backtop


Print 20 comment(s) - last by Moishe.. on Nov 2 at 2:07 PM

AMD anyone?

Advanced Micro Devices, Inc. (AMD) announced on Monday that it would be releasing new 64-bit Opterons in 2014 leveraging a radically different architecture to its current complex instruction set computer (CISC) x86 architecture -- ARM Holdings plc's (LON:ARM) reduced instruction set computer (RISC) ARM architecture.

At the tiime AMD mentioned in passing that the chips were 64-bit.  That led to a bit of mystery, as ARM had not officially announced a 64-bit intellectual property core yet, although one was widely rumored.

That mystery was laid to rest yesterday when ARM Holdings announced a new intellectual property core -- the Cortex-A50 -- a core which leverage ARM's previously announced ARMv8 64-bit instruction set extensions.

The chips will tackle the full range of applications -- everything from smartphones to servers.  It is the linear successor to the 32-bit ARM Cortex-A15.

ARM Holdings has announced several Cortex-A50 cores geared at different objectives. The ARM Cortex-A53 will be the most power-efficent ARM processor, and the world's "smallest" (according to ARM) 64-bit processor.  ARM pledges that the mobile-geared Cortex-A53 will offer "three times the performance" of current generation smartphone chips.

A second core, the ARM Cortex-A57, is a more powerful 64-bit core, aimed at "high-performance applications", such as heavily threaded server workloads.

Calxeda dense server
Server chip makers Calxeda (pictured) are among Cortex-A50's early adopters.

Wonder who is cooking up 64-bit ARM cores?  ARM disclosed that its initial licensee list indeed includes AMD.  Also on the list are Broadcom Corp. (BRCM)  Calxeda (Hewlett-Packard Comp.'s (HPQ) server chip partner, actually an ARM Holdings startup subsidiary), HiSilicon Technologies Comp., Ltd., Samsung Electronics Comp., Ltd. (KSC:005930), and STMicroelectronics N.V. (EPA:STM).

ARM says the new cores should ship in 2014.  The ARMv8 instruction set, though, is currently available for advanced developers and device implementers to start tinkering with.  Usually there's about a half year of lag time between the IP core announcement and the time when official speed and core count targets begin to trickle out from licensees: so chip buffs, stay tuned.

Source: ARM



Comments     Threshold


This article is over a month old, voting and posting comments is disabled

RE: .
By Schadenfroh on 10/31/2012 12:25:30 PM , Rating: -1
x86 is a dying architecture for consumers thanks to the smartphone / tablet shift to mobile platforms. Many folks that I know do not even use their PCs for personal use anymore. Their FriendFace and Gibber posts are all made from tablets / smartphones.

Once tablets and laptops merge for the consumer (Microsoft's surface design is the future), it will be completely dead to them.

In a few years, x86 will be like mainframes are today and will only be used by businesses for workstations / servers.

AMD needs to take the ARM and run with it, it is their only hope to remain competitive in the CPU market.


RE: .
By Reclaimer77 on 10/31/2012 1:54:08 PM , Rating: 3
quote:
Many folks that I know do not even use their PCs for personal use anymore. Their FriendFace and Gibber posts are all made from tablets / smartphones.


Yeeah well some of us use PC's for more than just inane social media and browsing. And that's why X86 isn't dying or going away.

quote:
AMD needs to take the ARM and run with it, it is their only hope to remain competitive in the CPU market.


And just hand Intel a monopoly...brilliant advice.


RE: .
By Arsynic on 10/31/2012 3:10:30 PM , Rating: 2
Having an x86 monopoly will be like having a horse and buggy monopoly...you better be doing something else.


RE: .
By Reclaimer77 on 10/31/2012 3:41:06 PM , Rating: 2
You're funny.


RE: .
By Jedi2155 on 10/31/2012 6:06:53 PM , Rating: 3
However it may become much less important market and a niched product relegated to professionals and nerds. That's the real fear. x86 dominated consumer space for over 20 years, but ARM has a real chance to taking that crown with all these smart phones, tablets and now look at Windows 8 RT..


RE: .
By xti on 11/1/2012 8:47:20 AM , Rating: 2
DT has a hard time realizing...it is the minority. a very small, small minority.

websurfing/social-media/watch a movie is an overwhelmning majority of the consumer breakdown.


RE: .
By Moishe on 11/2/2012 2:01:09 PM , Rating: 2
Good luck trying to do powerful things.

I know there are more phones, tablets, etc in numbers... but those numbers do not remove the need for powerful machines.

