×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

World's Smallest Dual-Core ARM Cortex-A9 Module?

timothy posted about 10 months ago | from the stick-it-in-your-ear dept.

Hardware 42

DeviceGuru writes "Variscite has unveiled what it claims is the world's tiniest Cortex-A9 system-on-module, measuring 52 x 17mm. The Linux- and Android-compatible DART-4460 board is based on a 1.5GHz dual-core TI OMAP4460 SoC, is available with up to 1GB of DDR2 RAM and 8GB eMMC flash, and can run at 400MHz on just 44mA. The module provides interfaces for display (HDMI, RGB, DSI), wireless (Bluetooth, WiFi), audio, camera, USB, and more, and it consumes as little as 5mA in suspend and 44mA while running from a 3.7V battery at 400 MHz, according to Variscite. And in case you were wondering, the iconic Gumstix form-factor is 12 percent larger, at 58 x 17mm."

cancel ×
This is a preview of your comment

No Comment Title Entered

Anonymous Coward 1 minute ago

No Comment Entered

42 comments

With that processor spec, maybe (0)

Anonymous Coward | about 10 months ago | (#44001703)

But there's also this: http://www.logicpd.com/assets/products/_gallery/OMAP35x_Tor_SOM.png

Why are they still sticking with Cortex A-9 ? (3, Interesting)

Taco Cowboy (5327) | about 10 months ago | (#44002671)

With A-7 available for almost 2 years already I can not understand why vendors still producing SoC based on the A-9 chip

http://www.carbondesignsystems.com/virtual-prototype-blog/bid/171998/IP-Selection-ARM-Cortex-A9-or-Cortex-A7 [carbondesignsystems.com]

Re:Why are they still sticking with Cortex A-9 ? (1)

ChunderDownunder (709234) | about 10 months ago | (#44002863)

R&D takes time; perhaps the concept and design preceded the ability of the A7?

From the summary, it seems based on an off-the-shelf OMAP, similarly found in the Beagleboard series whose community it can lure.

Re:Why are they still sticking with Cortex A-9 ? (1)

rahvin112 (446269) | about 10 months ago | (#44003295)

COST. Duh.

The chip is designed and that design is paid for. The cost of the part is the cost to manufacture plus a 1-3% margin. That probably puts the cost of the chip at 1/2 or 1/4 of A9 chips.

Re:Why are they still sticking with Cortex A-9 ? (1)

Anonymous Coward | about 10 months ago | (#44004215)

Uh. Because Cortex-A9 gets 50% more instructions per cycle than cortex-a7.

What I was wondering (2)

drinkypoo (153816) | about 10 months ago | (#44001705)

What I was wondering is whether this URL was supposed to be slashdotted at zero comments. Oh wait, a minute later, and enough of the page has loaded for me to get a title. It's not quite slashdotted yet.

Obligatory "are they running the website on the world's slowest dual-core arm webserver" comment here

Impressive but the market does not demand it. (0)

DumbMarketingGuy (171031) | about 10 months ago | (#44001709)

I'm sure this is very innovative, and shrinking things down to this scale is an amazing technological feat, but speaking as someone who has a masters degree in Marketing Science, I have to wonder, did these guys do any market research AT ALL?

Form factors of all devices have stagnated over the last few years, there's a limit to how small your cellphone can be for example (see: Zoolander). And in these recession-hit times there are few people who can afford to shell out for the latest and greatest, just for the sake of it. So whilst I wish these guys well, I hope next time they do some market research BEFORE wasting hundreds of millions of dollars on a product nobody asked for.

Re:Impressive but the market does not demand it. (2)

Score Whore (32328) | about 10 months ago | (#44001905)

What I wonder is why it's considered impressive that they take an off the shelf SoC, design a circuit board for it and then act like they are a two year old who just made a poo.

Re:Impressive but the market does not demand it. (1)

Anonymous Coward | about 10 months ago | (#44001913)

Small size often goes hand-in-hand with low power consumption. Which translates to long battery life. And that, my marketing friend, is something you shouldn't need a masters degree to sell.

Re:Impressive but the market does not demand it. (0)

Anonymous Coward | about 10 months ago | (#44001973)

I think maybe you missed that whole wearable computing thing with smart watches, Google Glass, etc?

Re:Impressive but the market does not demand it. (1, Interesting)

bobbied (2522392) | about 10 months ago | (#44001987)

I'm sure this is very innovative, and shrinking things down to this scale is an amazing technological feat, but speaking as someone who has a masters degree in Marketing Science, I have to wonder, did these guys do any market research AT ALL?

Whoa there partner! You've not done any market research, nor do you understand the technology.

