Not useful information unless we know the battery sizes with dock and without dock.Asus Transformer with Baytrail: 11h
Asus Transformer with Tegra 4: 16h.
Hm. D:
Yes this soc is missing sata and pci e. The tablet version will not have those two features for they destroy idle power consumption, the laptop and desktop versions will have sata and pci e.
Not useful information unless we know the battery sizes with dock and without dock.
If you do not know Asus is moving to smaller batteries inside their transformer docks.
So decent CPU performance in some tests as expected combined with embarrassing GPU performance. They probably needed to castrate GPU to get decent power numbers.
Am I missing something or is this SoC missing SATA and PCIe?
One must not forget - all of this has happened in two product iterations after the decision to focus on mobile was made.
Intel gave us about 2 hours of hands-on time with two reference-design tablets outfitted with 2.4GHz, quad-core Atom Z3770 processors and 2GB of DDR3L/1067 memory (one running Windows 8.1 and the other Android 4.2.2). The devices had 8-inch screens with a native resolution of 2560 by 1440 pixels. While Intel does not intend to bring these tablets to the consumer retail market, Walker said that they are representative of what Intel’s OEM partners will deliver later this year.
It came as no surprise that the Android version was capable of playing simple arcade-style games, such as the side-scroller Jetpack Joyride, but it was a revelation to jump into Valve’s multiplayer first-person shooter Team Fortress 2 on the Windows 8.1 version and experience relatively fluid gameplay on an 8-inch tablet.
AMD just announced an 8W TDP APU with 66% of CPU performance of A4 5000 (the 15W TDP chip that you see the tablet oriented Z3770 beating more often than not in the CPU tests) and 60% of A4 5000 GPU performance... you can do the math yourself, this would put it slightly ahead (graphics) of a Bay Trail-T chip that's likely using quite a bit lower power under load. Hardly amazing GPU performance at <10W TDP levels from both x86 giants but far from embarassing.
dont jump the gun, we havent seen any power numbers yet.
That's a <3W SDP for Temash. Bay Trail won't touch it in graphics, and that's on a process node advantage lacking both SATA and PCIe.
This and the fact that all thing being equal Jaguar has rougly
30% and 50% better IPC in integer and FP respectively.
AMD just announced an 8W TDP APU with 66% of CPU performance of A4 5000 (the 15W TDP chip that you see the tablet oriented Z3770 beating more often than not in the CPU tests
That's a <3W SDP for Temash. Bay Trail won't touch it in graphics, and that's on a process node advantage lacking both SATA and PCIe.
Why is that particularly important? Isn't performance IPC * Frequency? Normally trying to run at a higher frequency is a power problem, but the early reports for Silvermont suggest that's not an issue here.
Thanks to a 2.4ghz frequency both in ST and MT tests.
We ll see once verifiable testing environments are published
on diverse sites.
Why is that particularly important? Isn't performance IPC * Frequency? Normally trying to run at a higher frequency is a power problem, but the early reports for Silvermont suggest that's not an issue here.
Thanks to a 2.4ghz frequency both in ST and MT tests.
We ll see once verifiable testing environments are published
on diverse sites.
3-4W during video playback doesn't seem very good? http://techreport.com/review/25329/intel-atom-z3000-bay-trail-soc-revealed/4
That would be a valid concern if SoC power consumption wasn't still relatively low (2.5W at cinebench r11.5 multithreaded).