Qualcomm has had an incredible year. It wasn’t too long ago that I was complaining about Qualcomm’s release cadence, the lull between Scorpion and Krait allowed competitors like NVIDIA, Samsung and TI to get a foothold in the market. Since the arrival of Krait, the move to 28nm and the launch of monolithic AP/LTE solutions, no competitor has been able to come close to touching Qualcomm. These days the choice of integrating mobile silicon really boils down to what Snapdragon variant an OEM wants to go with. TI is out of the business, NVIDIA hasn’t seen much traction with Tegra 4 and even Samsung will ship Qualcomm silicon in many of its important markets. 
 
Qualcomm’s Snapdragon 600 was the SoC of choice at the beginning of the year, with Snapdragon 800 taking over that title more recently. Earlier this week, Qualcomm announced the successor to the 800: the Snapdragon 805. 
 
We’re expecting to see devices based on the Snapdragon 805 to be shipping in the first half of 2014, so Snapdragon 800 will still enjoy some time at the top of the charts.
 
The 805 starts by integrating four Krait 450 cores. Krait 450 appears to be an evolutionary upgrade over Krait 400, with no changes to machine width, cache sizes or pipeline depth. Qualcomm claims to have improved power and thermal efficiency, as well as increased maximum frequency from 2.3GHz to 2.5GHz. I suspect the design is quite similar to Krait 400, perhaps with some bug fixes and other minor tweaks. Qualcomm is likely leveraging yield and 28nm HPM process tech improvements to get the extra 200MHz over Krait 400. Krait 450 also adds 36-bit LPAE (Large Physical Address Extensions) to enable memory support above 4GB. This is a similar path to what we saw desktop PCs take years ago, although I'd expect the transition to 64-bit ARMv8 to happen for Qualcomm next year.
 
The GPU sees the bigger upgrade this round. The Snapdragon 805 features Qualcomm’s Adreno 420, designed 100% in house at Qualcomm. Adreno 420 brings about a D3D11-class feature set to Qualcomm’s mobile graphics, adding support for hull, domain and geometry shaders. Adreno 420 also includes dedicated tessellation hardware. Full profile OpenCL 1.2 is now supported. Texture performance improves by over 2x per pipe, and also gains ASTC support.
 
Adreno 420 is more efficient at moving data around internally. The GPU has a new dedicated connection to the memory controller, whereas in previous designs the GPU shared a bus with the video decoder and ISP. 
 
Qualcomm insists on occluding things like shader unit counts, so all we have to report today are a 40% increase in shader bound benchmarks (implying a 40% increase in shader hardware and/or more efficient hardware). 
 
Snapdragon 805 also features hardware accelerated decode of H.265 content. Hardware encode is still limited to H.264, but this is an awesome first for Qualcomm.
 
The Snapdragon 805 brings a much improved ISP. Qualcomm claims more than a 50% increase in ISP throughput (1GPixel/s class) compared to 640MP/s for Snapdragon 800. The 805 leverages its Hexagon DSP to deliver this level of performance. Qualcomm lists no change in DSP architecture between the 805 and 800.
 
Lastly, we see Qualcomm move to a 128-bit wide LPDDR3 memory interface for Snapdragon 805.  With support for LPDDR3-1600, the Snapdragon 805 features up to 25.6GB/s of peak memory bandwidth. It’s interesting to see Qualcomm go this wide just as Apple moved back down to a 64-bit wide interface. Qualcomm and Intel will be the only two shipping such a wide memory interface in the ultra mobile space come next year (although I do expect Apple to return to a wider memory bus at some point).
 
All of this makes for one beefy SoC, and a continuation of Qualcomm’s leadership in this space. I doubt we’ll see any slowing of Qualcomm’s roadmap after the 805 though. TSMC expects to be shipping 20nm wafers by the end of next year, and I wouldn’t be surprised to find a 20nm successor to the 805 in late ’14/early ’15. Remember that on the last process node shift we got Krait, I wonder what we’ll get this time.
POST A COMMENT

105 Comments

View All Comments

  • vcarvega - Saturday, November 23, 2013 - link

    I think I'm going to pass on this years Nexus 10 and upgrade next year, in hopes that it has a chip like this, or its successor. Reply
  • pbesk18 - Saturday, November 23, 2013 - link

    Vcarvega, you will most likely have to wait 2 more years if your skip the upcoming Nexus 10, I hardly doubt they will release another Nexus 10 next year. Reply
  • Krysto - Sunday, November 24, 2013 - link

    Snapdragon 805 is not a late 2014 chip, unless you mean they will move it to mid-range. They must have an ARMv8 core coming out late 2014, because it would be pretty stupid of them not to have one.

    Overall, I'd say Snapdragon 805 is even less impressive than Tegra 5. Adreno 420 will have ~150 Gflops, while Tegra 5 should arrive ~3 months later with ~300 Gflops, probably similar CPU performance, and much better computational photography features.

    Also, no VP9 support?!
    Reply
  • tuxRoller - Monday, November 25, 2013 - link

    Im not saying you're wrong but where are you getting those gflop estimates for adreno 420?
    Also, regardless of theoretical values adreno tends to do well in actual game benchmarks whereas mali and pvr tend to do extremely well in the 3dmark-type tests.
    Reply
  • Suneater - Monday, November 25, 2013 - link

    Actually by Anand's estimation Tegra 5 will be ~400 GFLOPS. 805 will be crap compared to tegra. Reply
  • blanarahul - Sunday, November 24, 2013 - link

    www.anandtech.com/show/7309/intel-14nm-progress-update-broadwell-airmont-on-schedule

    Tegra 6 with 16 nm comes 2015. Intel 14 nm Atom comes in 2014.

    It's impossible to beat Intel in terms of process node.
    Reply
  • Krysto - Sunday, November 24, 2013 - link

    Either way, the point is the full generation advantage Intel once had, has now been reduced to mere months with the adoption of FinFET by other foundries.

    Even at 22nm, Bay Trail can barely match last year's Exynos 5250 in CPU and GPU performance, while at 28nm. Intel will also not support more than OpenGL ES 3.0 anytime soon, either. So I can't wait to see how far behind Intel will be in mobile technology once ARM chip makers move to 16nm FinFET.
    Reply
  • melgross - Monday, November 25, 2013 - link

    And, according to teats performed here, the A7 is faster than the middle line Bay Trail, and close to the fastest version.

    Intel is pledging big improvements for next year. We'll see if they can make it, and barely keep ahead of the rapidly improving Apple and Qualcomm leading edge designs.

    It will be a good race.
    Reply
  • Kidster3001 - Tuesday, December 3, 2013 - link

    meanwhile, Baytrail uses 1/2 the power as the 5250. Reply
  • melgross - Monday, November 25, 2013 - link

    I like this "on time" description. Intel has now pushed 14nm back two times. I suppose if you take the latest timing by Intel, and it comes out then, it could be said to be on time. But, not really.

    I'm also not hopeful that any other bleeding edge node tech will ever be on time again. Remember that Intel's 22nm was also late.
    Reply

Log in

Don't have an account? Sign up now