CPU, APU & Chipsets News - Page 140

ARM have just announced its next-generation mobile GPU, the Mali-T658. ARM have said that this design is set to offer ten times the performance of their current Mali-400 MP which is found in smartphones such as the amazing Samsung Galaxy S II handset.

T658 is ARM's second GPU using an architecture it calls Midgard. Midgard is designed to support both 3D workloads using modern APIs, with support for both OpenGL ES and Microsoft's Direct3D 11, and computation workloads using OpenCL, Microsoft DirectCompute, and Google RenderScript. Compute tasks get some love, with Midgard supporting full IEEE 754 floating point.

The first Midgard design, the T604, was announced last year, and was licensed by companies such as Samsung and LG. The new T658 is quite the powerhouse, with each core having four arithmetic pipelines, and one each of load/store for texture, doubling the number of arithmetic pipelines found in the T604. Up to eight cores can be integrated into a single GPU, again representing a doubling in performance relative to T604, which allows up to four cores to be ganged together.

Ivy Bridge isn't even here yet, and we're seeing details on Intel's next-generation processor architecture, Haswell. Haswell will be a brand new architecture based on a 22nm fan process which will be matured by Sandy Bridge's successor, Ivy Bridge. This is Intel's "tick-tock" product development model, where we see a "tock" being a new x86 architecture, and a "tick" miniaturizes it to a newer silicon fabrication process.

If Intel's 22nm process all goes to plan, Haswell has an ETA of Q2 2013, with Ivy Bridge taking the lead from Q2 2012 until then. Haswell will feature an all-new socket, LGA1150, which means it will not be compatible with LGA1155 boards. The reason behind this is that Intel have implemented drastic changes in the pin map of the package.

There has been a major change with the component arrangement in the platform that is affecting Haswell's pin map, more specifically, it's Haswell's higher bandwidth chipset bus, rearranged PCIe pins (with FDI pins), rearranged power pins, and miscellaneous pins. It also throws away the separate power domain for the integrated graphics controller.

AMD's Bulldozer just wasn't enough to not only topple Intel, but to impress customers enough to grab it while it was hot. AMD knows they need to turn it around, we know they need to turn it around and you know they need to turn it around. AMD has always focused on performance-per-watt, and while Fusion has pulled this off, Bulldozer when overclocked is pushing past an astonishing 400W.

The Bulldozer team needs to put their hard hats on and get dirty, with 2012 rolling into view, it needs to happen very soon. Piledriver is an update to current Bulldozer cores and the expected advantages are increased core capacity by up to 25-percent, making more cores available more of the time. Reduced power requirement, making Piledriver more efficient, real world performance boost by somewhere around 10-percent, making each core do more.

Will this be enough for AMD to at least catch up and possibly take a few swings at Intel's upcoming Ivy Bridge? Can they bring it before Ivy Bridge, or at the same time? Can AMD release it at a competitive price point, seductive enough to draw customers into their web of CPU goodness? Bulldozer was roughly the same price as an Intel 2600K motherboard/CPU combo, so they really failed in that way.

We recently saw quite the breath taking overclocking record out of an AMD FX-8150 CPU from world renowned overclocker Andre Yang, cranking it up to a whopping 8.46GHz, above that of AMD's own previous record (at 8.43GHz).

With only 30 odd MHz separating it, Andre Yang has given it another crank and has produced a new record result, taking the processor to an incredible 8584.8MHz exactly; 123.3GHz higher than his previous achievement. We're still waiting to see the new overclock result appear on CPU-Z's validation database, but its authenticity looks good enough for us.

A bit about how the massive overclock was achieved, Andre used the same Crosshair V Formula for the job, though the core voltage was bumped slightly for this new record to 2.076V (up from 1.992V on his previous). Of course, extreme cooling measures were used for the task at hand; liquid nitrogen. But considering you can go one better by opting for liquid helium which liquefies at -269 Celsius (vs the -253 of nitrogen) in astosphere pressure, there's possibly higher clock rates to be seen from Bulldozer yet; not to mention the upcoming B3 stepping which may well improve overclockability even further.

ARM Ltd is getting serious against Intel and AMD, and are now going to compete with the chip giants in the high-end and server computing market, where competition is tight. ARM has announced their new ARMv8 architecture, the first one to include a 64-bit instruction set. ARM CTO, Mike Mullar says:

ARMv8 will enable the development of ARM architecture compatible devices that can be designed to maximize the benefits across both 32-bit and 64-bit application areas. This will bring the advantages of energy-efficient 64-bit computing to new applications such as high-end servers and computing, as well as offering backwards compatibility and migration for existing software through a consistent architecture.

