POST A COMMENT

81 Comments

Back to Article

  • Nintendo Maniac 64 - Tuesday, April 29, 2014 - link

    Hmmm, sounds like the AMD equivalent of an Intel "tick", especially considering that the IPC between Puma+ and Jaguar is unchanged.

    Interestingly enough, this would mean that the PS4 and Xbone could use Puma+ cores in the future (with turbo disabled obviously).
    Reply
  • nevertell - Tuesday, April 29, 2014 - link

    Why would they need to disable turbo? I believe nobody is hitting the CPU performance limits just to have a fps limit or rely on the raw performance for timing, whereas this could improve some load times or improve performance during context switching. Reply
  • mwarner1 - Tuesday, April 29, 2014 - link

    Consoles have fixed performance hardware to prevent games & applications performing differently on different hardware revisions. If you bought a PS4 today and then next week a new version was released (but likely not announced) that made games smoother / more playable then you would have the right to be annoyed. Reply
  • Havor - Tuesday, April 29, 2014 - link

    No the main reason for fixed performance hardware is that developers dont have too add code or scalable textures to adjust or performance differences.

    And thus they have a more efficient single spec code, that dose not have to adjust to hardware spec.
    Reply
  • nathanddrews - Tuesday, April 29, 2014 - link

    Yeah, I highly doubt they'll switch architectures since it has never happened before. Power savings and console redesigns come from shrinks and on-die packaging. Reply
  • Samus - Wednesday, April 30, 2014 - link

    There's nothing stopping Sony or Microsoft from launching a "performance edition" PS4 or XBOX One with a hardware bump that simply added antialiasing, etc, to games.

    This has already been done over the years with Nintendo offering the 4MB RAMBUS upgrade for the N64, and various performance storage options for XBOX 360/PS3 to assist load times of disc-based games. The SSD-edition of the 360 can load games/levels virtually instantly compared to running from disc or disk.
    Reply
  • mfoley93 - Wednesday, April 30, 2014 - link

    These aren't really higher performance though, just lower power. They could just lower the clock to offset whatever slight performance gains there are to equal the launch products. Reply
  • nathanddrews - Wednesday, April 30, 2014 - link

    1. When Microsoft or Sony want to increase performance, they only do so via software updates that don't destabilize the platform as a whole. Neither can afford to break millions of consoles with a bad update or segregate the community into two camps. On the other hand, if developers of individual games find a way to improve framerates or AA, they can submit updates for download - but only after it is tested by the console manufacturers.
    2. I have the 4MB upgrade for my N64, only TWO games required it, a very small percentage of N64 games supported it, and even fewer truly benefited from it. It's mild success was due entirely to ZMM, DK, and PD, but Nintendo hasn't tried anything like it since. (Lest we forget the 64DD...)
    3. Only Sony lets you install any drive you want. Most reviews from those that have upgraded to SSDs say it just isn't worth it. It's a consumer option, not something Sony changes at the platform level. The games still run at the same speed with the same textures.
    4. There is no consumer "SSD Edition" Xbox 360 and they won't let you install one (officially). Are you referring to the 4GB Slim? That's not an SSD and most 360 games are too big to install onto it.
    5. I have yet to see a console SSD upgrade result in anything instantaneously... except regret. :D
    Reply
  • Kevin G - Wednesday, April 30, 2014 - link

    Consoles are to be 'fixed spec' so that game developers know exactly what to expect in terms of hardware. The lone exception has been storage capacity. The N64 memory expansion is an excellent example of why developers aim for the lower guaranteed spec: only three games required it with a handful of games that'd use it if present.

