Print 24 comment(s) - last by bh192012.. on Mar 14 at 12:59 PM

Newly created carbon nanotube phase change memory sounds attractive, but it is much more expensive than traditional semiconductor-based phase-change memory. While it offers large energy efficiency gains, memory is typically a minimal part of power consumption on modern mobile devices.  (Source: Eric Pop, University of Illinois)
The root problem remains the power draw of the display

Researchers at the Beckman Institute for Advanced Science and Technology at Illinois have improved upon a novel, faster form of memory to make it use power more efficiently.

The new memory falls under a class of devices called power phase-change materials (PCM), which store data as resistance, rather than charge.  Previous attempts at PCM memory had been fast, but less power efficient due to large contacts.

The Beckman team, led by computer engineering professor Eric Pop [profile], explored the prospects of much smaller phase-change memory using nanoscale contacts and a gapped carbon nanotube (a tube with a missing segment in the middle -- essentially two extremely almost touching nanotubes).

The results were that power consumption was cut by a factor of 100 from current generation phase change memory (the team did not put this in context with traditional memory, rather than to say it was much more efficient).

Professor Pop optimistically describes, "We're not just talking about lightening our pockets or purses. This is also important for anything that has to operate on a battery, such as satellites, telecommunications equipment in remote locations, or any number of scientific and military applications."

While semiconductor-based phase change memory is arguably commercially viable due to its speed, this carbon nanotube phase change memory may be one case where the actual device fails to live up to the hype.

How big an impact will this really have on mobile battery life, if it makes it to the market?  It’s likely that it won’t improve battery life that much in the long run.  Currently, the biggest power wasters on your mobile phone are your screen/GPU, your CPU, and the wireless modem [source]. 

Furthermore, there's no established process to create chips with gapped carbon nanotubes affordably in a fab.  Creating carbon nanotube PCM chips would be extremely expensive with today's technology.

Ultimately, this could bump memory access speeds slightly, and bump the battery life 2 or 3 percent in the best-case scenario.  While it's true every bit counts, the cost may more that negate the minimal gains.  

In other words you're spending a tremendous premium to go from semiconductor to carbon nanotubes and the only reward for that switch (reduced power consumption) is extremely minimal.

And the real question here is "why memory?"  There are so many more attractive power efficiency targets -- like transistors (for CPUs), the battery itself, wireless modems, and displays.   

When and if carbon nanotube production becomes affordable, this idea may become marginally useful.  But for now chalk this one to a lot of bark but not much byte.

The team's study is published [abstract] in the journal Science.


Comments     Threshold

This article is over a month old, voting and posting comments is disabled

By GTVic on 3/11/2011 10:45:51 PM , Rating: 3
A non-volatile memory product has tremendous potential if some of the difficulties can be worked out.

The comment that the power savings are minimal and therefore the technology is pointless is ridiculous. The comment that the speed increase is minimal is baseless. And finally the comment that there are no fabrication techniques in place is common to all significantly different/new technologies and is hardly a reason to give up unless you are a couch potato.

Hard drives have issues that most people know about (speed, noise, physical vulnerability and size).

The current generation of flash is actually less reliable than the previous requiring more and more error correcting strategies. Plus the huge disadvantage of limited write cycles and block erase requirements.

PRAM memory has significant potential. Don't believe everything you read, in particular this article.

By bug77 on 3/12/2011 10:13:27 AM , Rating: 5
Read the article carefully. There is already some semiconductor-based phase change memory with similar benefits.
This one is just more efficient, but also more expensive (because we have little experience in working with nanotubes). All Jason is saying (and I agree) is that this doesn't seem to pay off.

By GTVic on 3/12/2011 6:35:12 PM , Rating: 3
When quoting someone who is promoting the technology, Jason Mick says "optimistically describes ...". Nice put down! Since he files this as a blog I guess we can throw journalistic integrity down the drain.