ARMv8 will have both 32- and 64-bit modes, like todays x86/x64 CPUs, and this should help transition ARM into the 64-bit world for existing applications. ARM have dubbed the 32-bit mode "AArch32" and the 64-bit mode "AArch64".

The folks over at OCWorkbench were lucky enough to have witnessed a Sandy Bridge-E processor being overclocked earlier today and kindly shared some details that help give us an idea of how well these chips will clock up.

The specific LGA2011 Core i7 model used was not mentioned, so it's possible it was only a quad core offering. Alas, using a "regular" air cooler it still managed to cruise along just shy of 5GHz (4.92GHz to be exact) with its idle temp holding at 45c.

This was made easily possible via the BClk that can be increased on Sandy E without any dramas. This particular result was achieved by setting the BClk to 120MHz, using a multi of 41X and a core voltage of 1.51v. As for the memory, this was DDR3-2400MHz RAM with a CAS latency of 10T.

A bunch more details have surfaced today on AMD's upcoming "Virgo" PC platform which comprises their next-gen mobile "Trinity" APU using an FM2 package. The latest details confirm that Trinity will in fact be compatible with AMD's current-gen A75 "Hudson-D" chipset, but we are yet to learn whether or not FM1 and FM2 are pin compatible.

AMD's Trinity mobile APUs are 32nm based and run a TDP of up to 35W. The four x86-64 cores used are based on AMD's next-gen "Piledriver" architecture (rumoured to be between 10 and 15% quicker on average when compared to Bulldozer) and these four cores are arranged in two modules. Each of the two modules holds two cores and is given certain shared and dedicated resources.

The new APUs sport an integrated DDR3-2133MHz memory controller, up to 4MB (2MB per module) of L2 cache and Radeon HD 7000 series IGP (said to be about 30% quicker than Llano's iGPU). AMD is talking a 20% performance improvement over current-gen Llano APUs which comprise K10 "Stars" cores. A new 3rd generation form of TurboCore technology will also be included that gives superior power management and overclocking features.

As we learned about a month ago, Intel has been planning the release of a new processor model to rest atop the i7-2x00 line; that being the 2700K. And we we had also established at that time, the processor's release date is today.

The Core i7-2700K looks not to seperate itself much at all from the 2600K on paper with its quad-core, LGA1155 package on 32nm Sandy Bridge silicon, 256KB L2 cache per core and 8MB L3 shared cache. It clocks in at 3.5GHz, but does of course sport an unlocked base clock ratio multiplier, as denoted by the 'K'.

While we won't know for sure this early into its release, it's been rumoured that these 2700K chips are essentially cherry picked Sandy Bridge processors that could well give even higher overclocking ceilings on average when compared to the 2600K. But while it would be nice to say you'll get a guarenteed 5.0GHz on air when picking up one of these puppies, we'll have to wait a little bit first and see what the general consensus is from peoples overclocking experiences.

ARM has unveiled a new chip that is set to enable more efficient and affordable mobile processors. Cortex A7 is based on ARM's 28nm fabrication and reportedly consumes five times less power and measured just one-fifth the size of the 45nm Cortex A8, which is found inside Apple's A4, Samsung's Hummingbird and Texas Instruments' OMAP3.

ARM will compliment the efficiency by pairing the Cortex A7 with quicker, more power hungry processing cores. The A7 is set to pave the way for sub-$100 smartphones, which will boost adoption rates in developing regions. ARM CEO, Warren East told the BBC:

We can see the developed world moving on and mobile being the nexus for all sort of consumer electronics. In the Bric countries (Brazil, Russia, India and China) we are seeing catch-up. As we look forward these smartphones are going to be totally ubiquitous and in the much less developed areas...

Given the ill received launch of AMD's FX (Bulldozer) CPU family no thanks to the lack of what mattered most, performance, it's hard to imagine how AMD can do much to rise above anytime soon, or even maintain the foothold they had in the market previously in terms of offering a more aggressive price/performance alternative.

However, it's looking like AMD still may have another playing card up its sleeve yet; this in the form of a new (B3) revision of the chips being worked on. A mention of the new stepping was spotted in a public AMD document entitled "BIOS and Kernel's Developers Guide (BKDG) for AMD Family 15h Models 00h-0Fh Processors"

This is potentially a good sign if what we saw from the Phenom launch is anything to go by. Many of you would recall When the first B2 stepping processors in the Phenom lineup were found to be flawed by the TLB (translation lookaside buffer) issue, which incurred quite the performance hindrance of at least 10%. It wasn't until AMD rectified the issue with the launch of the B3 stepping Phenoms four months later that Phenom became a lot more attractive.