    Both MS and Sony could come out with a hardware revision that does a bit more outside of gaming without impacting game developers. For example, MS could release an Xbox One with a digital tuner + DVR hardware. Such a change would have no impact to the gaming side of things. Ditto if MS or Sony were to add backwards compatibility via hardware: it'd be unavailable to use in an Xbox One or PS4 game.
    Reply
  • Arbee - Thursday, May 01, 2014 - link

    There was a late revision of the original PlayStation where the GPU got significantly faster for some operations, which resulted in higher frame rates in some games. (This was when the debug units switched from blue to green, in order to differentiate). Reply
  • superunknown98 - Tuesday, April 29, 2014 - link

    Although I don't think it would happen, or at least be publicly announced, Microsoft could use these new cores in Xbox One but could only enable turbo for the two cores that run the virtualization and Xbox OS. They would also benefit from the reduced TDP, which is something that eventually happens at some point anyway. Reply
  • lmcd - Thursday, May 01, 2014 - link

    Now that comment on the OS and virtualization cores was quite interesting. I now thing that a Puma-edition is likely (though I think a GPU switch-up is more likely if more efficient GCN variants occur. Reply
  • MikeMurphy - Monday, May 26, 2014 - link

    No sense revising an entire chip to save a few watts of power. They might revise it later provided that substantial power savings are attainable, otherwise will implement the usual die shrinks. Minor performance increases shouldn't be ruled out although focus will be on power reduction while maintaining similar performance. Reply
  • Rockmandash12 - Tuesday, April 29, 2014 - link

    And this is what happens when AMD gets into gear and makes a new architecture. Real improvement that's competitive with rivals. Common.... new Desktop flagship architecture that's faster and more efficient? please? Reply
  • Samus - Wednesday, April 30, 2014 - link

    This is pretty impressive. And honestly, out of nowhere. They all the sudden have an amazing tablet/uSFF SoC. Reply
  • silverblue - Tuesday, April 29, 2014 - link

    Didn't Kaveri launch with a fully enabled PSP as well?

    Judging by the performance story thus far, I think it will put to bed the calls for cat cores to replace AMD's higher powered offerings. Yes, we're past K8 performance levels now, but Llano and Trinity (let alone Richland/Kaveri) still have it beat. You'd need some serious clock speeds to get decent performance and it's the wrong silicon for that.

    I was disappointed to see that it's practically the same uarch as Jaguar, meaning we're still going to have a single channel memory controller, however the performance and power improvements are substantial, and the memory controller has been improved anyway which should reduce the need for said controller.
    Reply
  • ET - Tuesday, April 29, 2014 - link

    I think it's hard to draw conclusions of core performance when the RAM is a limitation. It's entirely possible that these cores are still pretty far from the big cores, but on the other hand it's possible that more bandwidth could up performance by quite a few percent. Reply
  • ssnitrousoxide - Tuesday, April 29, 2014 - link

    AMD has always done some impressive work to squeeze every bit of performance from an inferior node. How they managed to improve energy efficiency so much is beyond me. Reply
  • mfoley93 - Wednesday, April 30, 2014 - link

    It seems that most of the power reduction is at the manufacturing level; maybe it's more accurate to engineering tolerances or perhaps a more pure silicon, either way I don't think TSMC will be telling us what it is. The rest comes from eliminating circuitry that provides some more flexibility to OEMs, something nVIDIA has been doing for a couple years now, and while it doesn't really
    count, it's something Apple does very well.
    Reply
  • yannigr - Tuesday, April 29, 2014 - link

    Any idea if Beema will be compatible with existing AM1 motherboards?
    (with only a BIOS update of course)
    Reply
  • saneblane - Tuesday, April 29, 2014 - link

    Unlock That 2.4GHZ Beema Pls. It sure would be nice to get an Amd quadcore 3GHZ at 25 watts.
    Amd Should increase the highend on these and kill of Low Power Kaveri.
    Reply
  • MrHorizontal - Tuesday, April 29, 2014 - link

    Interesting little package that.

    But very little detail on the memory controller - how wide is it? From what I can see it's just single channel. Personally I think these Puma cores would serve AMD a lot better as a low power Opteron CPU than ARM would, equivalent to Intel's Avoton package, provided they can use a significant amount of memory on a system - a single channel won't cut it.
    Reply
  • MrSpadge - Tuesday, April 29, 2014 - link

    Single channel, since it's I/O compatible with the previous chips. Reply
  • mfoley93 - Wednesday, April 30, 2014 - link

    The author should add this, to the article, not everyone that is interested keeps up on this stuff closely enough to remember all of the details from 8 months ago. Reply
  • piroroadkill - Tuesday, April 29, 2014 - link

    Makes you once again realise how wimpy the CPU is in the Xbox One and PS4.