His other arguments are just as baseless. He says power savings are worthless because other components in portable devices using other technology are more wasteful. Huh? Is that how we are to judge whether we should investigate new technologies. Do we wait until the more wasteful devices are consuming less and, when this component, as a by-product of that research, consumes a larger percentage of the power, do we then start investigating improvements? Makes no sense.

He asserts that the gains will be 2-3% without anything to back that up and he minimizes their claims ("the team did not put this in context ...") while not providing any proof for his own, he has no data to back up his belief that it is more expensive, there cannot be any. It is a new technology, not in production, that will likely get less expensive as it matures.

By bug77 on 3/13/2011 7:12:02 AM , Rating: 3
Aren't you a bitter one?

Yes, in a typical device that exists today, memory is not a power hog. There's a link in the article supporting this claim.

And the researchers did indeed pull a stunt when they said their memory is 1000 times more efficient than current phase-change memory, but failed to compare their memory with mundane DRAM or Flash.

You know, it's not Jason's duty to do a thorough feasibility study about this. This is the researchers' job. Not providing one, automatically raises a few questions, no matter where your sympathies lay.

By GTVic on 3/13/2011 10:55:32 PM , Rating: 2
Not bitter, just annoyed that he uses this as a forum for his opinions that don't appear to be very well thought out.

Apparently Rush Limbaugh thinks this is a science site so this opinion might have some weight.

Everyone knows that memory is not a power hog (percentage wise in a system), supporting this conclusion is meaningless.

The researchers are illustrating that they have made a significant improvement in a technology that is in a research phase. That is not a stunt. There may be many applications for this result other than Jason's cell phone.

Also, no, they are not obligated to make further comparisons, that would be Jason's job, if he can do it objectively and he has some data to back it up.

To throw a few opinions around and basically say that the research is pointless is not responsible journalism.

By bug77 on 3/14/2011 7:42:18 AM , Rating: 2
Not bitter, just annoyed that he uses this as a forum for his opinions that don't appear to be very well thought out.

So, what's a "well thought out" idea then? Cause you still can't do anything with that memory right now. Nor in the foreseeable future.

By NellyFromMA on 3/14/2011 12:45:00 PM , Rating: 2
Why would you even bother to read if you are so annoyed that you have to unconstructively criticize him? Find a new site?

Do you not get enough recognition at work or home that you feel the need to post these things?

By GTVic on 3/12/2011 6:44:53 PM , Rating: 2
If that is all he is saying, he could have put it as succinctly as you did at the bottom of the article and clearly labelled it as his opinion.

What he did instead was a more subtle hatchet job on the technology and the people promoting and working on it by sprinkling his seemingly baseless opinions and put downs throughout the piece.

By Lerianis on 3/13/2011 12:50:58 PM , Rating: 2
Key term here: YET! Jason has a history of not using that term, the fact is that once we find one way to make something, it usually makes it easier by far to find other ways to make something.

By bryanW1995 on 3/12/2011 10:20:08 AM , Rating: 2
I agree. I don't understand why he even wrote the article if all he wanted to do was bash a new and potentially interesting technology. Not everything has to be directly applied to cellphone improvements, jason. There's a whole world/universe out there.

By bh192012 on 3/14/2011 12:59:31 PM , Rating: 2
Your comment that the power savings comment is ridiculous is pointless.

His commnet about speed increase does seem baseless, I see nothing in the abstract that mentions ANY speed increase.

He doesn't mention anything about giving up, what you did there is called a strawman argument.

P.S. wake me up when they start producing something. I've seen 192012 articles on carbon nanotube this and that, and have seen zero actual products. I'm not saying give up, so don't put those words into my mouth. I'm saying, you haven't made much progress until you can usefully make something.

Some Applications
By Flunk on 3/11/2011 5:00:32 PM , Rating: 2
I'm sure there will be some extreme case type applications. Deep space probes or something. Science doesn't have to be profitable or consumer marketable.

