Asus Promises 12-Hour Battery Life In New High-End Laptop
Asus' new high-end laptop could finally be the traveler's best accoutrement, touting twelve-hour battery life thanks to intelligent, second-by-second switching between the two GPUs and automatic, on-the-fly re-clocking of the Intel Core i7 CPU. All this also comes in with a price tag of just over $1,000. "ASUS's solution is different because it's user-transparent; even a novice user will get the fullest possible benefit because the laptop itself is deciding when to switch. The same principle applies to the dynamic CPU clocking. ASUS includes a desktop widget to track CPU clock speed. While using the UL80JT, I could see it moving up and down with what I did; up with program openings and CPU-intensive processes, and way down at idle. Between the GPU switching, dynamic clocking, and ASUS's other power management features, the UL80JT manages to consume less than half as much power as the unibody Macbook while browsing."
$0
$100
$200
$300
$400
$500
Take a check?
Does it double as a dinner plate?
"Dynamic clocking" my foot! I won't buy it unless it has a big, red "Turbo" button.
Today's Sesame Street was brought to you by the number e.
Why hasn't anyone come up with this before? Or if they have, where are the others?
Promising 12 battery life is one thing.
Actually delivering acpi that is not crap is another.
I guess we'll wait and see.
Vendors promise all sorts of things. That doesn't make them true. I'll believe it when I see Tom's Hardware or someone equally competent test one of these things and they actually get 12 hours.
Until then, I'll file this one under "vendors promise the world".
My blog
This thing is as old as my beat up Pentium III Inspiron 5000. Varying GPU clocks is also old.
What is interesting is seamless switching between GPUs. Everything else is just marketingese for "we do what everyone else does and we actually bothered to put some extra effort into power optimization".
I know netbooks have really pushed the low-end prices downwards, but is slight over a grand really considered "high-end" these days?
thanks to intelligent, second-by-second, switching between the two GPUs and automatic on-the-fly re-clocking of the Intel Core i7 CPU. All this also comes in with a price tag of just over $1,000.
So what they're saying is... as long as you don't use the laptop for anything more sophisticated than notepad, you'll get 12 hours. How about watching DVDs? How long will it last then? -_-
#fuckbeta #iamslashdot #dicemustdie
God I love asus, glad to see them actively pushing the envelope on the Windows side. Hopefully they'll kick apple into higher gear with this kind of stuff too. Their products themselves look and feel slick too, not a cheap part to be found.
ASUS's solution is different because it's user-transparent; even a novice user will get the fullest possible benefit because the laptop itself is deciding when to switch. The same principle applies to the dynamic CPU clocking.
So what they mean is that the laptop will be deciding when it should be fast or slow, with no input from the user? How's this different than the gazillion power management settings we have now (except switching between GPUs of course)?
I am also not sure I like the sentiment of "user-independent" is somehow more beneficial to the user. It sounds too much like the drivel from the RIAA/MPAA: "we will enhance customer value by increasing the price and decreasing what they can do with it."
These guys always seem to want to show speed and power in a laptop -- but what I need in a laptop is long battery life. How much CPU does it take to do a bit of web browsing, run up emacs & ssh. I have a PC at home or stuff that I ssh to if I need to do fast compiles or run databases & other heavy stuff. These guys just don't get it, I thought they had when they brought out the original eeepc -- but subsequent models have just turned to bloat (OK: I do like the larger screens, but that is all).
I have an Asus U80 laptop ($650) with a stated battery life of 7 hours (without Wifi). I've gotten 6 hours with Wifi and 8 hours without. Even with fairly heavy web browsing and CPU usage, I can get 4-5 hours in Battery mode. Thus, while many battery life claims are bullshit, I am inclined to believe Asus. Note that Asus uses some proprietary Windows software to reduce power usage. Without the software, the battery lasted 33-50% less.
As an aside, they also have excellent RMA service. I discovered that my laptop drive had several bad sectors. I called Asus, and after less than a 5 minute wait was talking to a human being. I explained that a low-level drive scan showed several bad sectors on the drive, and that this indicated a hardware rather than software problem. Rather than having me reinstall Windows, or some other bs, I was told I could return the laptop or the bare drive for service with a 2-3 day turnaround. I shipped the bare drive, and received a replacement 2 business days after they received the RMA drive. Not bad.
Yeah, and we'll have a man on Mars in 2035, Obama will change Washington, and Duke Nuke'em Forever will be out any day now.
SJW: Someone who has run out of real oppression, and has to fake it.
ASUS includes a desktop widget to track CPU clock speed. While using the UL80JT, I could see it moving up and down with what I did; up with program openings and CPU-intensive processes, and way down at idle.
