase.

While this is a significant increase, it’s not in the order of 10s or 100s – factors that have often been bandied about when talking about the benefits of using GPUs over CPUs. The significant thing here though is that Autodesk’s Moldflow development is the first implementation from a mainstream CAD/CAM/CAE vendor. From speaking to Nvidia and its rival AMD over the past year, both companies have maintained that GPGPU (General Purpose GPU) technology is very much on the roadmap for a number of CAE (Computer Aided Engineering) software developers, but until now both companies have been unable to name names.

Not all computational tasks can be offloaded from the CPU to the GPU, in the same way that not all computational tasks can be accelerated by multiple CPU cores. Simulation and rendering, however, are commonly referenced. For example, because a lot of CFD (Computational Fluid Dynamics) code is highly parallel and scales well over multiple CPUs, this should theoretically translate well to GPGPUs, which feature a massive array of parallel processors, in the order of 100s.

Only time will tell how important a role GPUs will play in accelerating computational tasks for CAD/CAM/CAE software. Most software developers have been very tight lipped about their plans, but I expect with Autodesk’s lead, the next six months will deliver some interesting news.

The next issue of DEVELOP3D is out for your enjoyment and we’ve got a wicked issue in store for you this month. Our cover story is about Marin Bikes and how they use Autodesk Inventor to develop their range of cycles. One thing struck us about this story, along with many others we run every month, is the passion behind the products, and the shear bloody enthusiasm people have for their job. As Jason Faircloth, Marin’s product manager and bike designer commented, “Working at Marin turns a hobby into an income producing job.”

Alongside the story, we’ve worked with the folks at Autodesk to give away a pretty sweet little deal. In return for some information and your thoughts on simulation and analysis, we’ll enter you into a competition to win the product on the cover – a pretty sweet Point Reyes Commuter bikes (worth a grand). If you haven’t registered for the magazine or PDF, visit here to reg up and enter. If you’ve already registered, then visit here and all you need do is enter the email address you registered with and enter the competition.*

Elsewhere we take a look at how Senz used SpaceClaim to help design a breakthrough product is a pretty static marketplace. How Hornby designs its Scalectrix cars based on real world data. Of course, no DEVELOP3D would be bunch of reviews for you. We take a look at what’s new in Solid Edge with Synchronous Technology 2.0, Inventor 2010, the new SpacePilot Pro, a bunch of new Graphics hardware from AMD and Nvidia.

I’m also delighted to announce the latest contributor to join the Team, Joe Moak from formlovesfunction.com who’s discussing assembly constraints design and how early decisions can influence manfacturability. Joe will be writing us his thoughts on a bi-monthly basis or more depending on how busy he is designing awesome stuff.

* Those inevitable Terms and conditions: This competition closes on 31st July 2009 and is only open to those over the age of 18. Entry into the competition and acceptance of the prize constitutes permission to use their name inside the magazine. The winners will be selected at random by X3DMedia Ltd. No purchase necessary. Only winners will be contacted personally.

#4: It’s been a frustrating few days for me. Not only because I had to leave sunny LA at the tail end of last week to return to wind and rain in London, but I got to see all manner of exciting technologies at HP’s global workstation event, which I’ve been unable to write about until now due to embargo restrictions.

One of these technologies is HP’s brand new workstation family, the Z Series (pronounced Zee Series by those the other side of the Atlantic). Comprising three models – the Z400, Z600 and Z800 – the new machines are something to get quite excited about. This is not just because of Intel’s new, incredibly powerful Xeon 5500 series processors, but because HP has completely re-thought the workstation and the way it can be serviced, upgraded and generally taken to bits.

To my mind, only Apple and Sun have delivered similar levels of serviceability offered by HP’s new Z series, which is completely tool-less in design. For its demonstrations, HP focussed on its top-end Z800 showing exactly how easy it is to service and maintain.

HP’s New Z800 workstation, can be taken to bits in minutes – and look, no wires!

