Sergio at SIGGRAPH: Day Five and Wrap-up
The fifth and last day of SIGGRAPH is always the one when you're already extremely tired, and the whole event ends early (some things close around noon, the exhibit ends at 3:30 and the last tech talks end at 5:30). Those issues, however, don't prevent you from having fun.
I went to the Fusion IO booth and checked out their NAND memory cards. If I had to pick one word to describe those cards, I'd say "wow." The Fusion IO is a card that you plug inside your workstation (just like any other pci-e card), and it can be used as either storage or hard drive cache, and basically it's used to increase your transfer rate for some applications.
As CPU and GPU power increases, hard drive speed hasn't really increased that much. This means that your computer could process way more data than what your hard drive is able to send, slowing down your workflow when working with video editing, image compositing, and similar applications. These cards are used to help solve the situation.
Due to the architecture, they are able to feed more data to the CPU, increasing your workflow, and thus, your productivity. For the demonstration, they launched Toxik and opened a 4K comp. If you're familiar with this scenario, you may imagine what happened when they hit the play button: playback rate from 0.5 to 1 fps, CPU barely used as it was not getting the data fast enough to process it. They then launched the similar comp, but using a Fusion IO card as cache for the application, and it all changed: nearly real-time playback, CPU used at max, and they could even apply filters while the sequence was playing.
There are different models, ranging from 64Gb to over 1Tb of storage capacity, depending on your needs. If you're like me, you would surely like to get the bigger card, but keep in mind that depending on your needs the bigger models may be overkill.
Studio|GPU was at the ATI/AMD booth, showing off version 2.0 of MachStudio Pro. MSP 2.0 has many new features, including GI, as well as integration with Mentalimages Mentalray and Pixar Renderman, making it a one-stop rendering solution.
MSP will now offer real-time rendering, as well as software rendering using either Mentalray or Renderman. If you're using Mentalray, you can also use the built-in MI shader creation tool to compile MI shaders. Those shaders can then be used for your stand-alone Mentalray renderer, or inside MSP.
MSP will include the new ATI FirePro 8800, a DX 11 card, and you will need DX 11 to run MSP 2.0 (it will not run on DX 10 cards).
Nvidia launched a new set of Quadro professional graphics cards (Fermi series), as well as a professional version of their 3D Vision system. The main difference between the 3D Vision and the 3D Vision Pro is that the latter one uses a high frequency signal, instead of the infrared one used by the 3D Vision. This means that artists will be able to use 3D Vision Pro systems in a facility, inside the same room, as each pair of shutter glasses synchronizes with its own emitter (on the current 3D Vision system, all pairs of shutter glasses inside a room will synchronize with the same emitter, so no more than one separate system can be used in the same room).
The Pixel Farm was showing their new node-based camera tracking software. Camera tracking software is usually very linear: first you detect features, then you track them, then solve the camera, test your solved camera and do any adjustments you may need, and then export. However, with a node-based workflow you can change and adjust your workflow depending on specific shots.
I visited the PixelActive booth to take a look at CityScape. Having used CityEngine before, I thought it would be a good idea to check on a different software that serves the same purpose. Even though they are both city creation applications, they work completely different, since CityEngine is more of a procedural approach, while CityScape is more "linear," so to speak (at least as far as I could see). I will check this software in more depth, but I can tell you the workflow is completely different.
SIGGRAPH is now over. I've said it before, but the feeling is always the same: I feel happy it's over and that I get to rest, but it's always sad since SIGGRAPH is just awesome. As I told a couple of people, "I feel exhausted, but extremely happy to be here," and I can't wait to be there again next year. See you in Vancouver. The face of an exhausted but happy SIGGRAPH attendee can be seen below :)
Terrence Masson said it was going to be the best SIGRAPH ever. If you attended SIGGRAPH, would you agree with that? I can't say it was really the best SIGGRAPH ever, although it was certainly awesome.
I was surprised to see that FJORG! was cancelled. I was also surprised to see the exhibit and job fair were so reduced (there were only a few companies at the job fair, and the exhibit was pretty much half of what I saw at my first SIGGRAPH). I did enjoy, however, the Sandbox, the Disney Learning Challenge and the SIGGRAPH Dailies, so some things added the counterbalance to others. I also missed the big SIGGRAPH banner that they usually put up outside the main lobby at the Convention Center.
This has also been the smallest SIGGRAPH I've been to. The ACM SIGGRAPH released the numbers, and while it's still a big number (22,000 attendees), it's less than what previous years (a few years ago there were around 30,000 visitors).
Some good things, some bad things, but as I said before, those don't make SIGGRAPH any less awesome. More pics below :)
Be sure to also read:
Sergio Aris Rosa [nemirc], is Sr. Staff Writer for the Renderosity Front Page News. Sergio discusses on computer graphics software, animation techniques, and technology. He also hosts interviews with professionals in the animation and cinematography fields. You can follow him on Twitter, and if you want to see what he's up to you can visit his blog.
August 2, 2010
Please note: If you find the color of the text hard to read, please click on "Printer-friendly" and black text will appear on a white background.