Archive Home arrow Tech Affiliates: arrow Tech News arrow NVIDIA nTeresting Newsletter - 8 April 2013
NVIDIA nTeresting Newsletter - 8 April 2013
Tuesday, 09 April 2013

NVIDIA nTeresting Newsletter - 8 April 2013

In this issue:

· GeForce 700M Series of notebook GPUs automatically maximizes consumer's notebook experience with the GPU Boost 2.0 for notebooks.

· GeForce GTX 650Ti BOOST redefines value under $200.

· Over 3,000 folks converged on GTC 2013 to cover all things GPU.

· NVIDIA had a strong showing of technology and platforms at GDC 2013.

Notebook GPUs Get Boosted to 700

On April 1st we announced the latest additions to our GeForce 700M line-up of notebook GPUs. Those GPUs include NVIDIA GeForce GT 750M, GeForce GT 745MM, and GeForce GT 740M in the performance segment and the GeForce GT 735M and GeForce GT 720M in the mainstream segment. These GPUs bring more Kepler goodness to our notebook product stack.

"On the whole, Kepler has been amazingly successful for NVIDIA, particularly in the mobile world. The bar for midrange mobile dGPUs was raised significantly with the GT 640M LE and above parts typically offering anywhere from 25% to 75% better performance than the previous generation, and that was accomplished along with reducing power use. It was NVIDIA's version of Intel's Core 2 launch, and the vast majority of notebooks with dGPUs seem to be using NVIDIA hardware these days."

They feature key NVIDIA technologies that automatically maximize consumer's notebook performance and experience, such as the new GPU Boost 2.0 for notebooks.

"The third, and perhaps most important, addition to NVIDIA's new cards is NVIDIA's GPU Boost 2.0, which offers dynamic performance optimization. GPU Boost 2.0 takes advantage of two underutilized aspects of PC performance. First, the gap between how much power a PC's components could potentially use and how much is actually being used, and second, the available thermal headroom-the amount of heat buildup that a laptop is designed to tolerate. By actively monitoring the power usage and available thermal headroom, NVIDIA's automatic optimization is able to ramp up the GPU clock speed for better performance while never exceeding a system's thermal or electrical limits."

More automatic goodness comes from NVIDIA Optimus technology.

"Each of these GPUs is based on Kepler and works with NVIDIA Optimus technology so if you're all about both next-generation power and long life in batteries, you'll be set!"

And they come with GeForce drivers!

"Much of that can also be attributed to NVIDIA's driver team, where Optimus support and usability still trumps AMD's Enduro alternative. AMD is still working to improve their drivers, but they're still not at the same level as NVIDIA's mobile drivers."

Last but not least, GeForce Experience automatically optimizes your games for the best performance and visual quality on your specific notebook and keeps your GeForce drivers up to date.

"The new NVIDIA GeForce 700M family holds a lot of promise for bringing GPU horse power to the mobile gaming arena with power optimizations to ensure the best balance between performance and battery life. Keeping in mind that these chips have yet to be independently reviewed, the Kepler-based GeForce GT 700M cards are shaping up to be a decent upgrade over previous-generation Fermi-based products. With the application-specific power boost and optimization technology inherent in GPU Boost 2.0, and the available gaming optimizations from the GeForce Experience, these new mobile GPUs will be a force to be reckoned with."

NVIDIA Optimus, GeForce Experience and the new GPU Boost 2.0 work so effortlessly they must be automagic.

A Sweet GPU at the Sub-$200 Sweet Spot

Gamers love a sub-$200 graphics card, and gamers looking to play this year's hottest PC games at a highly affordable price -- with in-game settings cranked up to high -- got their wish!

"The GeForce GTX 650 Ti delivers the best performance we've ever seen at the $150-160 price segment. It truly is phenomenal, and we are impressed at what has been delivered to gamers."

Best of all, the GeForce GTX 650 Ti BOOST GPU is available in 2GB and 1GB configurations and can be had for an affordable $169 and $149, respectively.

