D
Deleted member 13524
Guest
I'm pretty sure that what Jen-Hsu Huang meant was that Kepler's "footprint" will be in an upcoming Tegra SoC in the same way as the Adreno family has bits from Xenos, nothing else.
One has to keep in mind that with the coming of Windows 8, the target platform for Tegra won't be restricted to the sub-2w range anymore. Clamshell fall in the 5-10W range.
True. Yet Jensen's comment included superphones.
Now that the HTC One S/X's reviews are coming up, I have to say I'm impressed at how nVidia pulled off the marketing stunt for quad-core.
All the uneducated reviewers seem to underestimate all dual-core SoCs (even the superior dual-core Kraits!) because they're not quad-cores like Tegra 3, as if it won't be high-end enough if it's not a quad-core.
I wonder if this will put pressure on releasing quad-core Cortex A15s on all vendors.
verge;In my stress test with continuous streaming video at maximum loudspeaker volume and maximum screen brightness while connected to both Wi-Fi and 3G, I got four hours and 22 minutes of usage before the phone turned itself off. In a more reasonable test involving "normal" usage, the One X yielded 13 hours and 38 minutes, including a half hour of continuous Riptide gameplay (which looks amazing on this processor and display, by the way) and well over a dozen benchmark runs of various types.
Now that the HTC One S/X's reviews are coming up, I have to say I'm impressed at how nVidia pulled off the marketing stunt for quad-core.
All the uneducated reviewers seem to underestimate all dual-core SoCs (even the superior dual-core Kraits!) because they're not quad-cores like Tegra 3, as if it won't be high-end enough if it's not a quad-core.
I wonder if this will put pressure on releasing quad-core Cortex A15s on all vendors.
CPU performance for both phones is great and way above anything we've seen before. Battery life is acceptable for both. Graphics performance is very similar. The One X has a great screen. The One S has a pen-tile. Yuck.ToTTenTranz said:Now that the HTC One S/X's reviews are coming up, I have to say I'm impressed at how nVidia pulled off the marketing stunt for quad-core.
All the uneducated reviewers seem to underestimate all dual-core SoCs (even the superior dual-core Kraits!) because they're not quad-cores like Tegra 3, as if it won't be high-end enough if it's not a quad-core.
I wonder if this will put pressure on releasing quad-core Cortex A15s on all vendors.
CPU performance for both phones is great and way above anything we've seen before. Battery life is acceptable for both. Graphics performance is very similar. The One X has a great screen. The One S has a pen-tile. Yuck.
I don't know what your concerns are, exactly, with the Tegra 3 in the One X, but after reading both reviews I don't think there's much doubt that the One X is better than the One S?
I don't see a marketing stunt, one either side. Just 2 very capable phone SOCs?
Well, the quad-core is a technical reality, not a marketing stunt. They'd be crazy not to make a point of it.ToTTenTranz said:Don't get me wrong, of course I know both SoCs have stupendous performance for a smartphone. It's just that I'm seeing way too often comments like "the new Snapdragon S4 performs well, too bad it's not a quad-core CPU + 12-core GPU like Tegra 3".
Well, the quad-core is a technical reality, not a marketing stunt. They'd be crazy not to make a point of it.
A more interesting question would be to lock down 2 of the 4 cores and see if it makes any difference in real use cases. Since this is Android and thus totally open, it should be a piece of came to try this out.
Don't these optimizations end up in the open source tree eventually? Or is this similar to the binary modules of the regular Linux kernel?metafor said:While Android is open, the various versions the manufacturers run aren't. And love or hate them, they do bring about many performance optimizations -- particularly at the driver level -- that vanilla Android doesn't have access to.
Don't these optimizations end up in the open source tree eventually? Or is this similar to the binary modules of the regular Linux kernel?
While Android is open, the various versions the manufacturers run aren't. And love or hate them, they do bring about many performance optimizations -- particularly at the driver level -- that vanilla Android doesn't have access to.
Similar to what ASUS did with Transformer prime? and Sammy with GS2?
Seems to be implying that nVidia will be using the same core muxing strategy they used on Tegra 3 but with Cortex-A15s. If that's really what it means then I don't buy it. It works on Tegra 3 because the L2 cache is decoupled from the cores. That's no longer true with A15, so a "companion core" would be a heavy modification of what ARM's offering, and I doubt they're interested in providing such a thing instead of big.LITTLE.
Five A15s would also put them at a big die area disadvantage vs dual A15 SoCs, either in having much less to spend on other things like the GPU or IMC, going with low cache values (I can see it now, 4 cores sharing 1MB of L2 cache again..), or offering much bigger dies. Across their entire A15 product offerings, apparently.
I also wonder where that third option (screen size?) even fits into this. Are none of these products supposed to be meant for phones at all?