So basically it's like the task manager?
i already can watch my GPU and my CPU speed change as i do things....(im a linux/gnome user BTW...)
epic sig..... ya i got nothing
Saving power by way of second by second switching? I know it doesn't really apply, but Maxwell's Demon comes to mind.
EPIC WHOOSH
Looks to me like the software to accomplish this is one of those programs the manufacturer bundles on your computer, not an architectural change. If I have to tolerate a 6 month trial of MS Office, Norton Antivirus, several dozen casual games distributed as adware, and whatever other "productivity" software they decide I want, then no thanks. Bundled software should be possible to separate from each other.
Another important question: will it run Linux?
I sometimes ask revealing, often ignorant-seeming questions. Maybe they're harder to answer than you think.
People still buy $1000 laptops? How quaint.
Can't long endurance laptops be easily built by designing in a larger capacity battery?
How large a proportion of the laptop's cost is the battery?
Surely if there is a big demand for battery life the manufacturers can just make a thicker, heavier version of a current laptop for a hundred bucks more?
They whose government reduces their essential liberties for temporary security, receive neither liberty nor security.
And the first thing every executive that gets it will do is change all the power saving features because the screen is not bright enough for them.
WTB Mod: (-1) Feeding the trolls.
So, they have basically redefined ondemand governor and build Windows variant for CPU Frequency Applet I used to use in GNOME?
There's a review at:
http://www.silentpcreview.com/asus-ul80vt
They got a runtime of 9hrs 11min while web browsing, but it was running faster than expected so they thought that 10 hrs wasn't out of the realm of possibility.
12 hours then means 3 hours. Yay.
My laptop now is my mobile phone. It’s enough to watch movies with headphones in the bed, on the toilet or outside. And if I want more, I always have my full home cinema, server and workstation at home.
Any sufficiently advanced intelligence is indistinguishable from stupidity.
I suddenly want to dig out my old 286.
In all seriousness, with dynamic clocking, why use 2 GPUs and switch between the two? Why not just under-clock and under-volt the primary GPU when you're not gaming?
Doesn't the second GPU just add cost?
http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
Ok but does it run Linux?
A company can promise lots of things. Most companies hope that people forget those promises. The real question is whether they'll follow through or not.
A 310m is a mainstream graphics card for notebooks, not "high end".
Translation: this a mainstream laptop.
My Thinkpad T400 has two GPUs. One is an Intel GMA4500, the other is an ATI Mobility Radeon HD 3400. Running with the Intel GPU cuts off a full third of my power consumption because it uses the main CPU and main memory instead of dedicated chips. But switching between the two GPUs is iffy at best and usually requires a reboot. If I get stuck on the Intel GPU then I can't play games until I reboot and switch to the ATI GPU, so I usually stay on the ATI unless I really need battery life.
What is significant about what ASUS is doing is the PC will *automagically* switch to the high-performance GPU when you start up a game or a flash video, then switch back when you go back to word processing. This is something that has never been done before and is a major step towards making "switchable graphics" truly useful.
That is, of course, assuming that the ASUS power management app doesn't crash all the time leaving your system in an unstable state.
I have a T400 as well, but I can usually (like 98% of the time) switch between cards without a reboot. Are you sure you got the right drivers and stuff? I think if you remove all of the default Lenovo software, you end up with some problems. You need to keep the Lenovo battery/power management software (sweet! two battery gauges).
"Asus Promises 12 Hour Battery Life In New High-End Laptop"
did the forgot to mention "with display off"?
Marcin
I want 12 hour battery life in a low end laptop. Free phones will last 8 days on a single charge. I want my laptop to run for days, and I don't care if it doesn't have enough grunt to play Halo 3.
Oliver's law of assumed responsibility: If you're seen fixing it, you will be blamed for breaking it.
..when i got my first laptop in grade 8/9 (freshman in university now) i thought that all laptops had that long of a battery life:\
Silly wabbit. You're protected from having to choose whether or not you want to continue your service each month. I, for one, appreciate relief of this terrible burden as I have too many other difficult choices to make already: whether to get out of bed, whether to go to work, whether or not to kill myself... This brief respite from the tyranny of choices is a welcome break from choosing, and avoids the risk of anguish and self-torment at having once more chosen wrong.
Yes. If it works it's great. Wasn't it Dell that screwed up their CPU throttling a while ago, and a bunch of people ended up with laptops that decided to run at 2 MHz whenever you actually asked them to do something?
The lenovo T400 I have had for over a year now can switch between a low power Intel chip and GPU on command. With the 9 cell battery and ultra bay li-poly I saw actual usage well over 12 hrs of continous use before needing to recharge.