"We had to double check the price of the GeForce GTX 650 Ti with NVIDIA after experiencing the performance delivered. It seemed like the performance delivered for the price was too good to be true."

The new NVIDIA GeForce GTX 650 Ti BOOST GPU is based on the NVIDIA Kepler architecture and equipped with 768 NVIDIA CUDA cores. It delivers a wider memory interface, 60% more memory bandwidth and GPU Boost technology to give gamers 40% more performance over the original GeForce GTX 650 Ti GPU introduced last year.

"With one swift stroke, the company engineered a hostile takeover of the $100-$200 market, increasing graphics performance at any given budget in that space."

GTC 2013: Product Roadmaps, GRID Appliance, ‘Kayla,' Big Data Analytics, CUDA Python...and much more

Our 4th annual NVIDIA GPU Technology Conference (GTC 2013) was held in March at the San Jose Convention Center in California, and by all accounts this one was the biggest and best yet!

3,000+ attendees from all over the world were treated to over 400 sessions about the amazing new breakthroughs and advances enabled by GPUs.

This year, the sessions and news went well beyond the scientific, visual computing and technology fields normally featured at GTC. From GPU-powered automotive infotainment and driver-assistance systems and cloud computing, to computer graphics and the latest GPU-enabled movie and game effects, there was something for everyone.

Kicking things off with opening keynote on March 19, NVIDIA CEO Jen-Hsun Huang revealed a host of major new product and technology announcements:

First, he took the covers off NVIDIA GPU and mobile processor roadmaps, sketching out a future where mobile, desktop and supercomputer technologies intersect in powerful and surprising ways. Future GPU architectures include "Maxwell," the follow-on to the successful "Kepler" architecture, which will feature Unified Virtual Memory. Up next is "Volta," which will incorporate stacked DRAM to deliver 1 terabyte per second of memory bandwidth. Here's what 1TB/sec of blazing fast performance means for graphic designers, video editors, animators and gamers:

"... you could move the contents of a jam-packed Blu-ray disc through the chip in a mere 1/50th of a second. Consider that NVIDIA's Titan GPU - its latest high-end graphics card - can handle about 288GB/s, and you can only imagine the positive implications for a wide range of people depending on graphics processing horsepower..."

GTC attendees were all a buzz over the future Tegra mobile processors. First is "Logan," which will pair ARM-based mobile processor cores with Kepler GPUs to support high-performance PC and workstation features - such as PhysX, CUDA 5, and Open GL 4.3. And "Parker," which will join new 64-bit ARM compatible CPU cores with NVIDIA's next-generation Maxwell GPU.

"Thought the new Tegra 4i was the bees knees when it we saw it last month? Well, NVIDIA gave us a bit more info on the next steps in the Tegra roadmap, Logan and Stark Parker." It turns out that these next two mobile platforms will both utilize NVIDIA's CUDA technology, with Logan packing a Kepler GPU and Parker running a Project Denver 64-bit ARM CPU and a next-gen Maxwell GPU."

Next up was GRID VCA, the industry's first visual computing appliance that enables businesses to deliver ultra-fast GPU performance to any Windows, Linux or Mac client on their network. Fresh out of the gate, and accolades are already starting to stream in:

"Several NVIDIA customers and partners who have had trial runs with the GRID VCA gave it their highest praise."

GPU technology continues to make inroads into new markets to solve big problems. The latest is the field of big data analytics, with big-name enterprise and mobile application companies -- Shazam, Salesforce.com and Cortexica -- using GPUs to tackle big data analytics and advanced search for consumer and commercial apps.

"NVIDIA stretches Tesla GPU coprocessors from HPC to big data - ‘Anything a CPU can do, a GPU can do better'"

Jen-Hsun also revealed that GPU computing is now coming to Tegra processors, with the new ‘Kayla' development platform. Kayla roars with a powerful Tegra mobile processor and an NVIDIA Kepler-based GPU, supporting a powerful suite of technologies that have never appeared on mobile devices before - including CUDA 5 and Open GL 4.3.

