At GTC Europe 2017 in Germany, NVIDIA today has announced the Drive PX Pegasus, a new entry to the Drive PX family of computing modules for self-driving cars. Building off of the previous Drive PX family members, the Driver PX Pegasus is intended to be the next step in self-driving hardware by being the company's first Level 5 system, meaning it's capable of supporting fully automated driving for fully autonomous vehicles. Put succinctly, this is the holy grail of what NVIDIA has been building towards over the last few years.

The Drive PX Pegasus is very much a forward-looking product. While NVIDIA is announcing it today, they won't even have dev kits available until later next year, and any kind of commercial release is farther off still. Consequently, the specifications for Drive PX Pegasus are equally forward looking: the board features two  unannounced post-Volta next-generation discrete GPUs, which will be doing most of the heavy lifting. To put this in context, NVIDIA has only just started shipping Big Volta (GV100) for compute products, and smaller scale Volta GPUs are not expected until 2018, so we're looking at something quite far into the future. Meanwhile, rounding out the package and serving as the hearts of the Pegasus will be a pair of NVIDIA's upcoming Xavier SoCs, which combine an integrated Volta GPU (complete with tensor cores) with an unnamed octa-core ARM CPU design.

With the Drive PX Pegasus, NVIDIA is targeting commercial applications in robo-taxis and driverless long-haul. It will, in turn, coexist with the to-be-launched Drive PX Xavier, NVIDIA's previously-announced small-scale self-driving hardware that essentially packs Drive PX 2's capabilities into a small 30W board and features a single Xavier SoC. Drive PX Pegasus, by contrast, is essentially a next-generation Drive PX 2, utilizing much more powerful SoCs and GPUs than before.

NVIDIA has stated that Drive PX Pegasus will be air-cooled, although it's likely to be right at the edge of what's reasonable. WIth a 500W TDP, NVIDIA is pushing the envelope on performance in part by packing so much hardware into a single board. Not that a car will have any kind of trouble delivering that kind of power, of course, but it's quite the interesting change in car design when we're talking about significant amounts of power being dedicated to a non-mechanical operation.

Getting back to the tech specs, of the few details given, the Drive PX Pegasus possesses 320 TOPS of AI inferencing and combined memory bandwidth of over 1 TB/s. Designed for ASIL D certification, PX Pegasus automotive input/outputs include up to 16 cameras (6 of which are lidar). In addition, next-generation GPUs include NVLink and will come in the SXM2 mezzanine form factor also seen with Tesla V100 (as opposed to MXM like the PX 2). While TSMC remains the foundry of choice, the exact process node was not disclosed, although for the Xavier SoC NVIDIA previously announced that they were using 16 FF+. Particularly for the next-generation GPUs, no other information was given, despite being the first official and public reference to any architecture past Volta; on that note, NVIDIA has not yet revealed any detail of consumer Volta.

At 320 TOPS for a dual GPU plus dual Xavier system, this works out at 130 TOPS per GPU (the Xavier SoCs are already quoted as 30 TOPS at 30W). Meanwhile on power consumption, with Xavier already speced for 30W each, this means we're looking at around 220W for each GPU. In other words, these are high-end Gx102/100-class GPU designs. Coincidentally, this happens to be very close to the TOPS performance of the current Volta V100, which is rated for 120 TOPS. However the V100 has a 300W TDP versus an estimated 220W TDP for the GPUs here, so you can see where NVIDIA wants to go with their next-generation design.

NVIDIA DRIVE PX Specification Comparison
SoCs 2x Xavier Xavier 2x Tegra X2 "Parker"
Discrete GPUs 2x Post-Volta N/A 2x Unknown Pascal
CPU Cores 16x NVIDIA Custom ARM 8x NVIDIA Custom ARM 4x NVIDIA Denver &
8x ARM Cortex-A57
GPU Cores 2x Xavier Volta iGPU
& 2x Post-Volta dGPUs
Xavier Volta iGPU
(512 CUDA Cores)
2x Parker Pascal iGPU & 2x GP104?
TDP 500W 30W 250W

Rounding out the specifications, NVIDIA has told us that Pegasus units will be air cooled, not liquid cooled like the 250W Drive PX2 systems. Meanwhile as part of the Pegasus system, NVIDIA has stated that these also need to support failover, but how the failover occurs (either by NVIDIA hardware or other) is up to separate customers in their designs.

In terms of availability, PX Pegasus development kits will be available for select automotive partners in the second half of 2018. However it was not clear if this sampling involved full dev kits with prototype-grade silicon, or an emulation-style system with current-generation hardware to kickstart development. Otherwise, given the long testing and validation cycles in the automotive industry, Drive PX Pegasus may not show up in vehicles for several more years. Last year NVIDIA announced their Level 4 autonomous hardware, but we are only now seeing Level 3 vehicles being showcased in early development for deployment, emphasizing that embedded time scales are much longer than consumer hardware. NVIDIA has stated that their automotive platforms are designed for typical embedded-level longevity, in this case their 'minimum' is a decade of use.

Finally, it's worth noting that this year’s conference is just the second annual GTC in Europe, and going forward NVIDIA has confirmed that GTC Europe will be NVIDIA’s primary automotive-related event. In that vein, NVIDIA has also announced a partnership with Deutsche Post DHL Group and ZF in testing autonomous delivery trucks, starting in 2018.

Source: NVIDIA

Comments Locked


View All Comments

  • sunloveca - Tuesday, October 10, 2017 - link

    Good, but it's not all TOPS/TFLOPS. Firstly It is at 500W TDP, you need a separate battery to power this thing. Level 5 Automation is not all TFLOPS, it's the Algorithms. Nvidia is poor there (you can guess who has mastered the algorithms and the associated HW and why a premium is paid for it). Yeah they are mixing and matching their Pascals/Voltas/Tegras to get TOPS/TFLOPS
  • sunloveca - Tuesday, October 10, 2017 - link

    forgot to complete, Tesla Cars come with 1kWh battery packs, 1000Watts used for 1 Hour continuously. So if this Drive PX- Pegasus at 500W is used for 1 hour it consumes half the battery pack, for doing the same thing that other chips can do smartly. Hmm... I am surprised at the ignorance of the Wall St. analysts (in particular Vivek Arya) and Jim Cramer, they should just stick to what they are good at fudging the Revenues and Growth estimates and putting the estimated EPS.
  • BLee - Tuesday, October 10, 2017 - link

    Tesla S and X have 75-100 kilowatt/hour battery packs, not 1 kilowatt/hour
  • Yojimbo - Tuesday, October 10, 2017 - link

    Hmm. I think you need to check your numbers. Tesla Cars don't come with 1 kWh battery packs, they come with about 75 kWh batteries. Also, consider that 500 W is the max TDP. We don't know the average power usage. The self-driving computer would be using 500 W while it's driving, probably not while it's idling. A Tesla can only drive about 5(?) hours before using all its juice. In that amount of time the self-driving computer will have used at most 2500 Wh or about 3.3% of the battery, while the rest of the car will have consumed the remaining 96.7% of the battery.

    Also consider that, from what I can find, on a hot day a car A/C pulls well over 500 W (I've seen claims of 1.5 kW). If the power can be afford for A/C surely it can be afforded for something as transformative as self-driving.

    Perhaps more importantly, you made the tacit assumption that the automobile must be electric. If the power draw of a self-driving system proves to be too much for an electric car, which do you think is likely to be dropped first by a fleet manager, self-driving ability or electric drive train? I think the economics demands the manager choose the self-driving car.

    Finally, I'm not sure how you can judge the current status of the algorithms of the various players developing self-driving car systems. Why do you think NVIDIA is poor? Why do you think your mystery company is so good? (Who is it, by the way?) What's the evidence? Besides, NVIDIA isn't primarily trying to develop their own self-driving networks, anyway. They are trying to develop the platform (hardware and software) that other companies will use to develop their own self-driving networks.
  • danwat1234 - Friday, October 26, 2018 - link

    60-100KWh. But the computer is running from low voltage, the lead acid battery. So the unit must be shut off when the car is off unless the DC-DC convert is running all the time, which is very inefficient.
    So, I wonder how many seconds it takes for the self driving system to boot up (or more likely resume from hibernate)?
  • edzieba - Wednesday, October 11, 2017 - link

    Tesla already ship the Drive PX2 with a 250W TDP in all their vehicles. Scrounging an extra 250W is not too difficult.

    'The algorithm' is exactly what Nvidia are during off at GTC. They are demonstrating exposed capabilities based on their own NN training.
  • Yojimbo - Wednesday, October 11, 2017 - link

    According to reports I've seen that cite someone who opened his Tesla and took out the Drive PX 2 module, Tesla is using a configuration with one SoC and one discrete GPU, which is half what the 250 W full Drive PX 2 has. So Tesla's version perhaps has a 125 W TDP. So it's another 375 W they must scrounge, but it's still a good point.
  • syxbit - Wednesday, October 11, 2017 - link

    I wish they would release a new Nvidia Shield TV with the DRIVE PX Xavier. It sounds like it would hold up against a PS4!
  • Yojimbo - Wednesday, October 11, 2017 - link

    It's a huge chip though. I've wondered what they'll do with the SoCs for their Shield line and for future Nintendo products since their SoC announcements have all been geared towards autonomous vehicles. Maybe they are working on something that they haven't announced, yet. Maybe we'll see something at this coming CES.
  • chjingl - Tuesday, February 27, 2018 - link

    <a href=" yeezy boost 950 homme pas cher</a>
    Translate this page
    <a href=" nmd r1 homme pas cher</a>
    Translate this page

Log in

Don't have an account? Sign up now