Assuming they have made the processor smaller by using better technology to lower the die size in the wafer fab.... They will get lower power consumption for the same clock rates. This translates into lower heat dissipation (which may lead to a lighter device) AND LONGER battery life.

This is the same thing Intel and AMD have been doing with their processor lines for decades, spending billions on R&D and building out wafer fabs to sell smaller CPU's that run faster and cooler.

Re:Impressive but the market does not demand it. (1)

Anonymous Coward | about 10 months ago | (#44005047)

It's not a custom chip. The same Texas Instruments OMAP SoC is used in the Kindle Fire HD. All they did was take an off-the-shelf chip and make a board for it (but that's not a trivial task given the number of signals they need to route out of that small chip.)

Server powered by one? (0)

Anonymous Coward | about 10 months ago | (#44001711)

This story just hit, and their server is already slashdotted. I guess they use one to power their web server.

Aaand it's down... (1)

pieterh (196118) | about 10 months ago | (#44001723)

Site is down before first comment. Running on a DART-4460 maybe?

Couldn't get Coralcache to work? (1)

billstewart (78916) | about 10 months ago | (#44002439)

The standard Coralcache link for the main article would be http://linuxgizmos.com.nyud.net/tiny-cortex-a9-module-runs-linux-and-android/ [nyud.net], but it chokes when I try to use it.

The Variscite.com Product Link worked just fine for me, but in case it goes away, http://www.variscite.com.nyud.net/products/system-on-module-som/cortex-a9/dart-4460-cpu-ti-omap-4-omap4460 [nyud.net] worked fine also. I couldn't find the price, though.

Raspberry Pi next Gen? (0)

Anonymous Coward | about 10 months ago | (#44001849)

Could this be the upgrade to the CPU for the Raspberry Pi that the team talked about for next year?

Re:Raspberry Pi next Gen? (0)

Anonymous Coward | about 10 months ago | (#44001895)

We don't know.

Re:Raspberry Pi next Gen? (1)

bobbied (2522392) | about 10 months ago | (#44002035)

Um... Unless they intend to change CPU types, I don't think so. I also don't think that this module would meet the price targets for the Pi.

Of course, it is certainly an option to build a Pi like device on this platform and the WiFi being built in would be nice..

Re: Raspberry Pi next Gen? (0)

Anonymous Coward | about 10 months ago | (#44002195)

Highly unlikely. The Pi was developed by Broadcom engineers and uses a Broadcom SoC. The story is about a board which uses an OMAP SoC from direct competitor Texas Instruments.

Re:Raspberry Pi next Gen? (1)

Narishma (822073) | about 10 months ago | (#44004921)

The RPi foundation has said many times that they aren't planing on doing any spec upgrades any time soon.

Why can't Microsoft do this? (-1, Offtopic)

erroneus (253617) | about 10 months ago | (#44001951)

I started into computers before "computer" meant "wintel." In fact, my first real exposure was 6809 assembly language programming under OS-9 where I had to deal with 64K limits. All of my programs needed to be extremely efficient as did the drivers for the devices I used. It was an environment of extremely limited resources and wasting them meant there was less you could do.

Linux has been moving in the direction of bloat and yet still maintains some essence of the need to be efficient and tight -- it shows in its ability to be small when needed.

Every attempt we have seen Microsoft make in trying to adapt to smaller computing devices seem to have failed for a variety of reasons.

Here's the thing though. I don't "hate" Microsoft. I just wished they'd get their heads out of their asses. Their stuff is growing obsolete. And instead of saying "okay... we have a market that may or may not be relevant in 5 - 10 years... let's keep that going while 'starting over' in new directions by devoting our considerable resources into some heavy R&D." They need not to "adapt" their old stuff and instead need to build something new in the spirit of "lessons learned from decades of experience."

At the moment, though, it seems like they haven't learned a damned thing.

Re:Why can't Microsoft do this? (0)

Anonymous Coward | about 10 months ago | (#44002359)

MS has a new 2013 version of CE that starts w/ a 25MB foot print. The primary advantage of CE in this field is the availability of drivers for a variety of devices you might connect to. Embedded computing is still largely industrial in nature.

Careful if it runs too hot... (0)

Anonymous Coward | about 10 months ago | (#44002031)

there is nothing more dangerous than small ARMs fire.

Re:Careful if it runs too hot... (0)

Anonymous Coward | about 10 months ago | (#44003055)

Just don't let any other components get too close to it; you have a right to bare ARMs.

