What is Digg?68 Comments
- grgt1994, on 05/14/2008, -0/+32The post is a bit misleading. This development is a potential replacement for silicon in MOSFET transistors, power converters. Think of the power bricks and your computer power supply. Not your CPU, GPU or RAM. So this is not a replacement for the silicon chip in general - at least not yet. Still an interesting article.
- mojoe1185, on 05/14/2008, -1/+30Nerd!
/actually I am very impressed, and a envious - zerhynn, on 05/14/2008, -1/+20He must be great at dinner parties. "You know the silicon chip? Yeah, ::sniff::, well I bested it."
- Gusbob, on 05/14/2008, -1/+17His father woke him up every morning to do mental maths? I thought I had a good reason to complain about waking up early...
- goerg, on 05/14/2008, -4/+18anyway, isnt silicon still better because its deposits on earth are somehow inexhaustible?
i mean...there is really a lot of sand on this planet, but never heard of gallium nitride deserts - slashbot, on 05/14/2008, -0/+10Why are people burying? I think it is a valid question.
No, silicon isn't 'inexhaustible', but at least it is freely found in nature. Ga is not. - Frecklefoot, on 05/14/2008, -1/+12Sounds cool, but I'm suspicious about what the article DOESN'T mention. Like, perhaps, this gallium nitride costs 200x more than silicon? Or that it takes 2 months to fabricate one chip? Whatever the holes are, this material hasn't made a bump in the semiconductor industry. Intel and AMD are still squarely focused on silicon.
- Fordi, on 05/14/2008, -0/+8The same holds true for several known minerals. I think that around 2012, landfill mining is going to step up significantly. I don't know what process will be used to separate minerals from organic waste, but my money's on the TCP (thermal conversion process) developed by Changing World Tech. Separates out the minerals and has the added bonus of converting the organics into various petrochemical analogues - which gives the possibility of reducing the problems involved with peak oil, but at the very least would be useful as a genercized method of plastics recycling.
- RealmDown, on 05/14/2008, -1/+8Maybe on YOUR planet.....
- SaxWan, on 05/14/2008, -1/+9This is cool. They have known many semiconducting materials which have better electronic properties than silicon, but it has other advantages too. Silicon is a very stable and common element which can be easily refined (it needs to be pure) and also there have been billions and billions of pounds invested in r&d and equipment for making silcon devices.
Ergo, it stands to reason that there would have to be a significant improvement in the new material for it to be worth overhauling all that investment in silicon technology. Would be bad-ass tho.
Go materials science! whoop - slashbot, on 05/14/2008, -0/+6Digg the man up. This is exactly right.
The article is a bit misleading on this point. It also starts off implying that he somehow invented GaN technology, which has been the subject of research for quite some time. It looks like he was able to make the first power MOS from GaN.
Still an impressive accomplishment. I look forward to seeing some numbers. - teppicymon, on 05/14/2008, -0/+7According to wikipedia: http://en.wikipedia.org/wiki/Gallium
"One chemist estimated in 2007 that at the current rate of usage, the world's supply of gallium would be exhausted by about the year 2017"
That's even *without* its usage being ramped up by this new idea - malman4, on 05/14/2008, -0/+5Stupid response...He is an adult who received his doctorate, probably nearing 30 years old.
- JQP123, on 05/14/2008, -0/+5Not real sure what this article is talking about because GaN transistors have been in use for quite some time.
- RealmDown, on 05/14/2008, -0/+5He is going to get ALL the chicks now.
- slashbot, on 05/14/2008, -0/+6Those "older people" have been looking at the "same old materials" ie. GaN for years.
His accomplishment was taking that research and using it to make a transistor with it. - Zacko, on 05/14/2008, -1/+7I have met this guy, very humble. Great work from him and from RPI in general!
He was a finalist in the Lemelson-Rensselaer Student Prize competition earlier this year:
http://www.eng.rpi.edu/lemelson/finalist_Huang.cfm
Also check out his school's home page:
http://www.eng.rpi.edu/soe/ - BFisch06, on 05/14/2008, -0/+4Just one envious?