"'Kayla' - a processing beast that, when it's ready for action, will be extra-tiny and extra-powerful beyond anything we're capable of today."

On the high-performance computing side, NVIDIA GPUs continue to help researchers advance breakthroughs across a range of scientific fields. The Swiss National Supercomputing Center (CSCS) is building a new petaflop-class supercomputer with Tesla K20X accelerators to help Switzerland's national weather service predict local and national weather patterns days or even weeks ahead of time with the highest degree of accuracy.

And, CUDA programmers can now tap into the widely popular Python programming open-source language to accelerate high performance computing (HPC) and big data analytics applications on GPUs:

"NVIDIA CUDA Supports Python - Programmers Jump With Joy"

Finally, no NVIDIA GTC keynote would be complete without some breathtaking, eye-popping technology demos. One of the stars of this year's keynote was ‘Ira,' a startling digital model of a human head that not only offered unparalleled realism - but was generated in real-time using commercially-available GPUs and an NVIDIA innovation we call FaceWorks.

Check out the video of Ira here: https://www.youtube.com/watch?v=5d1ZOYU4gpo&list=SPZHnYvH1qtOY0ZrWQgnQlj4dwGZ1pVgoj&index=2

Unable to attend GTC this year? No problem, you can view the keynotes from Jen-Hsun and other guests here: https://www.gputechconf.com/page/gtc-keynote.html. And check all of the news at the GTC Press Room: https://www.gputechconf.com/page/gtc-press-room.html.

GDC 2013: If They Build It, Then They Came

Game Developers Conference 2013 was held in San Francisco March 25 to 29. GDC is the world's largest and longest-running professionals-only game industry event and attracts over 22,500 attendees. It is the primary forum where programmers, artists, producers, game designers, audio professionals, business decision-makers and others involved in the development of interactive games gather to exchange ideas and shape the future of the industry.

NVIDIA was on hand to show off what makes them the leader in game computing across all platforms, from super phones to high-end PCs. Attendees had a chance to get their hands on Project SHIELD.

"Personally, I can't wait to snag one of these and see just how well it will stream my PC titles. Down the line, it would great to have remote streaming so you could get your game on outside of the house, but for the time being, that's just wishful thinking."

They saw NVIDIA GRID in action and played PC games on a phone, TV and on any number of gadgets.

"As you can see from the above video, cloud-based gaming is in NVIDIA's future."

Highlights of the NVIDIA GDC presence included the first public unveiling of Epic Games' spectacular Unreal Engine 4 Infiltrator demo:

"The high-end PC used at GDC was made up of a single NVIDIA GeForce GTX 680 card coupled with 16GB of RAM and a "recent" CPU, the developer said."

Check out the stunning video here.

We showed a level of Hawken that NVIDIA PhysX made fully destructible!

"GDC may be over, but there's still the clean up operation to be performed. Discarded bits of news lay strewn across the show's floor, waiting for us to scoop them up and put them on their digital shelf. What's that in the corner? A burnt-out mechanical husk - misshapen and trampled by excited IGF nominees - informing us that NVIDIA's APEX demo featured a Hawken map that utilised the destruction tech in multiplayer."

Check out this stunning video here.

Our Real-Time Destruction demo was also a hit.

"If this was a movie, the end of this video would pull back to reveal a round-table of world leaders staring dumbfounded. Then, NVIDIA's PhysX SDK Research Lead Matthias Müller-Fischer, would appear on their screen (possibly with a cat), point his omnipotent crosshair of ultimate destruction at Big Ben and start reading out the transfer details for his Swiss bank account."

Check out this stunning video here.

They also got up close and personal with our FaceWorks-facial expression technology

"NVIDIA hopes to get one step closer to eliminating that boundary with their new FaceWorks technology in the video above. There's lot's of technical speak until the 8.5 minute mark, so if all you want to see is eerily-rendered CG faces, skip to then."

Check out this stunning video here.

Overall, the technobomb we dropped at GDC 2013 left the industry...... yipe.... stunned.


 

Comments have been disabled by the administrator.

Search Benchmark Reviews Archive