After all, their promises on the battery life of previous models have been so conservative!
Dell claimed theirs would last for 19 hours. Didn't believe them either. Testing is rarely done in real world usage scenarios...
http://www.pcadvisor.co.uk/news/index.cfm?NewsID=102941
My Panasonic Let's Note T6 was rated at 11 hours three years ago. With regular HDD. This was before the entire world started using stone age technology in "netbooks". Zzzz.
While the availability of seamless switching between the two subsystems would be very convenient, I think that the original question that Enderandrew asked is still valid: Why not just lean on the one heavier GPU exclusively, and re-clock *that* GPU in the same manner that the CPU is re-clocked, on-the-fly?
I think that the answer is that the current GPUs *can't* be re-clocked as easily, or as flexibly, as the new i7 CPUs are designed to be. Because if they were, it would seem that re-clocking should involve less risk, less complexity, and fewer components than the method of switching between two completely independent subsystems (especially since the two systems are manufactured by different companies, and you're going to need to update the drivers for each of them individually).
What if, for example, NVIDIA decides to release a driver down the road that interferes with the way that Asus designed the switching mechanism? If that happens, who gets the blame? NVIDIA would probably say that the drivers were meant to be used with their GPU in the manner that the GPU was *designed* to be implemented, and Asus may not have enough leverage to "force" NVIDIA to manage a branch of the driver implementation which still supports their switching mechanism. This is all speculative, of course.
Entomologically speaking, the spider is not a bug, it's a feature.
Most GPUs DO downclock dynamically to save power... take nVidia for instance: My POS GeForce 9300M GS has 4 modes... Standard 2D, Low Power 3D, High Power 3D and Full Power 3D.
Standard 2D (170MHz core, memory clocked down way far as well) draws about 15W at idle, while Full Power 3D (700MHz core, and IIRC about 1400MHz memory) draws about 22W at idle. Now take the same laptop with Intel integrated graphics - 11W at idle. Same settings, exact same components other than the graphics card.
"Performance" GPUs aren't made with heat and power savings in mind... or at least not nearly as well as onboard graphics.
Oh, and did I mention the nice little audio stutters and skips/pauses (high DPC latency spikes) every time the GPU changes clock speeds? Hope those aren't present on this newfangled switchable stuff (that's actually one of the reasons I'd like a manual switch - preferably a hardware one)...
A nvidia 310M is not high end. It's not even mid range.
I didn't say that they didn't auto-clock, I said that they can't be "re-clocked as easily, or as flexibly, as the new i7 CPUs are designed to be". I should have added a "probably" to that statement, since obviously I haven't looked into the granularity of the ability of different GPUs to re-clock themselves.
I think that the conclusion here is that GPU designs, including the ones meant to deliver power when it's needed, must also be designed to draw the minimal amount of current when they're idling. We're talking about a mostly parallel design when it comes to GPUs (broadly speaking), so it should be even easier to shut down modules that aren't needed. NVIDIA makes GPUs for cellphones these days, they should be able to allow the PC/laptop-targeted systems to drop into a near-dormant state. I'm sure this has quite a bit to do with intel's ability to manufacture at 32nm as well (or, more specifically, NVIDIA's inability to do so, yet).
Entomologically speaking, the spider is not a bug, it's a feature.
Ah, I misunderstood you a little there. FuzzyFungus posted a pretty good explanation (or at least a few assumptions that make sense a few posts down:
"One of the GPUs is, almost definitely, whatever GPU was integrated into the chipset. It'll be weak as hell; but use minimal power and be virtually free in terms of board space and bill of materials.
The second GPU will be whatever they picked for when actual performance is needed. It will add cost, space, and heat; but there is really no alternative if you want to have actual power available. Odds are, it uses more power in its lowest stable voltage/clock state than the integrated GPU does at full power."
It's the cheapest video card from nvidia. It's not even a card from the new lineup, it's an geforce 210 rebranded.
There should be a law against such false advertising.
It worked fine with the stock Vista install, doesn't work period in XP, and was unreliable in my self-installed Windows 7. I bought the machine a year ago so I'm not sure I can get a factory Win7 image for it. However I will check for updated drivers; last time I worried about it was in September and all the Win7 drivers were in beta.
Go ahead...just try to switch GPUs without a reboot and a trip to the BIOS if you run a real OS. Lenovo didn't bother checking with the community for support first with that feature.
Furries make the internet go.
The UL series (using CULV) has already been out for a while. Believe it or not, they actually do get 9-12 hours with typical usage -- just go read the Amazon reviews.