EmbeddedRelated.com
Forums
Memfault Beyond the Launch

32-bit Microcontroller for $1.00 -Guy Macon

Started by Guy Macon July 21, 2008
In article <181b6e7c-7f5c-495c-8c14-f11f91df9f30
@r66g2000hsg.googlegroups.com>, gnuarm@gmail.com says...
> Is this even news at this point??? Some two or three years ago I > pointed out that the ARM chips were getting cheaper and cheaper (at > that time around $2) commenting that there is getting to be no market > for the common 8 bit CPU. There were a couple who pointed out that > there are huge numbers of 8 bit devices used in the "invisible", mass > applications where even $0.01 makes a difference. Of course those apps > will always use 4 or 8 bit processors. > > But the trend has been pretty obvious for some time now. Mostly it > has been reported that the 16 bit parts are being skipped over > migrating from 8 directly to 32 bit. I have also seen predictions > that the 8 bit market will level off and start to decline although > keeping huge numbers. > > For designs that don't need to save every last penny, it really has > gotten to the point that the 8 and 16 bit parts have little need. > Maybe if power is the ultimate requirement a 32 bit part can't keep up > with an 8 bit part, but even there the newer processes that are being > used mostly with 32 bit parts are making them competitive in the > lowest power apps too.
I'm still finding niches for the MSP430 16-bitters. I've used them several times as smart peripherals for 32-bit chips and as stand-alone processors in some very low-power data loggers that store data on SD cards. They still seem to have about 1/10th the power requirement of an ATmel AT91SAM7 when clocked just fast enough to collect and store data. I expect that the M3 chips will fall somewhere between the Atmel ARMs and the MSP430 in capability and power consumption. Those chips are probably worth a look----especially after all the time and money I've spent on ARM compilers. Learning a new peripheral set will be the tough part. Mark Borgerson
> > <<SNIP>>


John Devereux wrote:

>I suspect some devious google-ranking scheme :)
I prefer openly admitting to google-ranking schemes... Plus, I always like to see who has a newsreader that changes the Subject line (collapsing multiple spaces to one space, inserting a CR/LF word wrap...) instead of folowing the de-facto standard and leaving that line unchanged. :) -- Guy Macon <http://www.GuyMacon.com/>
Jim Granville wrote:
> Jim Stewart wrote: >> Anders.Montonen@kapsi.spam.stop.fi.invalid wrote: >> >>> In comp.arch Joerg <notthisjoergsch@removethispacbell.net> wrote: >>> (Luminary Micro's website) >>> >>>> Log in just to get a datasheet? Yeah, right ... >>> >>> >>> If you just keep clicking on "I really really don't want to register, >>> just shut up already" you'll get a download link without having to >>> register. Annoying, but far preferable to those sites that actually do >>> force you to register (I'm looking at you, ARM.) >> >> >> Which is far preferable to the ones that >> make you register then promise to email >> you the datasheet and never do. (I'm looking >> at you, NXP) > > Which part of NXP does that ? > I can download any PDF just fine - PDF access at NXP is good, > it is the middle-ground, between the overview and detail, that > NXP needs to work on!
RF.
On Jul 22, 4:31 am, Joerg <notthisjoerg...@removethispacbell.net>
wrote:
> Guy Macon wrote: > > While researching something else, I ran into the > > following rather interesting opinions: > > > Consolidating the MCU market around the ARM architecture > > ("It's inevitable. ARM's Cortex-M3 processor core is going > > dominate the MCU market.") > >http://www.embedded.com/columns/guest/207001013 > > > Luminary Micro Announces 32-bit Microcontrollers for $1.00 > > -- First to Launch Products Based on the ARM Cortex M3 Processor > >http://www.design-reuse.com/news/12919/luminary-micro-32-bit-microcon... > >http://www.prnewswire.com/cgi-bin/stories.pl?ACCT=104&STORY=/www/stor... > > > Rich feature set key in MCU market: > >http://www.ciol.com/Semicon/Design-Trends/News-Reports/Rich-feature-s... > > > $1.68 per chip QTY 100: > >http://www.mouser.com/Search/ProductDetail.aspx?qs=6ARB0lp6jlUH%2fsxf... > >http://www.luminarymicro.com/products/lm3s101.html > > > $249.00 Development Kit: > >http://www.mouser.com/Search/ProductDetail.aspx?qs=6ARB0lp6jlWcf9eiWJ... > >http://www.luminarymicro.com/products/development_kits.html > > > $79.00 Daughter board: > >http://www.mouser.com/Search/ProductDetail.aspx?qs=6ARB0lp6jlX7a0oGmO... > >http://www.luminarymicro.com/products/expansion_daughterboards.html > > Log in just to get a datasheet? Yeah, right ... >
A few months ago, I had to sign NDA just to peek at the datasheets a couple of weeks in advance of public release. Unfortunately, to have good battery saving feature, we have to go with 100 pins 64K devices. The 44 pins 32K LM3S101 will kill the battery in no time. We won't be able to use LMIs until they have better battery saving modes for the low end, not the high end.

