Intel Releases Several Projects to Help Save Power 83
GeekyBodhi writes "LessWatts.org is Intel's new website that hosts several power saving tools. As Linux.com reports, it also shares tips and tricks to help optimize power consumption on hardware from portable devices running on batteries to large data centers. 'LessWatts.org is not about marketing, trying to sell you something or comparing one vendor to another. LessWatts.org is about how you can save real watts, however you use Linux on your computer or computers.' As reported on Slashdot earlier, this isn't the first time Intel has tried to help Linux users cut their power bills. In May, the company launched the PowerTOP program that monitors individual processes to keep track of power consumption. The project comes at a time when more vendors are pre-installing Linux on handhelds and laptops." Linux.com and Slashdot are both owned by SourceForge.
in other power news (Score:1, Troll)
Re: (Score:2)
I'd say the desktop platform is in a great position. speed has gone way up and cost is way down, and we have more options then ever in the history of computing. I propose you are just a whinger who will never be happy no matter what.
Re:less? (Score:4, Funny)
Or, they could have gone the nostalgia route and used "WattsYouTalkinAboutWillis.com"
Re: (Score:1)
Re: (Score:1)
Unless you are the guy who accidentally sticks his finger in a live light socket.
This happened to me once when the chain had broken and I used the twisting of the light bulb action to turn it on. Well early one morning I got the shock of my life. I was an instant human voltmeter.
Re: (Score:1)
Re: (Score:1)
Fewer Watts (Score:1)
Fewer Watts (Score:2, Informative)
Re: (Score:3, Funny)
Re: (Score:2)
Re: (Score:2)
That's like saying Voltage instead of potential difference, which is a generally accepted term, but unfortunately can lead to use of the term 'Ampage' instead of current which generally makes one sound ignorant. However, I am sure Intel (not to mention Linux fans) would have a problem showcasing a website proudly proclaiming that Linux and Intel means "Less Power".
eh? (Score:1)
powertop... (Score:5, Informative)
Regardless, Powertop actually is a useful tool, and it's helped me get much better battery life out of my T61 running Ubuntu.
Re: (Score:3, Informative)
Interesting question... (Score:4, Interesting)
Re: (Score:2)
Re:Interesting question... (Score:4, Insightful)
power supply calculator (Score:2)
eXtreme Power Supply Calculator Lite v2.5 [outervision.com]
Another thing to consider when buying a power supply is to be sure to get one that is 80+ certified, which means it has more than 80% energy efficiency, ie wastes 20% or less electric energy as heat, thus reducing electricity waste and bills.
The 80 PLU [80plus.org]
Re:Interesting question... (Score:5, Informative)
Basically Windows consumed significantly less power than Linux (about 30-40% less). From the lesswatts site it is good to see that more power saving will be in the Linux kernel in the 2.6.24/25 timeframe (eg SATA power saving) as well as user space (eg not polling SATA cdroms every 2 seconds looking for media changes). I guess we'll be seeing those updates in Hardy Heron.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Gutsy features a number of interesting problems. It seems the GNOME brightness feature doesn't dim the backlight, but rather fades the screen output closer to black. This energy saving method might work on CRTs. It works less well on LCDs. I noticed no immediate savings increasing or decreasing the brightness output, in direct
Re: (Score:2)
I was relying on the battery reporting via ACPI which is shown in the Gnome Power Manager under Linux and the Lenovo Power Manager under Windows. I believe both read the same underlying data from the battery. Measuring externally on a laptop (I do have the equipment) is harder since you'd have to plug a charger in at which point the software behaves differently.
Re: (Score:2)
And I suppose some sort of Linux Anti-Virus software would need to be procured and added to the Linux test systems to make it fair...
Not necessarily. All that's needed is equal resistance to viruses. If Linux can do that more efficiently than Windows, then it is unfair to discount that advantage by saddling it with an unneeded and unproductive overhead.
