Low Power Cortex M?

With some effort you can increase the power of these ones:

www.ehow.com/how_4791464_tone-flabby-arms.html

Does anybody know of one?<p>

You may want to look at Atmels Cortex M3 products.

Personally, I work with one of their ARM7 chips here, and as far as power consumption goes it beats most competitors Cortex M3s in the same CPU power/memory range hands down. If they can do a similar job with their (upcoming and already released) Cortex M3 chips, they should be pretty much as good as you can get.

All of this assumes that you're willing to deal with Atmels "quirks", though.

Really - at last?!

See: www.8052.com/.../165761

Aren't they supposed to be introducing some other Cortex-Mx which is supposed to be super low power...?

Most (all?) Cortex-M3 are quite nice when it comes to power/MIPS since the manufacturer buys the processor core as a macro cell and the macro cell for a given technology should have a quite fixed power consumption.

Err ... I don't think this generalization holds.

We were looking at Cortex-M3s before Atmel started offering theirs, and there was a factor of _10_ between the Cortex M3 of one of their competitors, and an ARM7 from Atmel, at a similar MIPS/memory/peripheral point. (The power dissipation for the Cortex M3 was about 250 mW according to the datasheet, for the ARM7 30 mW.).

Some manufacturers are better at peripherals and processes than others. Don't assume that just because it's a Cortex M3 chip, it will be a power-saving wonder. Read the datasheet first.

Aren't they supposed to be introducing some other Cortex-Mx which is supposed to be super low power...?

Cortex M0s, I believe. I read the announcement, but haven't heard anything since.

There are some power hungry devices in the system, and I am trying to address each aspect of the project. I've worked with the vendors of these power hungry devices, and, well..., physics still works.

I shall look at any claimed-to-be low-power Cortex processor that 'normal people' post.

Thanks,

--Cpt. Vince Foster
2nd Cannon Place
Fort Marcy Park, VA

Andy,

Aren't they supposed to be introducing some other Cortex-Mx which is supposed to be super low power...?

As far as I know, that is due to happen beginning 2010.

Oh and I did mean mAh and not mA/H (I had a sardine moment there)

Here is one you Mr. Black:
www.youtube.com/watch

--Cpt. Vince Foster
2nd Cannon Place
Fort Marcy Park, VA

Note that both the Cortex-M3 and the ARM7 chips are available in different geometries, and there are large differences in power consumption.

But the consumption of the processor core should not vary too much from different offerings using the same manufacturing process which was my claim from my first post.

And I did mention explicitly "But in the end, the peripherials may be the big difference, since they can stand for a significant percentage of the total consumption and are what differs between the manufacturers."

One manufacturer may have a tiny timer. One may have a timer with a ton of configuration bits.

The (non Cortex-M3) LPC23xx chips have a broken implementation of the RTC + battery-backed RAM, making the processor gobble a lot of current and making NXP officially drop the support for one power-save mode instead of adding an errata. For me, that little design misfeature represents about 50% of the power consumption at the lowest power-save mode for a design with a LPC2366. In the end, the 32kHz slow-clock mode wasn't meaningful to use because of power leakss inside the chip when the internal 1.8V DC/DC was turned off. And the power leaks required an external RTC because we couldn't afford the required battery/supercap size for using the internal RTC.

When reaching these low powers, it will be enough with quite small engineering errors to get the modules around the core to consume more than the core.

The problem is that chips mentioned to be released somewhere in 2010 may have still unknown errors in them, affecting their true power consumption. When will a buyer know that a "super" chip is really as good as the initial projection.

NXP recently introduced a broad line of lowest power Cortex M-series microcontrollers based on the industry leading ARM architecture. These new LPC1300 and LPC1100 microcontroller products are suitable for embedded consumer, industrial, and portable medical applications.
you can register for a webinar at www.techonline.com/.../220600840 the webinar is at

Date and Time

Greenwich Mean Time
Tue., Nov 03, 2009 16:00

Eastern Standard Time
Tue., Nov 03, 2009 11:00 AM

Erik

about Energy Micro:

www.newelectronics.co.uk/.../32bit-micros-consumes-25-the-power-of-rivals.aspx

Yes, but they are making the claim for an unreleased processor family, when comparing them with released versions from competitors.

We regularly gets visits from chip manufacturers, informing us about upcoming chips. But they never try to compare their existing chips with competing alternatives. They always compares the chips we will be able to get prototypes of in 3-18 months and introduce in real products even further into the future with the existing offerings from competitors. And the information they supply is of course under NDA, so we can't merge the information and present comparisons to others.

It would be interesting to know the comparison for this specific family compared to competitors processors using the same manufacturing process, and if any other manufacturer will have processors using this process available within a similar time frame.

I was just thinking the same thing!

So, what would be a good application that could easily be run on a number of Cortex-M3s to get some kind of a "real-life" comarison...?

I first heard of Energy Micro's Cortex-M3 over a year ago.

I was assuming this meant that they're actually releasing parts now but, reading more carefully, it doesn't actually say that...

:-(