Guy Macon wrote:

> While researching something else, I ran into the > following rather interesting opinions: > > Consolidating the MCU market around the ARM architecture > ("It's inevitable. ARM's Cortex-M3 processor core is going > dominate the MCU market.") > > Luminary Micro Announces 32-bit Microcontrollers for $1.00 > -- First to Launch Products Based on the ARM Cortex M3 Processor >
I agree that the Cortex is going to be an important 32 bit part and may actually evolve into a 8051 type general purpose part it is unlikely that it will completely dominate. The quest for computing power is now a lot about power performance and this is where the Cortex does well but will not in the long term dominate as instruction sets designed for machine generated software mature. Regards, -- Walter Banks Byte Craft Limited http://www.bytecraft.com walter@bytecraft.com Canada
Guy Macon <http://www.GuyMacon.com/> writes:

> John Devereux wrote: > >>I suspect some devious google-ranking scheme :) > > I prefer openly admitting to google-ranking schemes... > > Plus, I always like to see who has a newsreader that changes > the Subject line (collapsing multiple spaces to one space, > inserting a CR/LF word wrap...) instead of folowing the > de-facto standard and leaving that line unchanged. :)
Aha... so each and every newsreader does it a bit differently, creating new thread titles for you. Sounds pretty devious to me :) -- John Devereux
"Jim Granville" <no.spam@designtools.maps.co.nz> wrote in message news:4884e753$1@clear.net.nz...

> Conclusion: Yes the M3 is significant, but it is a > _very_ long way from dominating the MCU market, it is not even > close to dominating the 32bit MCU market.
"M3 is significant" - wow, that's quite a change from what you said a few years ago :-) Of course just about all of the MCU manufacturers have licensed the M3 since then as expected, including Atmel... Wilco
Wilco Dijkstra wrote:
> "Jim Granville" <no.spam@designtools.maps.co.nz> wrote in message news:4884e753$1@clear.net.nz... > > >>Conclusion: Yes the M3 is significant, but it is a >>_very_ long way from dominating the MCU market, it is not even >>close to dominating the 32bit MCU market. > > > "M3 is significant" - wow, that's quite a change from what you said a > few years ago :-)
Don't get too excited ;) I would also call the Coldfire, SuperH, MIPS, PowerPCB, Numerous DSPs (and even the AVR32) also significant, but do NOT confuse 'being on the radar', with dominating a sector. This recent news item, shows how much more power matters, than cores, these days.... "Texas Instruments (TI) is overhauling nearly all of its DSP lines with 15 new chips that slash power by up to 20X. In addition to cutting power, the new chips represent a number of firsts for TI, including the first chip to combine an ARM core with a floating-point DSP. As part of the rollout, TI is also re-uniting its fixed-point and floating-point C6000 families&#4294967295;two families that parted ways years ago." http://www.eeproductcenter.com/dsp/brief/showArticle.jhtml;jsessionid=0KQUA1SNGVHKEQSNDLPSKH0CJUNN2JVN?articleID=209400505
> > Of course just about all of the MCU manufacturers have licensed the > M3 since then as expected, including Atmel...
True, and NONE are yet offeringn pin-compatible second sources, but Freescale and ST ARE now offering PowerPC second sourcing, for their demanding Automotive customers. -jg
Jim Granville <no.spam@designtools.maps.co.nz> wrote:

>Wilco Dijkstra wrote: >> "Jim Granville" <no.spam@designtools.maps.co.nz> wrote in message news:4884e753$1@clear.net.nz... >> >> >>>Conclusion: Yes the M3 is significant, but it is a >>>_very_ long way from dominating the MCU market, it is not even >>>close to dominating the 32bit MCU market. >> >> >> "M3 is significant" - wow, that's quite a change from what you said a >> few years ago :-) > >Don't get too excited ;) > >I would also call the Coldfire, SuperH, MIPS, PowerPCB, Numerous DSPs >(and even the AVR32) also significant, but do NOT confuse 'being on the >radar', with dominating a sector. > >This recent news item, shows how much more power matters, than cores, >these days.... > >"Texas Instruments (TI) is overhauling nearly all of its DSP lines with >15 new chips that slash power by up to 20X. In addition to cutting >power, the new chips represent a number of firsts for TI, including the >first chip to combine an ARM core with a floating-point DSP. As part of >the rollout, TI is also re-uniting its fixed-point and floating-point >C6000 families&#4294967295;two families that parted ways years ago." > >http://www.eeproductcenter.com/dsp/brief/showArticle.jhtml;jsessionid=0KQUA1SNGVHKEQSNDLPSKH0CJUNN2JVN?articleID=209400505
This shows power matters to TI. Like someone already mentioned in this thread: TI's MSP430 series is as low power as you can get in todays microcontroller market. With all the competition moving to ARM, TI (and their competitors too) need to add something unique to their products which stands out in the crowd. -- Programmeren in Almere? E-mail naar nico@nctdevpuntnl (punt=.)


John Devereux wrote:
> >Guy Macon <http://www.GuyMacon.com/> writes: > >> John Devereux wrote: >> >>>I suspect some devious google-ranking scheme :) >> >> I prefer openly admitting to google-ranking schemes... >> >> Plus, I always like to see who has a newsreader that changes >> the Subject line (collapsing multiple spaces to one space, >> inserting a CR/LF word wrap...) instead of folowing the >> de-facto standard and leaving that line unchanged. :) > >Aha... so each and every newsreader does it a bit differently, >creating new thread titles for you. Sounds pretty devious to me :)
Why would you think that new subject lines are some sort of advantage? NUsenet newsreaders and Google Groups both thread by references, not subject lines. And doesn't the word "devious" imply that someone is harmed or at least slightly inconvenienced? -- Guy Macon <http://www.GuyMacon.com/> Guy Macon <http://www.GuyMacon.com/> Guy Macon <http://www.GuyMacon.com/> Guy Macon <http://www.GuyMacon.com/> Guy Macon <http://www.GuyMacon.com/> Guy Macon <http://www.GuyMacon.com/> Guy Macon <http://www.GuyMacon.com/> Guy Macon <http://www.GuyMacon.com/>

Memfault Beyond the Launch