RE: Some Applications
By smackababy on 3/11/2011 5:05:09 PM , Rating: 5
Or maybe it will eventually get smaller and cheaper like all other memory.

RE: Some Applications
By TSS on 3/11/2011 6:10:39 PM , Rating: 5
Alcohol accelerates this process.

... oh wait we where talking about electronics.

RE: Some Applications
By vol7ron on 3/11/2011 6:45:40 PM , Rating: 2
This is where NASA comes in to front the costs :)

Honestly, this technology could be very important in the future, especially as things move toward more SOCs and more 3D. I'm assuming this technology can be applied to more than just DIMM RAM, and possibly the on-chip cache. I suppose one day we'll see DIMM-less motherboards.

While there might be other technologies with higher power efficiency issues, I imagine this one was introduced because it isn't as complicated and there are several solution approaches.

RE: Some Applications
By mmatis on 3/12/2011 10:15:09 AM , Rating: 2
Except that, unless this memory is resistant to cosmic rays, there is very little chance it is usable for space applications. I believe you will find that successful space vehicles used hardened electronics much like those spec'd by the military for systems used in nuclear conflicts.

RE: Some Applications
By vol7ron on 3/12/2011 12:15:35 PM , Rating: 2
This is very true. The exterior disturbances affect electronic equipment, but I think that is where better casings are required.

Anyhow, I'm curious how DARPA's anti-matter shield could be used.

Oh come on...
By Iketh on 3/13/2011 5:30:19 PM , Rating: 2
It's brand new tech!! It doesnt have to be spectacular off the starting line. It will grow and mature if you give it a chance.

The remark about why not other applications?? It's brand new!! Let it mature with memory before you throw it at processor designs!

RE: Oh come on...
By bug77 on 3/13/2011 7:24:40 PM , Rating: 2
Are you willing to pay for all inventions that claim will do something useful in the future?
If we had unlimited funding, sure. But we must pick and choose.

By vanionBB on 3/11/2011 6:19:15 PM , Rating: 2
While it offers large energy efficiency games , memory is typically a minimal part of power consumption on modern mobile devices.

I think you meant energy efficiency gains, though energy efficiency games would be awesome...!

This quote says it all...
By integr8d on 3/11/2011 6:32:23 PM , Rating: 2
"We're not just talking about lightening our pockets or purses. This is also important for anything that has to operate on a battery, such as satellites, telecommunications equipment in remote locations, or any number of scientific and military applications."

They're already kissing up to the military for a phat contract;)

Chop Chop
By styrafoam on 3/11/2011 10:02:51 PM , Rating: 2
Get the scientists to work on the tube technology!

Say what?
By Reflex on 3/12/2011 5:29:26 AM , Rating: 2
And the real question here is "why memory?" There are so many more attractive power efficiency targets -- like transistors (for CPUs), the battery itself, wireless modems, and displays.

The real question is why the author things its an either/or situation? There are teams working on those other issues. That does not mean that work needs to halt on memory, or that it in some way takes away from work on other areas.

People can do more than one thing at a time. Companies can have multiple research projects. This is not an either/or situation.

10% facts, 90% opinion
By Robear on 3/12/2011 1:06:00 PM , Rating: 2
I'm sure that many people were saying that the abacus was a much more cost-effective solution when the CPU was invented, and there were the J.Micks of the time that were touting how much more expensive they were and how they were a useless invention.

This is stupid. Scientific progress is good, even if its benefits aren't immediate. The vast majority of this article talks about how this tech is essentially useless commercially because it's too expensive and won't immediately impact consumer electronics.

"And the real question here is "why memory?" There are so many more attractive power efficiency targets -- like transistors (for CPUs), the battery itself, wireless modems, and displays."

Memory is where design STARTS in the semiconductor industry because it's simple. Read a book. Please. Just one...


"If you can find a PS3 anywhere in North America that's been on shelves for more than five minutes, I'll give you 1,200 bucks for it." -- SCEA President Jack Tretton

Copyright 2016 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki