Intel Working to Double Battery Life

[Jonathon Watkins]

<div class='os_post_top_link'><a href='http://www.theinquirer.net/?article=16504' target='_blank'>http://www.theinquirer.net/?article=16504</a><br /><br /></div>Now this looks promising. The Inquirer have a story up about Zinc Matrix Power signing a collaborative deal with Intel, to work on new battery technology that could allow laptop batteries to run for ten hours. Hopefully PDA batteries etc. will be the next thing they turn their attention to.<br /><br /><i>"Zinc Matrix Power is developing a rechargeable, polymer-based alkaline battery which it claims has shown in lab tests approximately two times the energy-to-volume of today's standard lithium-based batteries. The polymer based technology uses a water-based electrolyte which, when combined with the plastic case, will minimise the risk of fire." </i><br /><br /><img src="http://www.pocketpcthoughts.com/images/web/2003/ZMPbattery.jpg" /><br /><br />The press release from <a href="http://www.zmp.com/news.html">Zinc Matrix Power</a> mentions that the company plans to introduce this technology commercially in 2006, so don't cancel your order for that second PPC battery just yet. :? We need *some* sort of decent increase in <a href="http://www.tropian.com/news/rcr123.htm">battery capacity</a> soon, so hopefully this partnership will prove fruitful.

[ricksfiona]

I would like to use my Pocket PC w/WiFi & BT for most of the day without getting 'low power' messages.

[maximus]

At least this is saver than the fuel cell alternative. Something that will not explode near one's face.

[Jason Dunn]

Very cool to see some improvements in this area! :-) Doubling the battery life would be an excellent improvement.

[Jonathon Watkins]

Very cool to see some improvements in this area! :-) Doubling the battery life would be an excellent improvement.

Well it would be a start. If they really can double the battery life in two years, then that will only just be outside of the Moore's (http://www.intel.com/research/silicon/mooreslaw.htm) law, range i.e. a doubling of processor speed every 18 months.

It's about time this happened!

[huangzhinong]

Very cool to see some improvements in this area! :-) Doubling the battery life would be an excellent improvement.

Well it would be a start. If they really can double the battery life in two years, then that will only just be outside of the Moore's (http://www.intel.com/research/silicon/mooreslaw.htm) law, range i.e. a doubling of processor speed every 18 months.

It's about time this happened!

It won't happen. It's about the difference between physics and chemistry. Moore's law only apply to physics engineering. Battary belongs to chemical engineering.

[kosmicki]

Now here is the problem, battery life doubles, how much do you wanna bet they think 'hey! we can make the batteries half the size now!'

Laptop computer life is already shi**y, I want to see one of those 1lb notebooks, with a 2lb battery. 24 hours AT LEAST per charge.

Remember, just IMNSHO.

[Blue Zero]

One word:

Ihatetorechargeeverytimeigohome !
Yeah, I know, such big word :roll:

[Jonathan1]

Very cool. This is prob going to be a good stopgap between current LI-Ion batts and fuel cells which are still a ways off from being practical. At this point anything that increases the batt life on my laptop is a good thing. Preferably I would like to get at minimum 24 hours off of a charge on a laptop but we are so far off from such a beast, if ever, that it remains in the realm of fantasy. :?

[jonathanchoo]

Once battery life doubles, processor speed would also double. No battery life in the future would equal battery life as today, just with more power hungry processors.

Glad for this news though cuz I am one of those fuel cells sceptics.

[unxmully]

... I want to take my T3 off it's cradle today, use it for three hours per day and get the low battery warning just after Christmas.

What's that, about 600 hours of usage. Only needs to go up by a factor of 150 or so. Sounds OK to me :D

Oh, and I want that with wireless so I can synch with my home workstation without needing to be physically connected to it.

Mind you if you could double the battery capacity and halve the power consumption of the components, 16 hours would be a big step forward.

Question, I'm sure I read a while back that the newer battery technologies could be "shaped" more than the old. If that is the case, does anyone know why the whole chassis of a PDA or laptop isn't just one big battery :?:

[Jonathon Watkins]

Very cool to see some improvements in this area! :-) Doubling the battery life would be an excellent improvement.

Well it would be a start. If they really can double the battery life in two years, then that will only just be outside of the Moore's (http://www.intel.com/research/silicon/mooreslaw.htm) law, range i.e. a doubling of processor speed every 18 months.

It's about time this happened!

It won't happen. It's about the difference between physics and chemistry. Moore's law only apply to physics engineering. Battary belongs to chemical engineering.

I am fully aware of this and was pointing out that we need similar battery life increases to keep up with the increases in processor speeds etc.

[Felix Torres]

Laptop computer life is already shi**y, I want to see one of those 1lb notebooks, with a 2lb battery. 24 hours AT LEAST per charge.

Remember, just IMNSHO.

This you can have today.
If you can afford the 1 lb notebook (cheap they ain't) then you can afford the $400 for an Electrovaya powerpad that gives you 16-24 hours of *extra* battery life.
Very thin and nicely implemented.
A friend of mine got one at work...

Electrovaya's TabletPCs all come with this kind of battery so, again, if you can afford it, you *can* get a fast computer with an all-day battery.
There's a review posted around here somewhere...

[hotweiss]

This is a nice improvement in battery technology, but lithium sulphur batteries promise a lot more power.

http://www.theregister.co.uk/2004/05/12/lithium_sulfur
http://www.mobilemag.com/content/100/102/C2838/

Although it looks like the zinc batteries will make it to the market first.