for desktop, power supply is biggest waste (Score:4, Informative)
Re:for desktop, power supply is biggest waste (Score:5, Informative)
Re: (Score:2, Interesting)
Re: (Score:2)
Re:for desktop, power supply is biggest waste (Score:4, Informative)
You'll find a variety of statements about this running around on the web. It's true that a switching power supply is a wildly nonlinear device (which is why, for example, you can't test it without a load). But there's no simple, reliable rule for when it will be most efficient. You'll hear some people saying a PS is most efficient when used near 100% of its rated capacity, and others who say 50%. Some usenet folks did systematic measurements (you can probably turn up the thread on google groups, I don't have it handy), and they found that basically none of these statements was generally true. There was no clear relationship between % load and efficiency that held true across a variety of power supplies. When people say that it's most efficient when almost 100% loaded (I know, you didn't say that, but other people do), there are two things to keep in mind: (a) it's not necessarily true, and (b) even if it was, you wouldn't want to design your system so that it used almost 100% of the PS's rated power. What typically happens if your PS is on the ragged edge of having enough power for your system is that you get random failures to boot which are hard to reproduce. The reason is that booting tends to require a lot of power (spinning up the drives, running the CPU full-out), and may go over what the PS can supply.
The main thing is to get an 80PLUS power supply. Not only can you be sure it will be efficient, but it won't contain any lead.
What Intel is doing seems laudable, but it seems like it's likely to be a very time-consuming way for an individual to cut one watt off of their system's power consumption. You want to pick the low-hanging fruit first. Get a power consumption meter such as a kill-a-watt, and take some measurements. I used to have a pair of speakers that drew 12 W, even when the computer was off, and I didn't know it. If you've got a CRT, replace it with an LCD. Another big issue is doing your word processing on a machine with a video card that gets hot enough to fry an egg on.
The other big issue on Linux is poor support for power management. There are some success stories, such as the fact that the kernel automagically supports AMD cool'n'quiet, but in general, there are serious problems with sleep and hibernation on linux, and the reason is that manufacturers of peripherals refuse to publicly release the documentation for all the registers that need to be saved in order to restore their states when the machine wakes back up. Personally, I've never had any luck with sleep or hibernation on any machine I've ever installed linux on.
move heat away of power supply (Score:3, Interesting)
Instead use a heat sink that have a vertical cooler like Zalman CNPS9500
what? (Score:2)
Real watts? You mean as opposed to imaginary watts?
Re:what? (Power Factor) (Score:2)
Shove an AC current into an RC circuit tht is resonant. Then you have
lots of Imaginary watts!
Wrong 'imaginary' (Score:1)
What about... Windows users :( ? (Score:3, Interesting)
Please?
Re: (Score:2)
just pull the plug, windoze luser
I do pull the plug from time to time. But it's a laptop, it just keeps running Windows on batteries.
As for you, I doubt you have laptop, let alone be able to install Linux on a laptop without hurting yourself fatally, seeing you have difficulty spelling "Windows".
trust them? (Score:1)
I am also wondering... isn't the tickless idle a standard thing on the newest kernels? so then aside from cutting dow
Re: (Score:3, Informative)
Re: (Score:1)
Re: (Score:2)
Re: (Score:2)
Grammar (Score:4, Informative)
Of the 291 million transistors (Score:5, Interesting)
on modern Intel CPUs, the overwhelming majority are used to compensate for the fact that the CPU is ridiculously faster than the memory to which it interfaces. IOW, the cache. These CPUs consume between 80 and 120 watts [intel.com] of power. The reduced power versions use only 50.
By way of comparison, the 1 GHz AMD Geode runs on about 1 watt of power, and ARM processors can get by for even less.
By way of further comparison, a register to register transfer can be completed in 1/2 clock cycle. Contrast this with a read-modify-write memory cycle where a word is fetched from one memory location, modified, and written to a distant location, which will take 4 memory cycles (which typically runs at 1/2 or less of the clock speed of the CPU).
The power consumption problem is due more to the fact that compensating for this difference in speed requires a large SRAM cache on the die. And even then, it's not perfect - if you do things which routinely involve cache misses (such as video encoding, etc...) the CPU is stuck operating at the effective speed of the memory bus.
The key to reducing CPU power consumption is to use lower-latency memories, which require smaller on-die caches for a given performance level. We could double the throughput of DDR SDRAM by simply demultiplexing the address and data busses, similar to the way SRAM functions. There's no requirement in the underlying storage structure of DRAM to require separate row and column addresses; it's just a historical artifact. Originally, before DRAM and SDRAM became popular, computers were built with SRAM because its lower latency allowed even slow CPUs to work efficiently. But DRAM promised lower cost (via fewer bus lines) and lower power consumption (bits stored in charged capacitors, rather than cross-coupled transistors), at the price of latency, and the rest, unfortunately, is history.
