NVIDIA To Hold CES Press Conference on 4th January 2016 – Here’s What To Expect From NVIDIA
NVIDIA has announced that they will be hosting their CES 2016 press conference on 4th January 2016. As every year, NVIDIA at CES will be introducing several key innovations in the graphics sector while a main focus of their conference would be autonomous driving which has been the case since several previous conferences as they aim to deliver better products that drive deep machine learning. NVIDIA will officially host their CES 2016 event conference on 4th of January, 2016, live from Las Vegas.
NVIDIA's CES 2016 Press Conference Takes Place on 4th January - Here's What We Expect From The Graphics Giant
Every year, the biggest tech companies come together at CES and announce a range of new products for the consumer market since the event itself is focused at consumers as the name suggests (Consumer Electronics Show). Last year, NVIDIA introduced their Tegra X1 mobile SOC which was embedded into two products, the Drive PX and Drive CX which are boards featuring the SOC to drive digital cockpits and auto-pilot car navigation. Later on, the Tegra X1 was embedded into a range of Shield devices which include the NVIDIA Android TV and the rumored NVIDIA Shield Tablet which hasn't launched yet. AMD and Intel are also going to make a major appearance at the event. Last year, AMD showcased their first Freesync and Carrizo samples while Intel took things ahead in the mobility sector with their low-power Broadwell-U and Curie (Quark based) SOCs.
— NVIDIA (@nvidia) December 13, 2015
Now people should be expecting to hear more in the field of enthusiast graphics offerings from NVIDIA at CES 2016 but those announcements have shifted to GTC (Graphics Technology Conference) where NVIDIA is going to announce details on their next generation Telsa, GeForce and Quadro parts. However, we know there will be some rumors to address prior to NVIDIA's CES 2016 press conference and some products which have been alleged to make an apperance at CES 2016. The following list of products are what we expect to be announced by NVIDIA at CES 2016 (top-to-bottom order):
NVIDIA Autonomous Drive Modules
One of the sector which is a major driver for NVIDIA's revenue is Auto which amounts to as much revenue as NVIDIA's Datacenter business. In the auto sector, NVIDIA has partnered with major brands such as Tesla and Audi. The automative products enable smarter, more sophisticated advanced driver assistance systems (ADAS) and paving the way for the autonomous cars. These products include the Drive PX (Auto-Pilot Computer), Drive CX (Digital Cockpit) and Jetson Pro (Automotive development kit) which lead the way for fully automatic, self driven cars. We talked about this tech in detail in our exclusive article covered by Usman Pirzada. At CES 2015, NVIDIA announced two new modules (Drive PX and Drive CX) utilizing their latest Tegra X1 processor, and this year, we can expect new products based on the next generation Tegra SOC. Details are unknown but they are expected to be more powerful and would make cars more autonomous than ever before.
NVIDIA's Next Generation Tegra SOC
One of the most anticipated reveal of the event would be the announcement of a next generation Tegra chip. Like stated, Tegra has become one of the major attractions from NVIDIA at CES and just like every year, 2016 would likely see the launch of a new Tegra chip which will be the successor to the Tegra X1. The Tegra X1 SOC was based on the TSMC 20nm process node and featured the latest ARM and Maxwell cores. While we know little about the new Tegra chip, but rumors are that it would be featuring a new FinFET process while featuring brand new ARM cores and possibly an updated graphics core. One more thing we expect to see from the new Tegra chip is that it might feature Denver CPU cores which are custom ARM cores developed by NVIDIA for the Tegra SOCs. We know from previous reports that NVIDIA made a Tick-Tock approach to their Denver design as mentioned to Arstechnica when talking to a NVIDIA rep:
“The reason we did not use Denver on Tegra X1 is that decided to take a “tick tock” approach to launching our processors. Since TX1 is on a brand new 20nm process we decided to use off the shelf ARM cores. But Tegra K1 was on a well understood [28nm] process and hence we decided to use Denver on it. We still have Denver on our roadmap and will be using it in our future chips as processes mature.”
As for the graphics core, we don't know whether we would see an updated Maxwell core or Pascal GPU. My bet is that NVIDIA will be using their Maxwell GPU core since Pascal is still away from launch and expected to arrive in first half of 2016. Of course, we can't be so sure about which GPU would they feature on the new Tegra chip. The latest Tegra will probably be embedded in a range of new devices under the Shield and Drive brands which are also expected to make an appearance at CES 2016.
NVIDIA's Graphics Update - Pascal and Dual-GPU Maxwell Card To Get A Showcase?
One of the least excepted bits we expect NVIDIA to cover at their CES 2016 conference would be details on the next generation Pascal GPU. No doubt, Pascal means a big deal to NVIDIA and we are likely to get an update in the form of roadmap and specification details sooner or later but I personally believe that GTC seems like a more obvious platform/event for NVIDIA to announce their next generation GPU rather than CES 2016. But NVIDIA likes to play spy now and often so we can't be so sure and Pascal just might make an appearance at the conference. Details we know so far about Pascal include a massive FinFET based graphics processing unit with up to 17 Billion transistors, 32/16 GB HBM2 VRAM that will be pumping out 1 TB/s bandwidth and a brand new CUDA architecture aimed across the board, with focus on Tesla compute parts. At SC15, it was revealed that Pascal would end up with more than 4 TFLOPs of FP64 compute performance while the specific Pascal SKUs are already making the rounds on the internet.
The second major update we could see at the event is the Maxwell based dual-GPU card which has been doing the rounds on the internet since several months. As a showcase of power, both NVIDIA and AMD are known to offer dual-chip cards based on their flagship chips. AMD has already previewed their Radeon R9 Fury X2 in June and the card is expected to launch in Q1 2016. NVIDIA has not officially announced their cards but we could see an announcement at CES 2016 as they did silently announce the Titan X at GDC last year. There are also some conflicting reports that NVIDIA might have cancelled their dual-chip card in order to improve multi-GPU scaling and operations on their existing cards as major AAA titles don't have support for multi-GPUs (Batman Arkham Knight and Just Cause 3). DirectX 12 is expected to change this scenario to some extent but games featuring the latest API won't be available later on in 2016 so it makes little sense to launch a dual-chip card much late and close to Pascal cards.
The NVIDIA CES 2016 will be full of Automobile and Tegra showcases, enthusiasts who expect to see NVIDIA announce new graphics cards at the event should also tune in for some surprises (if NVIDIA has some in plan) otherwise 2016's GTC in April will be the main event to see for details on Pascal GPUs and beyond.