- zerhynn, on 05/14/2008, -0/+4The only thing that can one-up that is that damn ***** who's walked on the moon. Ruins every dinner party man, every one.
- ketemphor, on 05/14/2008, -0/+5GaN is HUGE in the semiconductor industry. Maybe not the CPU side of the industry, but it's been used in a crapload of high-brightness LEDs and RF/microwave diodes. Mixed in with Indium or Aluminum and you have the basis of every blue LED on the planet. They might not be sexy semi's, but they're still semis ... :)
- JQP123, on 05/14/2008, -0/+4GaN technology has been in use for quite some time (see reference below) so something is definitely missiing in this article.
http://www.nitronex.com/about.html - slashbot, on 05/14/2008, -0/+4I saw that as well. I was going to cite that here, but the cited source looked dubious. I mean the citation's grammar doesn't exactly lend it credibility.
"Augsberg University Calculate When Our Materials Run Out" - JQP123, on 05/14/2008, -0/+3"Huang's father woke him up early every morning to practice mathematical calculations without a calculator."
So the key to new and creative ideas lies in rote repetition of old ones? Sorry, I don't buy it. - GeorgeStone2, on 05/14/2008, -1/+4I think older people look at the same old materials the same old way.
Young people come into it and take a whole new look at things.
I wouldn't be surprised if this is the best thing that dude thinks up... As if he would need to think anything else up after this. - spikespeigel42, on 05/14/2008, -0/+3RPI has become such a good school, alot of really cool stuff has come out of it. Remember that blacker than black thing they invented a few months ago?
- xptweakerntn, on 05/14/2008, -0/+3Perhaps make this the most dugg down comment in the history of digg? Why even waste your time leaving comments like the one above?
- Fordi, on 05/14/2008, -0/+3Designing a field-effect transistor at microchip scales is tedious work, requiring a very comprehensive knowledge and investigation of the electronic properties of GaN and a number of dopants. Of course, to go into widespread use, he also has to design functional resistors, diodes and capacitors that can be laid with it in order to form logic circuits.
Still, the higher melting point (>2500deg C) means a higher entropy temperature, and thus, electronics with better heat tolerance *small overclockgasm*. So this is definitely research worth doing.
The important question, though, is whether existing chip processing infrastructure can be parlayed into fabbing GaN-type chips with little modification. Low infra cost would be a huge boon to adoption. - GeorgeStone2, on 05/14/2008, -2/+6I have him beat.
"You know Metal Gear Solid on the PS1. :Sniff:, well I beat in in 2 hours." - raybury, on 05/14/2008, -0/+3GaN, SiGe, and other Si alternatives have been used for the past few years. This is not new.
- sgbooth, on 05/14/2008, -1/+5Marked as inaccurate. The article is full of will assertions that are completely false. GaN transistors have been around the lab for years and Huang did not invent them. In fact, they are commercially available from Nitronex.com. Perhaps Huang invented a method to make GaN transistors, but there is no detail as to what makes his transistors novel. Like what kind of substrate did he grow the GaN on?
- slashbot, on 05/14/2008, -0/+2Same here. This article is missing a lot of substance. I will wait before proclaiming this a "revolutionary" new technology
- slashbot, on 05/14/2008, -1/+3I'm sorry, but where is he conducting this research? That's right, an American university.
The American education system has some very fine schools. What you're referring to isn't a fault of the education system, but a product of the culture's emphasis on higher education. - MrFurious2k, on 05/14/2008, -0/+2Why is it you hear stuff like this - but then seldom see the products come to market? Sometimes these articles are too sparse on the information to know whether or not they're blowing smoke. Dugg the article, but I'd still like to know if this has any real world marketability.
- ketemphor, on 05/14/2008, -0/+2Ummm, you know those transistors in your CPU, GPU, and RAM? They're Complementary (meaning P and N) Metal-Oxide Semiconductor Field-Effect Transistors, i.e. CMOS / MOSFETs.