The whole system relies on clearly marked green levers and clips to give users visual clues as to how to take things apart. HP has done this before but never to this level of simplicity. First of all there are no wires inside the machine! These are all routed behind the back of the motherboard – not only to make things easier, but not to interfere with the CFD-optimised airflow throughout the machine. Even the dongles used to power high-end graphics cards, such as the Quadro FX 5800, are routed behind the motherboard and neatly clipped away when not in use.

Everything in the system, from graphics cards to hard drives and even motherboard, feature innovative mechanisms for easy and tool-less removal. Components that rely on power, such as the power supply and memory fans, use blind mate connectors, which are funnel-shaped to guide the components into place.

The power supply itself has been completely redesigned, and runs the whole length of the chassis so it can take in cool air from the front of the machine and as a result generate less fan noise. User maintenance and diagnosis has also been improved and should the workstation develop a power problem the supply unit can be removed simply by pulling it out by its handle. Then plugging it into the mains will help the user verify the cause of the problem. If the supply is indeed faulty the green light on the side of the unit will not come on. HP will then ship out a replacement power supply which can be easily installed by the user.

In its quest for even quieter machines HP has also innovated in its cooling mechanisms. Specially designed injection moulded ducting means each CPU in a dual socket machine receives fresh air from the front of the workstation reducing the load and improving acoustics. Dedicated fans for each memory bank also run independently from each other to minimise noise.

For those craving even fewer decibels, a liquid cooling option will also be available this summer and new Intel solid state hard drives, which not only run quieter than traditional physical drives, but consume less power and produce less heat, will also be introduced soon. These will also increase performance for those that access a lot of data, frequently.

Power saving

HP made some bold statements about power, claiming that on average its new workstations consume 35% less power than similarly configured machines in its previous generation. In terms of power efficiency, all of its power supplies now surpass the 80plus initiative and are now 85% efficient. The redesigned 1,100W unit in the Z800 is even 89% efficient.

HP has also done a lot of work on power saving and a new hibernate mode called HP WattSaver reduces power consumption to 0.8W, compared to the 2.1W in its previous generation workstations.

In addition, HP claims that all HP workstations are to 90% recyclable by weight and the HP Z line is registered as Electronics Products Environmental Assessment Tool (EPEAT) Gold, the highest rating available.

DesignWorks was instrumental in the development of the innovative new chassis for HP’s new Z Series workstation. Foam protoytpes pictured.

Industrial Design

HP engaged BMW DesignWorks as part of its industrial design team and to do the global customer research into what its customers want from workstations. In addition to the totally tool-less design to transform the way users re-configure systems, the Z600 and Z800 feature integrated handles making it easier to move them around on site. Also, because the workstations don’t have rubber feet, they can be slid around on carpets and tiled floors.

Styling was also high on the agenda and the ‘design language’ developed by DesignWorks expresses clean lines and premium materials. The grill that runs down the entire front of the Z600 and Z800 was made possible by incorporating a slot loading optical drive and brushed aluminium side panels also feature. Finally, with its injection moulded components, the machine also looks elegant from the inside, as opposed to the usual mess of wires and clips. It’s a welcome change from HP’s ‘xw’ chassis, which was getting a bit long in the tooth and while the Z Series has a refined look, it’s still no Mac Pro in terms of character.

Alec Bernstein, Senior Director, Strategy, Research & Strategic Partnering, DesignWorks USA, also explained how the Z800 applies a lot of the principles of green design. For example, minimizing the amount of screws and fasteners can have a huge impact on the environment as screws travel more than any product in the world as they are mined, manufactured and then shipped globally.

The Z family

HP’s new Z workstation series feature three models, the Z800, Z600 and Z400. The entire family is based around Intel’s new Xeon processor 5500 and 3500 series (codenamed Nehalem), which feature integrated memory controllers and Turbo Boost Technology.