[omikron.sk]

Does 10AH on the picture really mean Amper-hours ? 10 AH = 10 000 mAh, right?
:jawdrop:

[torok]

Well, I'm impressed. I don't think I've ever seen battery energy density double. If they can pull it off, of course.

[torok]

Does 10AH on the picture really mean Amper-hours ? 10 AH = 10 000 mAh, right?
:jawdrop:

Maybe that cell is the size of house cat.

[Talon]

Or you could just use a lower power PDA :-)

Where does the power go?
The CPU, the memory, the display and wireless stuff.

Intel ARM based parts have always used more power that other manufacturers ARM parts.
Yes an Xscale is clocked faster than other ARM9 cores so if you ran it at the same speed the power would be around the same. But the Xscale is not a true ARM, it gets a lower throughput per clock than any other ARM9. Plus it's the idle power that matters anyway, the CPU is idle most of the time on a PDA. Things like the OMAP are an order of magnitude lower power when idle than the xscale.

The memory, well if you use the new lower power mobile SDRAM then the power is a lot lower but it costs a lot more. The last PDA I took apart was still using standard SDRAM but that was a while ago.
Plus if you halve the RAM you halve the power it takes. Not a big deal when the system is busy but a big factor on the standby life of the battery. With demand pageing now in PPC you can use NAND flash and only use up around 2MB more ram than if you were running from NOR flash. Expect to see more people using that.

As for displays, the backlight is the biggest power user on the whole system. But people want bright backlights. Not much chance of that changing much.
One thing that will change, the higher the resolution the display the darker it will be. Which means higher resolution displays need more power in the backlight to appear the same brightness. There is no getting away from that, it will be canceled out a bit by better display technology but not completely.

Finally wireless. Idle it doesn't take much power. As soon as you start using it you can expect to see the battery percentage fall. Bluetooth is bad, 802.11x is worse. There isn't a lot you can do about it and while the power used will go down with new generations it is never going to be low power.

[Jonathon Watkins]

Does 10AH on the picture really mean Amper-hours ? 10 AH = 10 000 mAh, right?

Hmmm, that looks too good to be true..........

[Talon]

Does 10AH on the picture really mean Amper-hours ? 10 AH = 10 000 mAh, right?

Hmmm, that looks too good to be true..........

It does mean that but it is too good to be true.
That cell is 1.6V, most Li-Ion cells have a nominal voltage of 3.7V.
So a 10Ah 1.6V cell contains 16Wh of power, the same as a 4.3Ah (or 4324mAh) Li-Ion cell.

[Jonathon Watkins]

Plus if you halve the RAM you halve the power it takes. Not a big deal when the system is busy but a big factor on the standby life of the battery. With demand pageing now in PPC you can use NAND flash and only use up around 2MB more ram than if you were running from NOR flash. Expect to see more people using that.


Could you expand on that Talon? Why does NAND memory take up more space than NOR?

[omikron.sk]

Does 10AH on the picture really mean Amper-hours ? 10 AH = 10 000 mAh, right?

Hmmm, that looks too good to be true..........

It does mean that but it is too good to be true.
That cell is 1.6V, most Li-Ion cells have a nominal voltage of 3.7V.
So a 10Ah 1.6V cell contains 16Wh of power, the same as a 4.3Ah (or 4324mAh) Li-Ion cell.
4300mAh - that's more than enogh for me. I just hope it's not too big (physically). It's hard to tell its dimensions from the pic. If it would be the same size as my iPaq2210 battery, I'm impressed.

[Talon]

Plus if you halve the RAM you halve the power it takes. Not a big deal when the system is busy but a big factor on the standby life of the battery. With demand pageing now in PPC you can use NAND flash and only use up around 2MB more ram than if you were running from NOR flash. Expect to see more people using that.


Could you expand on that Talon? Why does NAND memory take up more space than NOR?

It is to do with the way the internals of the memory work. NAND flash is a serial based system. You read or write a block (normally 512 or 2k Bytes) at a time. You can't just jump to the exact point. Think of it like a CD with no fast forward. You can jump to the correct track but you have to play the whole track in order to listen to the end of it.
NOR flash is random access, you can access any location in memory just as easily and quickly as any other.

If you want to run a program you need to have random access to the memory that holds it. In other words you can run a program that is stored in NOR flash, you can't run a program that is stored in NAND flash.

So why use NAND? It's a lot cheaper. Plus SDRAM is faster than NOR flash so once you program has been coppied into RAM it will run faster.

If windows is stored in NAND then traditionally you have to copy the whole thing into RAM before you can run it. Windows mobile is around 30MBytes, so your 64MByte PDA only has 34MBytes free for the user to make use of.
This doesn't make people happy. Unfortunatly until WM2003 that was the only way of running with NAND flash.

If windows is in NOR then it can run from flash. Windows will use around 2MBytes for storing odds and ends (filesystem lookup tables etc.) but that leaves the bulk of your RAM free. The downside is you system is slower and more expensive.

Demand pageing is a bit like virtual memory on a desktop PC. The OS copies the bit of the windows OS that it is using into RAM and leaves the rest in the NAND flash. If it needs a new bit of the OS then it copies that section as a block from the NAND to RAM and throws away something that it hasn't used for a while. That is exactly what NAND flash is good for, reading a block at a time. A PPC system using demand paging will use the 2MB of RAM that windows needs anyway plus another 2M or so for the bit of the OS that it is using.
The performance hit of having to keep stopping to copy new blocks from NAND into RAM is balanced out by the faster speed of the OS when it is running from RAM so the overall performance is roughtly the same.

Does that answer your question?

[Johnathan]

All we can do just waiting :?