Any word on cost? (-1)

Anonymous Coward | about 10 months ago | (#44002041)

How much do they cost?

price? (0)

Anonymous Coward | about 10 months ago | (#44002097)

has anyone asked them for a quote yet? they dont seem to show the price

Small until you try to use it (4, Insightful)

BobboBrown (541913) | about 10 months ago | (#44002201)

Boards in that sort of arrangement are common as muck and have been for a long time. Then when you want to use it, first you need a baseboard which is beyond the construction capabilities of most hackers (Variscite don't show you the back of the board which will have several specialised, very fine-pitch header connectors) and once you break out the "real-world" connectors, you've got a much larger beast.
If you want something to hack around with, go for a SOM like an RPi, or a Wandboard, or an ODroid, or a BeagleBone, or a Cubieboard, or an OLinuXino, or any of a huge number of other products.

Pretty soon, disposable computing (4, Interesting)

bobjr94 (1120555) | about 10 months ago | (#44002407)

Not to much longer these types if systems will sell for $20, then $10 until they are no longer seen as computers and just in everything. Special box of Captn' Crunch that include include cartoons or a movie on a little system like this, your kids watch them while eating breakfast, once it goes dead you throw the box away like anything else.

Re:Pretty soon, disposable computing (0)

Anonymous Coward | about 10 months ago | (#44005013)

You mean, like this? [cnet.com]

Arm Cluster (1)

EEPROMS (889169) | about 10 months ago | (#44003623)

I'm surprised no one is making an arm cluster solution based on cheap $30 modules. The modules only need a single fast IO interface that can fit into a backpane. This would allow computer hobbyists to build a cluster over time, they can start with a few modules and work their way up as they get more money.

Re:Arm Cluster (1)

drinkypoo (153816) | about 10 months ago | (#44008367)

I'm surprised no one is making an arm cluster solution based on cheap $30 modules.

No one is doing it because for a mere $300 or so you can build a multicore amd64 cluster node that will probably do more than ten $30 arm cluster nodes.

This would allow computer hobbyists to build a cluster over time, they can start with a few modules and work their way up as they get more money.

For what purpose?

Re:Arm Cluster (1)

White Flame (1074973) | about 10 months ago | (#44012245)

There have been small Raspberry Pi and even AppleTV clusters made. However, these types of small-as-possible modules aren't suited very well to hobbyist clusters:

1) These systems don't have fast interconnects. Networking is either wireless, which doesn't scale well; or through USB->Ethernet interfaces, which are slow.

2) This isn't a cheap $30 module. Gumstix are still in the $170-200 range. I'm not seeing any prices here for this one, but it's in the same hardware style.

3) You need a lot of custom hardware to mount these together, as the connectors are weird high-density custom jobs.

Why is this a question? (2)

wonkey_monkey (2592601) | about 10 months ago | (#44004561)

World's Smallest Dual-Core ARM Cortex-A9 Module?

Variscite has unveiled what it claims is the world's tiniest Cortex-A9 system-on-module

Is there any particular reason we shouldn't believe them?

Time for a standard small-size docking connector (3, Interesting)

knarf (34928) | about 10 months ago | (#44005375)

With all these small, smaller and smaller yet again modules popping up left and right for often lower and lower prices - or at least higher performance/price ratios - it gets more and more interesting to decide upon a standard small-size docking connector. This connector should offer access to a display interface, fast external storage, USB, power, etc. Everythin needed to interface it to a 'slab' containing a screen, UI actuators, camera(s), sensors, a power supply and the like.

Using these two components you can 'build' or 'upgrade' those mobile internet devices which currently get left on the shelves because their processing capacity is deemed to anaemic by modern standards. Done right it would not add much to the total price of the device but it would expand the capabilities markedly.

Something like this would not be launched by any big brand as they like the buy-use-retire cycle as it is more profitable - for them. If the standard existed and their was demand in the market that demand *would* be fulfilled from suppliers in China - just look at the amount of Arduino-compatible hardware which is produced there. Since nearly all hardware is produced in China anyway the existence of such a standard would only make it easier for the designers of those low-price mobile internet devices which are currently on offer. It would make those devices more interesting as well. Since there is as good as no brand loyalty to these suppliers they will not care about any reduced profits stemming from the replacement of buy-use-retire by buy-use-upgrade-use-retire - they'd just supply materials for the upgrade cycle.

Time for YetAnotherKickStarterProject...?

Re:Time for a standard small-size docking connecto (0)

Anonymous Coward | about 10 months ago | (#44012275)

It's called EDM standard and is used by Wandboard, Technexion, etc.. It's a fine-pitch monster with 314 pins, however, so there may be room for a smaller standard.

Re:Time for a standard small-size docking connecto (1)

rdnetto (955205) | about 10 months ago | (#44013339)

Someone already tried to do this with EOMA-68 [elinux.org]. How successful it will be remains to be seen.

Check for New Comments
Slashdot Account

Need an Account?

Forgot your password?

Don't worry, we never post anything without your permission.

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>
Sign up for Slashdot Newsletters
Create a Slashdot Account

Loading...