    8 Jaguar cores. More crap is still crap.

    These cores are far better than that, and still wouldn't be considered high end.
    Reply
  • H2323 - Tuesday, April 29, 2014 - link

    8 Jag cores that have access to fast memory are plenty good for a multi-threaded game Reply
  • tipoo - Tuesday, April 29, 2014 - link

    Disagreed. It was a matter of performance for cost, and even though each Jaguar core may perform half as well as a high end desktop core, they were able to fit 8 cores into less die space than those bigger cores would have taken. If a developer takes good advantage of multithreading they'll be fine for a while. Reply
  • Antronman - Wednesday, April 30, 2014 - link

    But they won't considering most engines used are multiplatform engines and the majority of the games made with those engines are single threaded.

    The performance sucks, no matter how you look at it.
    Graphics have a big effect on performance, and the graphical power on the new consoles is...less than enough.
    Reply
  • lmcd - Friday, May 02, 2014 - link

    Are you kidding? Engines mean that things can be more threaded as it's more abstracted and so more time can be spent on threading (as opposed to rewrites and higher-level coding that a smaller team would deal with). Reply
  • coburn_c - Tuesday, April 29, 2014 - link

    Oh look another AMD product launch with regressive performance... and yet another AMD product launch where their power usage is double that of the competition. Still I suppose it's an improvement, they could be at 4x the power usage now if they hadn't finally done some work. Well I suppose they can just lay off another 25% of the company, that'll sure help close the gap.. Reply
  • duploxxx - Tuesday, April 29, 2014 - link

    Rival really here is intel, nowhere compared in this review Reply
  • ssj3gohan - Tuesday, April 29, 2014 - link

    You do realize that this is an AMD advertisement, not an actual journalistic article? Look at the top of the page: it says AMD Center. Reply
  • DanNeely - Tuesday, April 29, 2014 - link

    Anything tagged AMD gets the AMD Center branding, regardless of the content (which's lead to an occasional giggle-snort situation in with articles that are 99% Intel/nVidia but mention AMD for comparison purposes). Look at the by line, this was written by Anand, not one of the AMD PR drones. Reply
  • mfoley93 - Wednesday, April 30, 2014 - link

    There is also an Intel Center and nVIDIA Center that all articles about their respective products go in, does that make all articles about them advertisements as well? Reply
  • ruggia - Tuesday, April 29, 2014 - link

    If you want a rough comparison with BayTrail, go check out the benchmarks at Toms' (or pull up Anand's old Kaveri articles). You'll be amazed. Reply
  • ruggia - Tuesday, April 29, 2014 - link

    *Kabini Reply
  • duploxxx - Friday, May 02, 2014 - link

    there is info here and you see that it can very well compete with intel parts or even better in many parts. the problem is that many bench just wont run on the low end atom GPU wise :). you need the celeron version to compare performance to mullin Reply
  • Anand Lal Shimpi - Tuesday, April 29, 2014 - link

    All of the CPU tests are compared to Intel parts, including the direct competitor: Intel's Bay Trail based Atom both in the ASUS Transformer Book T100 as well as numbers from Intel's Bay Trail tech day using the Atom Z3770. Reply
  • Stuka87 - Tuesday, April 29, 2014 - link

    Intel chips are in several of the graphs, and is talked about specifically in several places.

    Did you even read the article?
    Reply
  • MrSpadge - Tuesday, April 29, 2014 - link

    A substantial performance increase over its predecessor is regressive? And if you insist comparing to AMDs big cores.. well, have fun putting these into a tablet!

    2x the power consumption? You seem to refer to idle power compared to Snapdragon 600, a SoC which is significantly slower and hence - just like the big cores - in a different category.

    Beema / Mullins seems to occupy a very interesting point on the performance/power/price curves. It's obviously not for smartpones yet and neither for big gaming rigs. But for anything in between it's at least an intersting option.
    Reply
  • vlad42 - Tuesday, April 29, 2014 - link

    Hey Anand,
    Do you know if connected standby was enabled on the Mullins tablet you tested? I’ve heard that AMD has not yet developed the connected standby drivers yet. If this is the case, then shouldn’t that be noted in the power consumption test? Given the improvements Intel sees when connected standby is enabled, it definitely looks like the Mullins tablet was not using connected standby.
    Reply
  • Anand Lal Shimpi - Tuesday, April 29, 2014 - link

    No CS, although I believe Mullins could theoretically support it. For this particular test there should be no advantage to having CS, we're just looking at short/active idle power usage. Reply
  • Anand Lal Shimpi - Tuesday, April 29, 2014 - link

    Correction - there are no plans to support CS, AMD doesn't see value in it. Reply
  • tomsworkshop - Tuesday, April 29, 2014 - link

    this low power but high performance tiny little chip can be fit on something like the raspberry pi, we can build a really small micro pc with such hardware. Reply
  • mfoley93 - Wednesday, April 30, 2014 - link

    They aren't low power enough yet, so in the mean time, I suspect nVIDIA's Jetson board based off the Tegra K1 might fit the spot you are thinking of. The board is significantly larger, but its powerful enought to justify that. It's also more in the Beagleboard/Pandaboard segment, being aimed at embedded development and not education.

    http://www.nvidia.com/object/jetson-tk1-embedded-d...
    Reply
  • tomsworkshop - Thursday, May 01, 2014 - link

    The Tegra K1 rated at 5W TDP, the Mullins rated at 3.95W - 4.5W TDP, i think they should be low power enough for a single board computer, i saw the price for the Jetson TK1, it was 3 times higher than the Raspberry Pi, hope that AMD will come out something at the middle of the Raspberry Pi and Jetson TK1, with the price lower than the Jetson TK1, and the performance better than the Raspberry Pi. Reply
  • nemi2 - Tuesday, April 29, 2014 - link

    It's good to see AMD catching up but the power consumption may still be a deciding factor when choosing between this and Baytrail - I look forward to a full review of release hardware. The other area of concern is that the Baytrail successor Airmont (with shrink to 14nm and ~30% power savings) will also be out in 2014 so AMD may only have 0-3 months at parity/competitiveness with intel. Reply
  • H2323 - Tuesday, April 29, 2014 - link

    Yes these are shorty terms wins, but 14nm airmont is late and will miss back to school and be just in time for christmas, AMD will net some revenue on this and still retain GPU lead with baytrail Reply
  • R3MF - Tuesday, April 29, 2014 - link

    a four-core mullins in an 8" tablet chassis with 4GB of memory and bios support for a 64bit ubuntu would be an awesome thing. Reply
  • MrSpadge - Tuesday, April 29, 2014 - link

    I'd rather have a Thinkpad 10 + Win 8.1 with this, at least as an option instead of Bay Trail. If the actual product is as impressive as the preview. Reply
  • H2323 - Tuesday, April 29, 2014 - link

    I think that is already a reality, Lenovo will be going with beema and kaveri for a variety of laptops, same with dell, AMD needs some HP wins though Reply
  • artk2219 - Thursday, May 01, 2014 - link

    Thats interesting, normally I see that AMD has chips in lots of HP products but very rarely in any Dell products. Reply
  • ecmtb - Thursday, May 08, 2014 - link

    According to this, Dell is planning on putting out a laptop with these new parts:
    http://www.cpu-world.com/news_2014/2014041801_AMD_...
    Reply
  • formulav8 - Tuesday, April 29, 2014 - link

    This sentence doesn't quite sound correct.

    "The result is a memory interface that shaves off move than 500mW when run in this more strict, low power mode."

    Move and Run mess it up. :)
    Reply
  • HalloweenJack - Tuesday, April 29, 2014 - link

    oh look the intel fangirls have arrived - a 4.5w part is matching a 10w intel part and they cry `its crap`... Reply
  • MrSpadge - Tuesday, April 29, 2014 - link

    Seems more like the Intel hooligans ;) Reply
  • Kjella - Tuesday, April 29, 2014 - link

    Yeah, power consumption has been AMDs Achilles heel for some time now both on the high end for performance and on the low end for battery life. And suddenly they can put the performance of a 15W Kabini in a 4.5W Mullins? If this holds up when it's actually released and reviewers can run real tests on it - battery tests would be extremely interesting - it could be a real winner. Assuming that people want x86 tablets though, I think the jury is still out on that. Reply
  • zodiacsoulmate - Tuesday, April 29, 2014 - link

    How could this compete with Atom?????
    Atom have way lower TDP SDP or whatever power consumption they call it...
    Atom have no problem surpassing the performance of this thing...
    Price maybe AMD's advantage, but low power devices are still filled with high profit devices, also it's TDP may cause bad battery life, with may not be taht cost efficient afterall...
    So why AMD still trying to compete here??
    Reply
  • H2323 - Tuesday, April 29, 2014 - link

    It wrecks anything silvermont based in benchmarks, Thing is mullins will not make a lot of gains in tablets, no android support and free intel chips, Beema will make some gains on baytrail until christmas. Reply
  • rocketbuddha - Tuesday, April 29, 2014 - link

    Hi Anand,
    Can you confirm if Beema & Mullins are made at TSMC 28nm HKMG? Hardware France seems to think it is Global Foundries.
    Thx
    Reply
  • H2323 - Tuesday, April 29, 2014 - link

    everyone else is saying gloflo Reply
  • Anand Lal Shimpi - Tuesday, April 29, 2014 - link

    It's GF, I've updated the piece :)

    Take care,
    Anand
    Reply
  • raghu78 - Tuesday, April 29, 2014 - link

    anand
    do you have information on the exact GF 28 nm process used for manufacturing Beema and Mullins. Is it GF 28 SHP (Super high performance) or GF 28 SLP (Super Low Power).
    Reply
  • Da W - Tuesday, April 29, 2014 - link

    Yeah but current gen console have PC-like architecture because developper wanted easier PC porting. Most will already port their game for the Mullins/Beema architecture anyway. Reply
  • Da W - Tuesday, April 29, 2014 - link

    Man, i'm dying for an AMD-powered surface mini with Xbox/PC streaming capabilities like Nvidia Shield! MAKE IT HAPPEN! Reply
  • H2323 - Tuesday, April 29, 2014 - link

    That is a real possibility, might be one of there two semi-custom wins from earnings Reply
  • littlebitstrouds - Tuesday, April 29, 2014 - link

    I wish you would provide the Sony Vegas render test in anything that will run it. I'm looking for a tablet that's a dailies workhorse. Reply
  • Dawgmatix - Tuesday, April 29, 2014 - link

    Any idea on if this will support OpenCL 2.0? Reply
  • MarionHNunn - Tuesday, April 29, 2014 - link

    I think it will put to bed the calls for cat cores to replace AMD's higher powered offerings. Yes, we're past K8 performance levels now, but Llano and Trinity (let alone Richland/Kaveri) still have it beat. You'd need some serious clock speeds to get decent performance and it's the wrong silicon for that. http://sn.im/28v1ckw Reply
  • wow&wow - Tuesday, April 29, 2014 - link

    "Mullins seems like a good fit for a high performance Android tablet, but ..."

    Mullins seems like a good fit for a high performance Android tablet, but HOW BIG IS THE HIGH PERFORMANCE ANDROID MARKET?

    With limited resources and wafer supplies, led by Rory Read with business sense, AMD has been focusing on making money (the #1 common business sense) instead of BLINDLY COMPETING IN EVERY MARKET for the sake of conpeting and failing like before. Every business is a good business is a sales talk, but in reality not every business is a good business and pick the proper ones!
    Reply
  • name99 - Tuesday, April 29, 2014 - link

    "I’d expect a similar die size to Kabini/Temash. It’s interesting to note that these SoCs have a transistor count somewhere south of Apple’s A7."

    Isn't this something of an apple's to oranges comparison?

    This AMD SOC is basically CPU+GPU+memory controller.