The Z400 is a single socket machine taking the price/performance position in the range (though HP will continue to produce its Core 2 Duo-based xw4600 for those that want an even lower entry-point). The Z600 is a compact dual socket workstation, but for ultimate expandability the Z800 offers up to 192GB memory and 7.5TB of internal storage.

All workstations are armed with 1,333MHz DDR-3 memory and new professional graphics solutions have been introduced across the line. These include Nvidia’s Quadro FX380, FX580, FX1800, and FX3800, which join the established FX4800, and FX5800 in the high-end machines. The choice of AMD graphics cards has also been expanded over previous generation workstation families with the ATI FirePro V3700, V5700 and V7750 all offered as standard options.

Conclusion

With the introduction of the new Xeon 5500 Series, it’s an exciting time for workstations in general, because performance is about to go through the roof, particularly when users are multitasking or running multithreaded applications.

However, while this level of performance will be seen across new product introductions from all the major workstation vendors, what makes HP stand out from the crowd at this moment in time is its innovative approach to chassis design, which looks set to make the servicing of workstation an absolute breeze. Check back in the coming months where we plan to get our hands dirty with full reviews of the systems.

With two of these new Nvidia Quadro FX3800 graphics cards you can run Windows and Linux at the same timer on the same workstation with little or no drop in 3D performance

On what is arguably the biggest day for workstation-class hardware releases for some years Nvidia has unveiled its new top-to-bottom line of Quadro FX professional GPU solutions, many of which will soon be available in new workstations from all the major manufacturers.

Joining the Quadro FX 5800 and Quadro FX 4800, which were announced last year, are the Quadro FX 380, FX 580, FX 1800, and FX 3800. Apart from promising better performance than its previous generation cards, one of the most interesting new features is the introduction of Nvidia SLI Multi-OS, a technology that enables workstation virtualisation.

Supported in the Quadro FX 3800 and above, SLI Multi-OS works in association with Parallels Workstation Extreme virtualisation software and Intel’s VT-d technology, assigning both the host and guest virtual machine its own dedicated GPU and boasting close to 100% the 3D performance that you would get from running each OS on its own workstation.

Click here for my first experiences with Parallels Workstation Extreme virtualisation software

Dell officially unveiled its new range of workstation-class systems today with the launch of the Precision T3500, T5500 and T7500. Like all new workstations being announced this week, the new Precision family is based on Intel’s Xeon 5500 Series (Nehalem) architecture, which offers incredible power for multi-threaded applications; particularly those that take advantage of HyperThreading, such as 3ds Max and HyperShot.

Like most of the major workstation vendors, Dell has bypassed Intel’s Core i7 platform for its new Precisions, instead waiting for ECC (Error Correcting Code) memory for better accuracy, which is not supported on Core i7. The new Xeon platform also, uniquely, supports Direct Cache Access (DCA) which enables the cache of inactive cores to be accessed by those that are active.

Dell has expanded its range of graphics options with new cards from Nvidia including the Quadro FX 580, FX 1800 and FX 3800, but has also increased the number of AMD ATI FirePro cards it carries in the range with standard options available on the FirePro V3750, V5700 and V8700. Up to two Quadro FX4800 and FX5800 are available in the high-end T7500.

Acoustics has been a major design concentration for the new Precisions with low duty fans aiding the CFD (Computational Fluid Dynamics) designed airflow. Dell has also done a lot of research into ‘what if’ scenarios, modelling what would happen if one vent was blocked off, for example.

In terms of machine positioning, the single socket T3500 will take up the entry-level role, but Dell will continue to offer the Core 2 Duo-based T3400 for those on incredibly tight budgets. The mid-range T5500 is a particularly interesting machine, specifically because Dell has managed to pack so much technology into such as small chassis. Dell gave DEVELOP3D a sneak preview of the machine and we were astounded by the engineering that has gone into this, with the second processor and memory located at 90 degrees to the motherboard on a riser card. With such a small footprint, however, the T5500 is fairly limited in its expandability and this is where the T7500 fits in with capacity for up 192GB RAM and a ridiculous amount of hard drives. It also includes an on-board SAS controller.

