backtop


Print 33 comment(s) - last by xype.. on Dec 1 at 9:24 AM

Exynos 5250 SoC is based on Cortex-A15 architecture

The mobile market is heating up with consumers flocking to smartphones and tablets in record numbers. Samsung is looking to strengthen its position in these markets thanks to the new Exynos 5250 SoC which is the follow-up to the Exynos 4210 used in the popular Galaxy S II.
 
The Exynos 5250 is built using a 32nm low-power HKMG (High-K Metal Gate) process and is based on ARM Cortex A15 architecture. The dual-core Exynos 5250 operates at a speedy 2GHz and has twice the processing performance of 1.5GHz, dual-core Cortex A9 processors according to Samsung.

 
But CPU performance isn't the only thing that's been improved; Samsung says that the Exynos 5250 delivers four times greater graphics performance than Cortex A9 designs (memory bandwidth has doubled to 12.8GB/sec). It also adds in Stereoscopic 3D functionality and support for resolutions up to 2560x1600 (WQXGA).
 
The Samsung Exynos 5250 SoC is being targeted at the tablet market and will go into mass production during Q2 2012.

Sources: Samsung, SammyHub



Comments     Threshold


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

RE: Incremental improvement
By TakinYourPoints on 11/30/2011 2:18:01 PM , Rating: 2
Since when did clock speed mean everything? AMD proved it nearly 10 years ago when a lower clocked Athlon XP outperformed a Pentium 3, and it's happening now when you see an AMD Bulldozer that is barely competitive with a lower clocked i7 from two years ago.

It's a similar case with the A5. Part of the issue is obviously pre-ICS Android being so inefficient, but there are still hard computational limits that current Android hardware has. Doesn't matter how highly it is clocked if practical performance doesn't measure up. Even the Tegra 3 in preliminary GPU benchmarks is falling short of the A5. We'll see how the official ones are soon enough.


RE: Incremental improvement
By steven975 on 11/30/2011 3:05:15 PM , Rating: 2
Well, the Tegra3 is worse because Nvidia pumped it full of CPU cores for marketing reasons, and then went with a lower-end GPU. Ideally, they should have made it dual or triple-core with a SGX543 GPU. More overall performance from the same die. But a "quad core" has marketing oomph I guess.

I find it odd that Nvidia won't use it's own GPU design for their ARM offerings.

Also, the AthlonXP was pretty comparable to a P3 on a clock-clock basis...just the P3 was phased out at something like ~1.3Ghz (and only then in a very niche market). The P4 was the XP's competitor, and AMD outdid it heftily...unless you considered the not-oft-used for the time SSE2 functions where the P4 blew it away.


"Spreading the rumors, it's very easy because the people who write about Apple want that story, and you can claim its credible because you spoke to someone at Apple." -- Investment guru Jim Cramer














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