I can almost, with 90% certainty saythat the p2 should use an amd apu!

[comradekingu]

So if you halve the size of that heatsink, what are you left with, (other than something that isnt mobile anymore.)
Remember you are trying to put that in a enclosed space.

There is a reason you only hear of amd tablet on amd press releases.
4 watt amd-tpd is at least ten times the amount of power you want a full set of just the a9 cores to use peak, and thats where amd cant compete with AMD64. And if you can afford to go a15, why would you even with amd, i dont even. The power scaling is where i think you will run into the most problems. A top notch krait 400 uses 2.5-3W for the whole SoC and you wont see bulky heatsinks on those due to the aforementioned problem. How about being able to accommodate a screen without under-clocking everything.
Also, there are no good drivers for for current amd hw.

tl;dr see you in 2-4 threads.

Edit: How many hours does the Quanta BZ1T last?

Edit2: From my understanding of how nobody builds amd tablets to how nobody wants to buy them, i pondered the idea as to why it seems to be difficult to find real life usage numbers. And these things seem to all be closely connected.

Only thing i could find was "up to 12 hours of idle battery life" which is respectable, werent it for the fact its idle, once you start _actually using it_ you are looking at 5 hours tops, which is beyond bad.

 

[monstercameron]

ok here is closure...

OpenGL: 7,18 FPS Avg power usage: 10 Watts
CPU: 0,45 PTS Avg power usage: 9 Watts
(Cinebench: A4-1250 9 Watt TDP: OpenGL: 9FPS, CPU 0,48 PTS, -geleende externe data-)

http://tweakers.net/productreview/83223/medion-akoya-e1318t.html amd

http://www.notebookinfo.de/artikel/bay-trail-benchmarks-zum-intel-atom-z3770-fuer-windows-8-und-android-tablets/3995/ z3770

fixed

 

[comradekingu]

That was two links to the same AMD machine

 

[monstercameron]

another revelation

[toshiba sat. click review] http://www.laptopmag.com/reviews/laptops/toshiba-satellite-click.aspx

[toshiba sat click product page] http://www.toshiba.com/us/computers/laptops/satellite/Click/W35Dt-AST2N01

Without the keyboard dock, the Click ran for just 4 hours and 1 minute [workload: Continuous Web surfing over Wi-Fi until the battery is completely drained]

Dual Lithium Ion (23.6Wh, 2-Cell)—*Not user replaceable

unknown]

so that is around 6W soc+platform power [reminder that there is a 5400rpm HDD and a 13.3" ips display]

 

[_wb_]

another revelation

[toshiba sat. click review] http://www.laptopmag.com/reviews/laptops/toshiba-satellite-click.aspx

[toshiba sat click product page] http://www.toshiba.com/us/computers/laptops/satellite/Click/W35Dt-AST2N01

Without the keyboard dock, the Click ran for just 4 hours and 1 minute [workload unknown]

Dual Lithium Ion (23.6Wh, 2-Cell)—*Not user replaceable

so that is around 6W soc+platform power [reminder that there is a 5400rpm HDD and a 13.3" ips display]

How can you deduce anything from those numbers? Unknown workload, a display that is probably consuming most of the power anyway, many unknowns (e.g. hard disk).

For all we know, that SOC could use 5W just to be idle, and it would be consistent with the above numbers. Or it could use only 0.5W at high load, and that would also be consistent with those numbers.

It is not that hard to benchmark the power consumption of a SOC, at least to get a ballpark number. Something like what I did in this thread:

http://boards.openpandora.org/index.php/topic/14436-perfwatt-comparison-between-rebirth-and-1ghz-units/?hl=%2Bperf+%2Bper+%2Bwatt

 

[monstercameron]

How can you deduce anything from those numbers? Unknown workload, a display that is probably consuming most of the power anyway, many unknowns (e.g. hard disk).

I updated it, this was just a browser test, though not alot of setup info was given.

