LinkBack URL
About LinkBacksRead more.The first test engineers thought that the Windows 10 battery meter had a bug.
Read more.The first test engineers thought that the Windows 10 battery meter had a bug.
*Shocking News* Low power SoC means longer battery life... as expected? Those test engineers must have been unaware what they were testing.
Tablets/netbooks using ARM SOCs or the previous Atom solutions have always had generally better battery life than laptops using a much more performant Core i3/5/7... that is the whole point isn't it - that they have long battery life and just about acceptable performance for lightweight tasks?
These ARM Windows laptops are just a refinement of the old netbook idea. The only 'innovation' here is Windows running on ARM with x86_64 emulation.
I thought the backlight was the main power sink when watching videos. Are these figures with the backlight turned right down?
A nice information, to finaly have full windows 10 on ARM PC for semi light workloads. That may put some more presure to Intel and AMD, which is a good thing.
Despite kingpotnoodles pessimism i think this is pretty great. Its the combination which is a boon; we've had notebooks before but never with this much battery life and the cellular capabilities of having a snapdragon soc whilst able to run windows.
Still I wonder how smoothly the emulation will work and how limited performance will be.
It is only the Windows bit which is really new. I presume this just has a 3 times bigger battery than a 10 hour Chromebook: http://chromebookreview.org/hp-chromebook-14/Originally Posted by chj
ISTR the emulation is only 386, presumably because the 386 is out of patent protection, so I wonder how useful this will really be.
More to the point, does it run U-boot or Coreboot so can I put Linux on it?
For me, the best part is that it can do some stuff (like watching youtube) that my 2011 MacBook Air struggles with when it starts to thermally throttle. This is very frustrating when I know my phone can do these things with ease. Whilst I can encode video on my Air (before it throttles itself to performance oblivion or ends up on the floor because it has made you jump by scorching your genitals), the odds of doing anything more taxing than web browsing, word processing and watching a video or two are very slim. If I could get what amounts to a mobile phone with a bigger screen and amazing battery life I'd be actually pretty happy. The problem of course is that my mobile is a £700 relative beast (S8) whilst for a laptop that only does the very basics you'd want to be paying maybe £300 tops. For most people I'm not sure they'd want to pay a lot of money for something so limited in performance. I'm no expert in this area but I thought decoding video, etc used specific instruction sets and hardware which allowed it to have smooth playback even on low powered devices. Would these things still work properly if you're running it as an x86 emulator?
A good compromise might be resurrecting the old idea of a "shell laptop" that amounts to a screen, keyboard and battery that you can plug your phone into and use as the processor but I think for that to actually work you'd need the hardware to be standardised and for Google to ensure the software capability is baked straight into Android. I think if an individual manufacturer did it, there wouldn't be the market share (i.e. enough people with the particular flagship phone required and who need a new laptop of the specification on offer who are also willing to sacrifice their ability to use their phone whilst using the laptop) to allow it go gain traction. The other thing of course is that you can already do this by getting a tablet and a case with a Bluetooth keyboard - I do this and it allows me to use it very much as a low end, but very capable laptop or as a tablet (and the keyboard is usable with my tablet). Add in my portable battery pack and that'll last me all day (and the next) quite happily, even watching films. If I'm in a hotel room without wifi, I just share my phone's internet connection, set my tablet up in the laptop configuration and attach both to a charger (or my battery pack if there's not a socket conveniently located).
Considering the above, I do wonder how much of a market there is for this kind of thing. The only real advantages are having proper word, excel, etc, but the mobile versions really aren't that bad. The other advantage is true multitasking but the latest version of Android does this pretty damned well enough anyway, especially given the screen real estate limitations you're going to have on an ultraportable laptop.
</ramble>
The market, for me, is having one of these devices and travelling a lot. Document writing, mild store gaming, means that a device such as this is perfect for the person that travels or just sits on the sofa not wanting to do anything special. It just might start to replace iPads.
It should be interesting to see how Intel reacts when this reaches market, they can be rather touchy about their X86 license and when XDA-Developers reached out to them back in June they said the following...
Intel respects intellectual property rights and we expect others to do the same. x86 technology is both proprietary and central to our business, and we’re concerned any time it appears that others may be copying it inappropriately. We will thoroughly evaluate any products that claim to emulate x86 technology, and vigorously enforce our intellectual property rights if we believe they are infringed.
It uses video device drivers, where you basically pipe data into it and tell it which bit of the screen to scale the output onto. So an x86 emulator could just tell the ARM device driver and that bit can run native, so it is an ideal case.
Having said that, the likes of Chrome have been native ARM for some time.
Well it would be nice if Intel have started respecting other people's intellectual property rights. Oh hang on, they didn't actually say anything about other people did they, perhaps no change then
I can't see how Intel can do anything about basic emulation. Apart from it being old technology that has been done on SPARC, DEC Alpha etc processors in the past to run DOS and then Windows code the 386 is now a really old bit of kit so patents won't help them and they will have to try and claim some sort of copyright infringement. They could bog Qualcomm down with lawyers until the product goes away, ultimately lose the lawsuit and ultimately pay out a couple of billion just before they get a ruling against them so they can claim no fault was ever found in their actions.
Edit: Stuff like SSE will still carry patent protection, I have to wonder just how useful emulation of old 386/x87 code would be in the modern world.
May X86 die. It is time we need an open source architecture
Are you telling me my 2w phone has longer battery life than my 180w i7?? NO WAY!!!!!
This article is void of information. It's like comparing a n3150 based laptop with a i7 based laptop. One is a "mobile" Celeron, meant for light tasks, the other is a powerhouse meant to crunch numbers. Yes, both will run a 480 movie and yes the i7 will deplete a battery faster, but the i7 will also run 4k and the n3150 will maybe even crash under the load, 1fps if you're lucky...
30 hours on battery means nothing, benchmark that cpu against a similar powered intel/amd and then we have something we can talk about. Till then, this is vapourware.
Wouldn't trying to avoid all the extensions and their associated pattens be a bit like trying to tip-toe through a minefield, added to that is as you say how useful a 30 odd year old ISA would actually be for running modern day programs.
Cheers!It uses video device drivers, where you basically pipe data into it and tell it which bit of the screen to scale the output onto. So an x86 emulator could just tell the ARM device driver and that bit can run native, so it is an ideal case.
Having said that, the likes of Chrome have been native ARM for some time.
Just what's needed to help the locked-down windows edition compete with chromebooks
Smartphone chipsets are generally much better than i7's at running video, IIRC intel have only just caught up with hardware decoding of 4k. Also, with modern power gating, better performance in a benchmark doesn't justify worse power consumption under light load to this level
There are currently 1 users browsing this thread. (0 members and 1 guests)
Posting Permissions
Reply With Quote
