Thread: Anandtech News

Paul Otellini showed this tablet off earlier today but we got a little more detail on it just moments ago. Intel has put together a 32nm Atom (Medfield) based reference design which will be available to OEMs should they be interested in it. The only physical specification we have is its thickness: 8.9mm.

Intel is running a build of Honeycomb on the device however it's still in alpha. Intel is currently working on improving performance, power consumption and stability of the platform at this point. Intel is expecting to see Medfield based designs in the market in the first half of 2012.
I'm not expecting huge performance out of Medfield, but Silvermont will likely bring out-of-order execution to Atom and really start to redefine its performance levels in 2013. At this point all Intel needs is a design win so it can start iterating and executing in this space.


More...

Amidst a full day of updates from IDF, I'd like to draw your attention to some more Intel-related news: Intel has refreshed its generic Sandy Bridge graphics driver to version 15.22.50.2509 for 32-bit Windows Vista and 7 and version 15.22.50.64.2209 for 64-bit Windows Vista and 7.
You can read the full release notes from Intel here, but I'll try to highlight some of the more interesting fixes and improvements: non-gaming-related fixes include fixing crashes when using an HDMI monitor, WebGL rendering problems in Firefox and Chrome, and various OpenGL optimizations. On the gaming side, several games have issues with "rendering artifacts" addressed, and a few games have had their performance improved when compared to the previous 15.22.1.64.2361 driver released in April.
The new driver should support all desktop and mobile variants of the Intel HD 3000, Intel HD 2000, and Intel HD Graphics processors included in Intel's Sandy Bridge processors. The Intel HD Graphics chip in older Westmere-based processors is also supported, but no fixes or enhancements targeting it are listed in the release notes.
Gaming on Intel's integrated graphics is generally not an optimal experience, but to anyone stuck with a low-end or ultralight computer that lacks a dedicated GPU, these drivers should improve the situation somewhat. You can download the 32-bit drivers here, and the 64-bit drivers here. Windows XP (32-bit here, 64-bit here) drivers are also available, but these are maintained separately from the Windows Vista and Windows 7 packages and include a different set of improvements.
Finally, Intel notes that specific OEMs may make changes to the generic driver to suit their purposes and that the generic driver omits these customizations; I've never had a problem running Intel's generic drivers on systems I've worked with, but your mileage may vary.


More...

Earlier this morning Paul Otellini showed off Intel's Android smartphone reference platform running a version of Gingerbread. We just got up close and personal with the device as you can see in the gallery below. The phone is 9.5mm thick and features Intel's 32nm Medfield SoC although there's no word on the specifics of the architecture. Intel also told us that Medfield includes a newer generation GPU, but it's unclear whether Intel is using something as powerful as the PowerVR SGX 543MP2 or if it's a bit more tame in the graphics department.
Intel did give us a demonstration of the new ISP (Image Signal Processor) in Medfield that's capable of capturing up to 20 fps at 1600 x 1200. The reference platform could only deliver 10 fps from its 8MP rear sensor, but there's apparently headroom in the architecture.
Intel is claiming competitive battery life/power for Medfield but it's unclear how it will perform compared to the next-generation of ARM based SoCs (Krait, Kal-El, A15). We won't see the first Medfield based solutions until the first half of 2012. Apparently most device manufacturers weren't too pleased with Intel's close partnership with Nokia and Google was likely unhappy with Intel's work on MeeGo. With the Nokia partnership dissolved, Android prioritized and a unified design infrastructure, we may finally see some real momentum behind Atom in smartphones starting next year.
{gallery 1373}

More...

After at the Samsung Galaxy S 2 launch announcement, I made an educated prediction that the T-Mobile variant would come with a Qualcomm SoC and not Samsung's own Exynos. The combination of no hands-on, T-Mobile's desire to have a DC-HSPA+ phone for its DC-HSPA+ network, and also talk of the Samsung Hercules all clued us in. Further, to use Qualcomm's MDM8220 would necessitate use of a Qualcomm SoC to enable voice. Just as we predicted, Samsung today indirectly confirmed our suspicions by tweeting that the T-Mobile SGS2 will indeed include a 1.5 GHz Qualcomm APQ8060 SoC. The GalaxySsupport account purports to be official Galaxy S support for Samsung.

As a reminder, the APQ prefix in Qualcomm's lineup denotes an AP-Only (only the Application Processor inside) SoC, with no cellular baseband. The reason for the T-Mobile variant including an SoC without baseband is that it is highly likely to include an MDM8220 for DC-HSPA+ connectivity, which would make it the first shipping DC-HSPA+ enabled smartphone.
Source: GalaxySsupport


More...

We're still working on our deep dive on Intel's Ivy Bridge architecture but for those who are interested the slide deck that Intel just presented is in the gallery below. Expect more details in our full overview.
{gallery 1374}

More...

Ivy Bridge has three new features for overclocking:

1. The max CPU ratio is now 63x up from 57x
2. You can now adjust CPU ratios without a reboot, just via a register write.
3. DDR3-2800 will be the maximum DRAM frequency.

More on Ivy Bridge later today!


More...

We just finished going over the GPU portion of Ivy Bridge in Intel's IVB tech session. If you're interested, the slides are below.
{gallery 1375}

More...

IDF isn't a traditional trade show in the sense that the majority of the show isn't built around a wide open show floor with tons of booths and product demonstrations. The majority of the show is built around Intel's own tech sessions that offer the audience information on things Intel is involved in. There is a daily tech showcase however to allow Intel's partners to show off some of their goods.
Although Sandy Bridge E is still a couple of months away from launch, Intel's partners were allowed to show systems as long as they didn't divulge clock speeds or allow for anyone to play with the USB ports. Both Gigabyte and MSI had boards on display which you can check out in the gallery below.
Major themes? Cheaper boards have 4 DIMM slots (1 DIMM per channel) while the more expensive boards will have 8 DIMM slots (2 DIMMs per channel). All mentioning of PCI Express Gen 3 seems to have been forbidden (note the last image where PCIe Gen 3 has been marked out on the board). There are currently no PCIe 3.0 devices on the market which makes validation a bit of a problem at this point.
Pricing is still unknown but we can expect motherboards to be priced north of X58 solutions.
{gallery 1376}

More...

I dropped by Kingston's booth at the IDF tech showcase to check out two things this evening: Kingston's SSDNow KC100 and another Sandy Bridge E demo. The KC100 is another SF-2281 SSD but aimed at business users with a 5-year warranty instead of the 3-year warranty that comes on the HyperX. Performance should be identical to the HyperX. SandForce has a new firmware revision that is in testing now (3.30) which should fix some issues users have been having, but no word on whether or not it'll address all issues at this point.
Kingston also populated a Sandy Bridge E motherboard with 8 x 8GB DIMMs just to show what's possible with Intel's next high-end enthusiast platform.
{gallery 1377}

More...

At the start of 2011 Lucid announced their Virtu software to go with Intel’s Sandy Bridge CPUs. With Virtu users would be able to use a discrete GPU and Sandy Bridge’s integrated GPU simultaneously in order to use the features of both GPUs. This normally meant either using a dGPU as the primary GPU while still being able to access Sandy Bridge’s QuickSync (D-mode), or using the iGPU as the primary GPU while having the dGPU power up and render frames to the iGPU when the dGPU’s capabilities were needed (I-Mode).
More recently in June at Computex they announced Virtu Ultimate. Virtual Ultimate added support for using Virtu with AMD’s integrated GPUs, and introduced the company’s Virtual Vsync technology for preventing screen tearing without using vsync.
Lucid has continued to work on Virtu since then, and finally at today’s IDF they’re announcing their latest rendition of Virtu: Virtu Ultimate MVP. Virtu Ultimate MVP adds Lucid’s latest technology, which the company is calling HyperFormance technology. Lucid is claiming that with HyperFormance they will be able to significantly improve I-mode rendering performance by “removing redundant rendering tasks”, or in essence they will be improving performance by rendering less.

So what exactly is HyperFormance? That’s a great question, and we don’t have a great answer. Lucid has decided to keep the mechanisms of HyperFormance under wraps for patent purposes, so what we have to work with amounts to a very high level overview. In this overview Lucid has made it clear here that the principle feature of HyperFormance is that it’s reducing the rendering workload on the dGPU by intercepting and removing some rendering tasks within a frame, which in turn would improve framerates by allowing a GPU to finish faster and move on to the next frame.
The long and short of it is that Lucid seems to have integrated parts of Hydra – their largely ignored multi-GPU rendering technology – into Virtu, and are using Hydra’s ability to split up drawing commands to remove drawing commands heading to the GPU entirely. The unknown part of the equation is how Lucid is deciding what to remove. Based on what we’ve seen of Hydra it seems unlikely that this is a hidden surface removal technology, in which case it’s more likely that it’s an attempt to calculate and only render parts of an image that have changed from a previous frame – a GPU form of delta/inter-frame compression.

What’s also clear is that HyperFormance goes hand-in-hand with Lucid’s Virtual Vsync technology, which appears to do something like that (as with HyperFormance, they aren’t saying how Virtual Vsync works either). The significance here is that while HyperFormance would be looking at a frame based on draw commands, Virtual Vsync works off of a finished frame. So HyperFormance is getting some kind of feedback for future frames based on Virtual Vsync’s analysis of the completed frames.
Beyond this we know very little about the technology, or more importantly what the pitfalls may be. The immediate concern is that not rendering something could very result in the final composited image being wrong in some manner, or that doing all of this analysis may increase input lag more than a faster framerate decreases it. More interestingly though, because they need to analyze both draw commands and the rendered frame, HyperFormance only works in setups with multiple GPUs (i.e. iGPU + dGPU), as Lucid needs to be able to completely control the input and the output of the dGPU. So I-mode could potentially be faster than using just a discrete GPU, and Lucid is promising just that. It certainly seems farfetched, but not impossible with the little information we have to work with.
In any case as Lucid has just announced Virtu Ultimate MVP, HyperFormance is not yet available for testing. Once Lucid makes it available for review it’s definitely something that bears following up on, so stay tuned.


More...