For all we know, that SOC could use 5W just to be idle, and it would be consistent with the above numbers. Or it could use only 0.5W at high load, and that would also be consistent with those numbers.

very true as there isnt any analysis as to where the power is going in the testing, but I am fairly confident that the HDD and display are eating a significant portion of the 6W power draw...

this is just an idea of what this soc is capable off...I am still searching for more reviews and testing.

 

[monstercameron]

http://www.anandtech.com/show/7514/amd-2014-mobile-apu-update-beema-and-mullins

interesting, it seems AMD has not let me down...cant wait for shipping products!

 

[comradekingu]

I like the prospect of TrueAudio.

 

[3_bit]

Now looking at the now outdated Z-60's specs... I imagine you could really grab some x86 linux gamers (because TF2, Gmod and Minecraft do run on Linux) as I used to game with a 1st gen A8 APU computer with just the internal GFX and got really nice frames... I'd pay $800 if I could get 4-6GB of RAM and at least 64 GB of storage, as well as 3G and a camera facing out.

 

[monstercameron]

samsung ativ book 9 lite with amd a6-1450

http://www.notebookcheck.com/Test-Samsung-ATIV-Book-9-Lite-905S3G-K01DE-Subnotebook.101920.0.html?utm_source=TwitterFeed&utm_medium=twitter
 

Economical in the stress test was only the Asus T100 (Baytrail atom), which draws only 11.8 watts from the mains. In idle this device requires even only 1.7 watts. 

under load the asus t100 with atom z3740 can use upto ~12W and idles ~2W...in a mobile optimized tablet formfactor

Compared to the Acer Aspire V5-122P , with the same SoC inside, the power dissipation of 6.0 watts in idle by almost 50 percent below the mentioned competitors.

How can already hinted at the low temperatures, the Samsung ATIV Book 9 also shine off the idle operation: Our review unit needed under full load (Prime95 + Furmark) and WLAN enabled, almost unbelievable 13.4 watts

so under max load the amd a6-1450 uses ~13.5W and idle can range from 4-7W. but in a notebook formfactor.

The Surface 2 does not consume much energy during idle with 1.8 - 4.1 Watts. Here we can see the benefit of the Tegra 4's frugal companion core. We can measure up to 15.8 Watts under load if we run the Relative Benchmark with maximum display brightness; the less powerful Nexus 10 (up to 9.4 Watts) is significantly more frugal.

Our review unit consumes 7.2 Watts during video playback (1080p), both the Nexus (9.2 Watts) and the Toshiba eXcite (9.3 Watts) need more energy, despite the identical SoC. The power supply unit is more than sufficient with 24 Watts, even under maximum load. 

Nvidia's previous SoC, the Tegra 3, was already not renowned for its energy-efficiency. The Tegra 4 seems to carry on this tradition: While measuring an acceptable 2.5 to 6.1 watts during idle, we measure an enormous 9.3 to 13.2 watts during full load. Even the iPad 4 is with a maximum of 12.5 watts a little bit less power thirsty, let alone the Nexus 10 (9.4 watts) and the Xperia Z (5.6 watts). The included power supply is also surprisingly powerful: 36 watts would be sufficiently sized for some ultrabooks.

for reference tegra 4

http://www.notebookcheck.net/Review-Toshiba-Excite-Pro-AT10LE-A-108-Tablet.98828.0.html toshiba

http://www.notebookcheck.net/Review-Microsoft-Surface-2-Tablet.105780.0.html ms surface 2

When power consumption , the Pipo Max M7pro arrived too vigorously . In class comparison, the tester cuts consistently worse off than the competition . In Idle mode, we measure 4.3 to 7.2 watts under full load values ​​increase to a whopping 9.8 to 11.3 watts.

for reference rockchip rk3188 http://www.notebookcheck.com/Kurztest-Pipo-Max-M7pro-Tablet.102451.0.html

So the a6-1450 is very frugal, it goes to show that amd parts aren't as bad as many would think.

the a4-1200 while it isn't as powerful, should still perform well, use relatively little power and allow x86 apps to operate.

 

[comradekingu]

Worst case whole SoC vs something thats worse on its own.