Re: (Score:2)
The key to reducing CPU power consumption is to use lower-latency memories, which require smaller on-die caches for a given performance level. We could double the throughput of DDR SDRAM by simply demultiplexing the address and data busses, similar to the way SRAM functions. There's no requirement in the underlying storage structure of DRAM to require separate row and column addresses; it's just a historical artifact. Originally, before DRAM and SDRAM became popular, computers were built with SRAM because its lower latency allowed even slow CPUs to work efficiently. But DRAM promised lower cost (via fewer bus lines) and lower power consumption (bits stored in charged capacitors, rather than cross-coupled transistors), at the price of latency, and the rest, unfortunately, is history.
Well, it would be great to have a gig of SRAM as your main memory, but that wouldn't be too good for overall power consumption, not to mention price. Besides, modern processors can turn off the power from parts of the cache, in order to save more power.
On the other hand, I agree it's been a dumb decision to favour throughput at the expense of latency. DDR2 and DDR3 are unfortunately good examples of this. I guess we're back to the Pentium 4 -style marketing where megahertz is king, no matter what the
Re: (Score:2)
>By way of comparison, the 1 GHz AMD Geode runs on about 1 watt of power, and ARM processors can get by for even less.
And? You forgot the other part of the comparison: the performance part, otherwise an unplugged CPU used 0W (and no a clockspeed is not a performance indicator).
>We could double the throughput of DDR SDRAM by simply demultiplexing the address and data busses
Which would have a significant cos
Re: (Score:2)
And? You forgot the other part of the comparison: the performance part, otherwise an unplugged CPU used 0W (and no a clockspeed is not a performance indicator).
Naturally, a slower CPU will tend to have lower performance unless you're running a VERY atypical application that gains nothing from cache in the first place.
However, many many computers (especially in business) are used for nothing but some basic word processing, email, and web browsing. The most demanding computational task they perform is m
Re: (Score:1)
By way of comparison, the 1 GHz AMD Geode runs on about 1 watt of power, and ARM processors can get by for even less.
Great, but where do I get a AMD Geode and a motherboard for it? I'd love to have a silent PC (actually, server: I made the mistake of believing AMD's Cool 'n Quiet hype and it really isn't all that quiet... but enough ranting). I know about Soekris Engineering [soekris.com]. The fastest they have is a 600MHz Geode. Probably more than enough for my needs, but they aren't exactly cheap. (I know why,
Re: (Score:2)
EBay has several sellers with Geode+Mobo combos to be found.
The mobos, though can be any Socket A that can supply the extra-low voltages needed.
Re: (Score:2)
That's a really terrible comparison...
First off, your power figures on AMD Geodes are entirely wrong. The only 1GHz Geode I was able to find is the "AMD Geode(TM) NX 1500@6W," and beyond 6 being a lot bigger than 1, is much lower than the actual wattage rating anyhow, it's just like that 1500 PR number. For comparison the AMD Geode LX 800@0.9W processor is actually 500MHz, and operates at a
Linux-Ecology-HOWTO: More than Power Saving (Score:2)
This site is NOT an Intel project (Score:1)
The site is subscriber only, so here is one of the comments making it clear that this is not an Intel only site:
"Posted Sep 21, 2007 4:04 UTC (Fri) by subscriber arjan (#36785) [Link]
The site is by absolutely no means limited to Intel.
If you have suggestions etc for other systems by all means contribute and I'll add them personally to the website [*]
[*]
Re: (Score:2)
OSX (Score:1)
Re: (Score:2)
Re: (Score:1)
Cut FB-DIMMS (Score:2)
Random glitches (Score:1)
On the subject of power savings, I tried enabling the on-demand frequency governor when I last upgraded my system. Turned out that enabling AMD's PowerNow feature in the BIOS, which enabled the on-demand governor to work, caused my VMware virtual machine to randomly hang and/or stop its clock [vmware.com]. So I had to turn it off.
I'm willing to try some of the other power-saving suggestions, but only as long as they don't cause problems in any of the applications I use.
Tip for AMD users (Score:2)
You can reduce the power used on an AMD CPU by having it ramp the clock down when the OS is in an idle loop.
To find out the current setting, you can use lspci.
lspci -xxxs 00:18.3 | awk '($1 == "80:") {print $9}'
This number will tell you what divisor the CPU is running at.. 0 is no clock ramping, and 61 is the maximum value.
to set this:
setpci -s 00:18.3 87.b=61
If you have a second CPU socket (single/dual/quad core doesn't matter) you will have to also adjust 00:19.3. This may caus
Re: (Score:2)
correction (Score:2)