While the new systems will ship with Windows Vista by default, Dell will continue to offer Windows XP downgrades (with XP recovery disks) as well as Linux. However, thorough its custom factory integration program Dell can supply workstations with XP pre-installed, and it is also possible for customers to supply disk images for Operating System, network and applications, which Dell installs prior to shipping.

Dell is also in the process of developing a new Flash-driven workstation advisor website, which is designed to make it easier for customers to choose workstations according to which applications they use. This is coming in Q2 2009.

Look out for review of the new Precision range soon.

Today Lenovo became the first of the dedicated workstation manufacturers to announce its new generation Intel Xeon (Nehalem) based workstation range. Pipping Dell and HP to the post, Lenovo’s new ThinkStation S20 and D20 feature single socket and dual socket versions of Intel’s new Core i7-based Xeon chip.

In addition to offering both AMD (ATI) and Nvidia graphics cards, the big news is that Lenovo is pushing Nvidia’s Tesla GPU platform to supplement the jaw dropping performance of Intel’s new chips. For those that don’t know, Nvidia’s Tesla cards look like a graphics card and feature virtually the same technology as a graphics card, but are designed specifically to carry out compute tasks usually done on the CPU. Like all new technologies though, we are still waiting for the applications to come, most likely in the areas of simulation and rendering.

Elsewhere, Lenovo is boasting some pretty impressive green credentials, claiming that both workstations use 50% recycled content.

Look out for a full review review soon.

www.lenovo.com

Nvidia has again upped the ante in the professional graphics sector with the launch of a new monster of a board, which is likely to set you back around £2,000.

With 4GB RAM, the ultra high-end Quadro FX 5800 has the biggest amount of memory on any graphics card, doubling the previous 2GB record held by AMD’s ATI FireGL V8650. However, this amount of memory and the high-level performance that this card boasts, is only likely to appeal to a small proportion of users, with nVidia touting the medical imaging, oil and gas, and automotive styling sectors, as key markets. Additional interest is likely to come from high-end CAD and design visualisation users with products such as NX and 3ds Max, particularly if these companies need one or two top-end workstations to complement their mid-range machines.

Built using the nVidia’s parallel CUDA architecture, the Quadro FX 5800 is also set to play a key role in the company’s drive to move complex computational problems away from the CPU and onto the GPU (Graphics Processing Unit).

While nVidia has done much to promote this technology, which is specific to nVidia hardware, little progress has been made in the mainstream CAE sector, with the majority of developments coming in the more niche areas of science and finance.

One development that should help bring CUDA more into the mainstream, is the launch of nVidia’s new Quadro CX card, which is a dedicated graphics accelerator for Adobe Creative Suite 4. With the Quadro CX, CUDA is used to encode H.264 videos in Adobe Premiere at what is claimed to be lightning-fast speeds. The card also powers real time image manipulation in Photoshop for the first time, though this feature is actually supported by all OpenGL 2.0 compliant graphics cards.

While the £1,000+ price tag is likely to put off all but the most power hungry users of Creative Suite, nVidia Quadro supplier, PNY, told DEVELOP3D that this card would also deliver excellent performance in 3D CAD/DCC applications. This could make it an attractive proposition for design visualisation specialists who use Photoshop and Premiere alongside products such as 3ds Max.

At the other end of the spectrum, nVidia’s Quadro business is also concentrating on the lower end of the market with the launch of the Quadro FX 470, the company’s first integrated professional motherboard GPU, and Quadro FX 370 Low Profile (LP), an entry-level Quadro graphics board for small form factor systems. While Nvidia has not yet signed up any of the major workstation OEMs for the Quadro FX 370 LP and Quadro FX 470, specialist workstation manufacturer CAD2 told DEVELOP3D that it was currently investigating the new technologies and hoped to be able to offer small form factor workstations in the New Year.

www.nvidia.com