- kurtwinter, on 05/14/2008, -0/+2Here's the reason why: silicon cannot continue to shrink in dye pitches much further due to its fundamental properties, and the fact that extra deposits in the fab process have to be cleaned with water. Lots and lots of water. The use of clean freshwater in the fab process, and its ability to remove excess particles on higher density chips is driving this development.
- chuckDontSurf, on 05/14/2008, -0/+2"i bet you you don't even know whats 18*18 is out of your head..."
who cares? I can't speak for slashbot, but I'm a computer engineer, and I have no need to know 18*18 off the top of my head. I have a calculator; I just plug it into there. Now, if my job involved a lot of multiplication of invovling 2 and 3 digit numbers, you can bet I'd get a lot better at it quickly. - Fordi, on 05/14/2008, -0/+2On the shoulders of midgets, we built up this machine! (YEAH!)
- Shadowgamers, on 05/14/2008, -0/+1I wonder where I can find some of his papers...
- wonderchemist, on 05/15/2008, -0/+1Well magnets would get the ferrous bits.
- EpicSelekta, on 05/14/2008, -0/+1He wants us to suck him off because no one else will. In fact, even Miss Teen South Carolina's ugly cousin turned him down.
- steveoco, on 05/14/2008, -2/+3Lets hope this leads to something revolutionary.
- slashbot, on 05/14/2008, -0/+1True. But this would be for power mosfet applications, where scale is less important.
- Shadowgamers, on 05/14/2008, -0/+1Eh. Looks like I'll just wait for it to drop on Wikipedia
- stefang7, on 05/14/2008, -0/+1As a lot of people mentioned GaN transistors have been around for a long time. Actually 10 years ago I was running experiments on GaN transistors to characterize breakdown behavior at very high voltages. Since is has a very large bandgap the excitement was about high voltage high temp applications (space, missiles, etc...). Since it's also a direct bandgap semiconductor (conduction band is aligned to the valence band so electron recombination results in photons) it's probably now most used for LEDs and lasers. The article is not clear, but my guess is that this guy figured out a way to make a cheaper and better oxide for GaN. The reason Si is in 99% of all chips is because is has a cheap and excellent oxide (SiO2). But for small dimensions (
- MindlessSpartan, on 05/14/2008, -0/+1"Huang first developed a new process that demonstrates an excellent GaN MOS (metal/oxide/GaN) interface. Engineers have known that GaN and other gallium-based materials have some extremely good electrical properties, much better than silicon. However, no useful GaN MOS transistor has been developed. Huang's innovation, the first GaN MOSFET of its kind in the world, has already shown world-record performance according to Huang."
The article mentioned exactly what he did. If you want to know the details, go read the paper he either has or will publish on this. ScienceDaily summarizes major events in science, it doesn't copy/paste somebody's research paper. I'm also guessing the exact details as to how he made this will be released in the patent the University will immediately be getting on this, he just sailed his way into insta-tenure if it pays well enough.
On another note, if he hopes to make use of this he needs to find some investors and go into business himself. If he winds up working as a product engineer in the corporate world he can forget about ever being relevant again. The best engineers don't invent, they innovate. - stefang7, on 05/14/2008, -0/+1(cont from above - Digg error on submission)
... for small dimensions - Thuktun, on 05/14/2008, -0/+1Damn kids, get off my lawn!
- Iztikeit, on 05/15/2008, -0/+1That's why people don't understand why some people are terrible at math. It's because math geniuses work their asses off, or brains rather. I hate it when people complain about not being "good" at math.
- smallwang, on 05/14/2008, -0/+1Good for him
- Smuikas, on 05/14/2008, -1/+3Combined with my loathing of blue LEDs, I hate GaN.
(seriously, whoever thought that having 500 intense blue lights all over everything would be cool? they hurt my eyes.) -
Show 51 - 68 of 68 discussions
Browsing Digg on your phone just got easier with our enhancements to the 
© Digg Inc. 2009
— Content created and posted by Digg users is