    A7 is all that plus secure storage, ISP, h264 encoder/decoder (the genuine low power deal, not some "hardware assisted" frankenstein that runs the CPU and GPU [together, both at high power] to do the job) along with god knows what else --- flash controller? fingerprint recognition cell?
    Reply
  • mczak - Tuesday, April 29, 2014 - link

    Kabini / Temash also full custom hw video encode/decode (all gcn based chips do), though if you want some hybrid mode is still available, so that should be pretty comparable. Flash controller and the like, too. Yes no ISP, but OTOH there's quite a lot of stuff the A7 won't do too (like 2xsata, the 4x1 and 1x4 pcie 2.0 connectivity, 2xUSB 3.0, high-speed i/o isn't exactly cheap). Anyway, the transistor count and die size is comparable after all (based on the official numbers, Kabini is slightly larger, but the a7 has slightly more transistors, though there's both different methods to count transistors and measure die size, not to mention they come out of different fabs), and it shouldn't be a surprise. Reply
  • lmcd - Friday, May 02, 2014 - link

    AMD should try partnering with Broadcom (as Broadcom has no real SoCs for smartphones). Reply
  • 200380051 - Tuesday, April 29, 2014 - link

    I am eager to see how Mantle-enabled games will perform on these Mullins tablets. It seems a good fit from a technical standpoint. It might just push the PC gaming sphere to dig into tablet space. This in turn directly expands the market of game studios.

    Also, I wonder if AMD's mobile lineup is to be the first product they'll roll out on Samsung's 14nm FINFET process. The process will be available starting 2015, as per their agreement. Its up to AMD to cook us a shrinked revision of these chips in a timely fashion.

    Things are getting interesting.
    Reply
  • MartinT - Wednesday, April 30, 2014 - link

    It seems to me that performance numbers for these parts don't tell even half the story without the accompanying power readings, considering the 'use whatever power until the chassis burns the user' approach of AMD's turbo implementation. Reply
  • kirilmatt - Wednesday, April 30, 2014 - link

    How AMD did this is amazing. Imagine if this was released instead of kabini/temash. This destroys Bay Trail. I only hope that it gets released soon so it doesn't have to compete with Intel's 14nm SoCs. Anyways, good job AMD! Reply
  • R3MF - Wednesday, April 30, 2014 - link

    Ubuntu tablet please... Reply
  • purerice - Wednesday, April 30, 2014 - link

    Would one way to test the "non-turbo" performance be to loop some test 100 times and see the performance decrease over time? Considering the turbo would decrease as the CPU/APU heats up we could see the performance difference and also how long you really get "turbo" turned on for. Reply
  • azazel1024 - Thursday, May 01, 2014 - link

    I am impressed, but I am curious as to both why Bay Trail beats it in the PCMark testing by a fair margin, but not in individual CPU benchmarks. If that is thermal limits...well, I will say that a lot of tablet workloads are very short term. Windows tablet workloads (at least mine)...not so much.

    Enough of what I do would likely hit those thermal constraints and at least in my testing, my T100 doesn't clock down even under very prolonged workloads, like 15+ minutes of converting RAW to JPEG images. Or long gaming, like an hour or two of KSP.

    That and I have concerns about that idle and low power use. Seems to be pretty good under higher load and performance seems to be there (with caveat/concern)...but idle and low power could be an issue. According to those AMD specs, the APU itself is using darn near 2w of power streaming 1080p. Based on my math, my T100 TOTAL uses around 2.4w of power when streaming 1080p (around 13hrs of run time, 31whr battery). I assume that the display, wifi, signal processor, memory, etc, etc are consuming more than .6w of power.

    Having a much bigger battery or much shorter run time could be a big sticking point for a lot of tablet users (I know I'd have an issue if my 6-7hrs gaming/10hrs normal use/13hrs video turned in to more like 3hrs gaming/6hrs normal use/8hrs video.
    Reply
  • FITCamaro - Wednesday, May 07, 2014 - link

    My next tablet will likely be a Windows 8.1 tablet. I'd love the high end AMD CPU tested here even if it doesn't do as well on power as Baytrail but bests it in GPU performance. Would be nice to be able to do better light mobile gaming. Reply
  • enihcam - Monday, May 19, 2014 - link

    AES benchmark is missing. Reply

Log in

Don't have an account? Sign up now