Mini-vans are high in utility and don't perform well otherwise. Their numbers outpace sports cars... but there is still a place for sports cars.

If anything, x86 would be supplanted by a powerful ARM, so that the power is available when it is needed.

I have a powerful PC that I use often, but I also have a tablet, and a phone. Clearly there is overlap, but none of the three device classes can replace the other.


RE: .
By xti on 11/2/2012 2:06:52 PM , Rating: 2
you are still assuming the average consumer is a mirror of you.

the average consumer checks email. very few in the grand scheme of things, play crysis at 1080p pushign 100fps.


RE: .
By Shadowself on 10/31/2012 9:55:58 PM , Rating: 1
Yes, the x86 market will eventually go to just compute intensive work. Eventually even things like Photoshop or video transcoding will go to tablets. Desktops and laptops will vanish except for a niche of professionals who absolutely require them to do compute intensive work.

Yes, a large fraction of users don't do anything compute intensive today. Quicken (or MYOB or whatever) or social media don't take much. But a lot of people like to convert their photos (point and shoot or prosumer) to formats that are more distributable. Today's tablets are barely able to do that without a significant lag. Upload RAW format images (and yes, a lot of people go on the 'net and read that they should ONLY shoot in RAW and do so without understanding what it really means) from their Nikon or Cannon prosumer device to a tablet then convert it to a more manageable image and it's not so trivial. For the next few years it's going to be much, much easier on a desktop or laptop. And with the ever increasing pixel count of prosumer cameras, as well as bit depth of each pixel, the tablets are chasing a moving target.

Eventually tablets, and to a lesser extent smart phones will win out. However, tablets and phones replacing desktops and laptops won't happen for at least 10 years or more. It could even be 15 years or more.

The processing power of a current high end Xeon chip or current top end i7 is so much higher than anything ARM is currently putting out -- or even has planned in the next couple years -- that the difference is going to insurmountable for many, many years. Add to that the highest end GPUs in ganged modes (running 2 or 3 or 4 GPUs in the same desktop) and you've got machines that that ARM based hardware just fantasizes about being when they grow up.


RE: .
By someguy123 on 10/31/2012 10:41:28 PM , Rating: 2
You say "eventually" yet even the top of the line consumer CPUs struggle depending on software, even in photoshop editing. One of the reasons why adobe added OCL to their software suite.

I find it sort of confusing how people can claim ARM will dominate so soon when the chips are so slow. Ask someone if they want a dual core atom and they'll say it's much too slow, but dual core krait would have people drooling even though they're similar in performance. People seem to think that the benchmark for high CPU performance is web browsing on a cellphone.


RE: .
By inperfectdarkness on 11/1/2012 2:37:04 AM , Rating: 1
get back to me when crysis can be played in wqxga on a tablet. i swear by a 15" laptop--not because it's the fastest or most powerful pc, but because 15" has always been (imho) the perfect balance between power and portability. i spend a LOT of my time "away from home". i'm also a religious gamer (that's gaming religiously, not someone who plays religious games).

now i've got nothing against half-life 1, but if that's the most "modern" game that a tablet can handle, then it sure as sh** isn't going to cut it for me.

i will always feel my computing needs are adequately met between a laptop & a smartphone (if i choose to get a smartphone). i can derive little additional benefit from procuring a tablet, nor can a tablet offer nearly enough functionality to sway me away from a dedicated laptop.

i've said the same thing about owing a wii/laptop & not buying a ps3/360. sure there's a market BETWEEN the two, but for someone who owns both ends, there's virtually no additional benefit.


RE: .
By Moishe on 11/2/2012 2:07:51 PM , Rating: 2
Agreed. People don't understand that there are multiple markets, and they vary in size. Tablets will always sell more than PCs because they are cheaper and more portable and take care of the basic needs.

But good luck doing intensive things on tablets.

I use mine as a DVR, gaming machine (BF3), development, video processing, etc. Those won't be taken by tablets anytime soon. At best development, video processing, and low end gaming can be done on tablets. I think the video processing will always be slower.


RE: .
By Wolfpup on 11/2/2012 9:50:21 AM , Rating: 1
Ooooh, I see. So people will stop browsing the full web, writing emails, writing documents, tracking their finances, buying movies/shows, taking photos and videos, etc., and just passively thumb through a few web pages. Good to know! I had no idea people didn't like modern technology until you pointed it out!


"Game reviewers fought each other to write the most glowing coverage possible for the powerhouse Sony, MS systems. Reviewers flipped coins to see who would review the Nintendo Wii. The losers got stuck with the job." -- Andy Marken














botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki