digg

Join DiggAbout

Discover the best of the web!
Learn more about Digg by taking the tour.

RISC vs. CISC in the mobile era

arstechnica.com — For any Ars Technica readers who've been with us since 1998 and who fondly recall the "RISC vs. CISC" wars of yesteryear, I've got great news: the battle is back on. Here's a look at thew new state of play.

Bury
show profanity settings
expand all
  • maxino, on 05/19/2008, -16/+3Are still You MrBabyMan? ......DUGG
    The story is great!
    P.S. I think the answer is ARM 7-9-11...... , it's not enough?
    P.P.S Look this -> More MIPS, less mA - get AVR32! http://avr32linux.org
    • Kennerk, on 05/19/2008, -1/+1talk about dyslexic..... are still you?! i think thats supposed to be "you are still" but im not sure....
  • HafizD, on 05/19/2008, -27/+43Sorry MrBabyMan, but this is a dupe.
    http://digg.com/tech_news/RISC_vs_CISC_in_the_mobi ...
    • MrBabyMan, on 05/19/2008, -19/+34By the timestamp, it looks like we submitted this at the exact same time.
      • HafizD, on 05/19/2008, -9/+5My bad... You are just too quick :)
        Btw nice video.
        • MrBabyMan, on 05/19/2008, -9/+17Thanks. My apologies for a double submission. That's a major flaw Digg has yet to fix. That's the reason there's always 6 or 7 Zero Punctuation submissions every Wednesday at the same exact time.
      • Fergy, on 05/19/2008, -7/+5You beat him by 1 minute :)
      • noseeme, on 05/19/2008, -1/+27Why did your story get the "2" after it?
    • Sabin, on 05/19/2008, -10/+6Submitted at almost the same time and only one of you is a whiny ***** that isn't getting my digg.
  • QGYH2, on 05/19/2008, -7/+6Atom will change the world..
    • noseeme, on 05/19/2008, -1/+1World will be comprised by atoms.
    • Tyr7BE, on 05/19/2008, -0/+8Not until its power requirements and pipeline complexity rival what ARM puts out. Speaking as someone who works with these sorts of processors a lot, Atom simply will not catch on until they address these issues.
      • shredswithpiks, on 05/19/2008, -2/+3While this is true, the issues will eventually be addressed and someday and the idea behind the Atom will take over the mobile market. The benefits of one common architecture across the board is just too attractive to not work towards...
        • MoofTheStoof, on 05/19/2008, -1/+2I think there was something in that Kool Aid...
      • grumpyrain, on 05/20/2008, -0/+2Correct. Most developers couldn't give a crap about RISC vs CISC. There are pros and cons to both. RISC is simple and as such does not require the power overhead that is the bain of many embedded devices. But it also requires more instructions to do the same thing, and removes the possibility of a particular optimisation being discovered that drastically improves speed.

        The challenge that intel needs to face if they want x86 to hold its own against ARM and others is consumption vs speed. If they can produce something which is just as quick as ARM (in real terms, not clock speed) and can keep the wattage in check, and can also compete on cost, then they have a good shot. Otherwise, they don't.
        • Ademan, on 05/20/2008, -0/+3This is why C and letting the compiler sort out architecture specifics is the way to go, so that the underlying architecture doesn't really matter from a coding standpoint, that decision can be reversed, the platform can be changed ad-nauseum with little need to change code (you'd need to change things like code generation, and possibly tweak the standard C library, but aside from that...)
  • maxino, on 05/19/2008, -4/+6The stories have the same links.... there is something wrong......
  • MapShatter, on 05/19/2008, -24/+32So Mr. Babyman admits that this story, and the story submitted by hafizD here : http://digg.com/tech_news/RISC_vs_CISC_in_the_mobi ... were submitted at the exact same time. Yet HafizD has 10 Diggs while babyman has 91 already... anyone else see a problem with this? If I could change it, I would only allow stories to be submitted without showing who submitted them, that way the articles would truly be Dugg by merit alone. People who crave popularity via an internet website are kinda pathetic to begin with. And honestly, did any of us actually think Mr.Babyman would stop submitting to Digg? Even though he told us he would stop, he is right back at it. If the rest of Digg is going to support/condemn articles by the person submitting them, then I will bury Mr. Babyman because he doesn't come accross as that honest of a person to begin with. Sorry your story will probably not hit front page HafizD
    • HafizD, on 05/19/2008, -0/+15[Probably] not.
      lol.
    • chanop, on 05/19/2008, -4/+11Get over it, HafizD is being a good sport about it, so no harm, no foul.....
    • Modestexcuse, on 05/19/2008, -14/+12I'm very tired of the MrBabyMan bashing. It's really getting old. His submission got more traction.

      He has more friends. He has gained friends from submitting good stories. It is quite logical.

    • vinnyvenus, on 05/19/2008, -3/+8Why do you guys hate mrbabyman? He seems like a really nice guy. He even makes fun of himself in this video.
      http://www.youtube.com/watch?v=PhLoghrZ9vE
    • schoate09, on 05/19/2008, -3/+5Shut the ***** up about it. MrBabyMan has more friends, therefore, more people saw his and dugg it. That's the way the system was designed.
      • sleepwalkers, on 05/19/2008, -0/+4That's exactly what was just said. I believe he was saying that is part of the major flaw in the system.
        • mikesbaker, on 05/19/2008, -2/+1then f off and go to fark or reddit or some *****
    • sleepwalkers, on 05/19/2008, -0/+2To be fair, MrBabyMan never said "I'm not submitting to Digg ever again!"
      He said if a majority of Digg dislikes him, then he'll stop.
    • dopplerdog, on 05/20/2008, -1/+1Play the ball, not the man.
  • noseeme, on 05/19/2008, -4/+6You know he was just sitting there, waiting for the next Arstechnica story to pop up...
  • Goblin, on 05/19/2008, -2/+21RISC is good

    /ZeroCool
    • cao4, on 05/19/2008, -1/+1Indeed. RISC architecture is gonna change everything
    • jabberwolf, on 05/19/2008, -1/+3That was such a sad sad SAD movie where the hackers were using older macs to hack HAHAHA

      Was great for laughs and a first look at Angelina Jolie! YUM !
    • warkwark, on 05/19/2008, -1/+0No it's not

      /ZeroCool
  • mark076h, on 05/19/2008, -1/+2i thought i was a huge geek but i get lost and mind boggled in arstechnica articles like these
  • gizzymo, on 05/19/2008, -2/+2RvC 2 > AvP 2 ;-)

    p.s. think I'll get dugg down big time by all the people who dont get the reference....
  • mark076h, on 05/19/2008, -7/+5MrBabyMan has decided to stay on digg http://twitter.com/MrBabyMan/statuses/815084938
  • tomazkovacic, on 05/19/2008, -1/+6RISC ... FTW!

    Keep it simple and smart in the future. (Ex: ARM latest development)
  • chuckDontSurf, on 05/19/2008, -0/+5"No, in today's world, when Intel aims both barrels of its formidable fab capacity directly at a new market segment, it's not a joke to anyone who ends up in the crosshairs."
    Ain't that the truth. I hope ARM can keep them at bay.
  • Aerandir, on 05/19/2008, -11/+20Welcome back MrBabyMan.
    Now i have to bury this, just because of you.
  • shredswithpiks, on 05/19/2008, -2/+7oh hey look, it's this discussion again.

    the thing about RISC vs CISC is that either one is just a methodology or "general guidelines" for designing an ISA. They both have their trade-offs and in the end the "correct" answer is always going to be "somewhere in the middle, adjust depending on your application."

    On a related note, I don't know if anyone else has been keeping up with the Intel Atom (http://www.anandtech.com/cpuchipsets/intel/showdoc ... but I'm putting my money on x86 for the future mobile devices. Maybe it will be a few years in the future, but the compatibility between hardware platforms of having everything on a compatible ISA is more important than RISC vs CISC (or x86 vs Sparc vs Whatever).
    • dkapuchino, on 05/19/2008, -3/+6Dugg down for an actual comment on the article. Comments should be restricted to issues revolving MrBabyMan. Thank you, and have a good day.
    • RungeKutta, on 05/19/2008, -1/+2Yea I agree with you there, the familiarity of x86 will probably make some big inroads into this market. Not because x86 is so great (I'm not an ISA evangelist) but because so many engineers already know it and the tools are there too. If Intel can get MS to jump on Atom, doubly so. The possible popularity of Atom will probably for the same reason that MS could attract so many game developers to the Xbox, DirectX. I doubt ARM and the others will ever go away but their market-share in the area might drop quite a bit.
      • dkapuchino, on 05/19/2008, -0/+3As far as embedded systems go, PowerPC, ARM and MIPS have been the rulers of the market. So many embedded engineers already know them, and the tools(cross platform compilers/debuggers, In-Circuit-Emulators, Board Support Packages) are there too. It's just plain wrong to consider a PC as the platform from which the people developing mobile phones will come from. They'll come from other embedded platforms, and they'll be familiar with other processors. Application programmers on the other hand, will prefer Java solutions, so their application can be portable.
  • altjeringa, on 05/19/2008, -6/+5While intel has always designed really awfull processors, there is one thing they do very well. They have good salesmen who know how to treat their clients with nice kickbacks. Intel will destroy ARM just as it did those in the RISC camp on the PC, and we'll all be left with cpu's in our mobile devices that cook our hands and drain our batteries. The game isn't about what technology is the best, it's about who plays ball. Get the cash flow first, stack layers of crap on the technology latter and make it work.
    • shredswithpiks, on 05/19/2008, -1/+4the core 2 series are really awful processors?
    • ElectricKetchup, on 05/19/2008, -0/+4Intel will destroy ARM?!? Did they quit making their ARM CPUs? I hope they're not trying to destroy the ones they make... that would be stupid.

      Intel tried to destroy the RISC camp on the PC? I remember when they were trying to push the Itanium processors (which were RISC), and they eventually lost to AMD's AMD64 (which is CISC).
      • JayD16, on 05/19/2008, -1/+4The reason Intel still uses CISC as well as the reason Itanium failed is because the x86 has such a huge code base that companies aren't willing to design entire systems just for a faster RISC machine.
  • greeniemeani, on 05/19/2008, -5/+6***** THE RISC.
  • cbeach, on 05/19/2008, -11/+12Buried for MrBabyMan. His botnet (or huge social network, or whatever it is) is an unfair advantage over other submitters, and I'm fed up of seeing his stupid name, and fed up of all the silly drama.
    • Junior612, on 05/19/2008, -5/+4Thanks for adding to the silly drama you stupid ***** twit. Just shut up and DIGG.
    • MacParrot, on 05/19/2008, -4/+17Here's an idea. Digg the story up or down based on its content NOT on who submitted it. All this whining about MrBabyMan just seems like jealousy.
    • mikesbaker, on 05/19/2008, -4/+1in two years only 265 comments? I will give a crap about your opinion on the matter after you break 1000 until then keep crying
    • HonoredMule, on 05/20/2008, -0/+2Here's another idea. I come to digg for the content. I don't care who submits it, but if a character (i.e. MrBabyMan) has a reputation for submitting a large quantity of only high-quality articles, then I do well to follow his submissions...it reduces the amount of crap I have to sift through to get my digg fix.

      I care about fair content promotion because I'm here for the content. If that results in "unfair" submitter promotion, that's just fine. I really couldn't care less, and can only assume that those who do care fail to appreciate digg's content-aggregation capabilities, preferring to see it as a popularity contest. But guess what? The consumers here vastly outnumber the producers (as in any market). It's really not about the popularity you (producers) want, it's about the content we (consumers) want.
  • dwhitbeck, on 05/19/2008, -1/+3The CISC architecture of the x86 family was initially an attempt to mimic main frame architecture, which made no sense then and makes no sense now except for the need for backward compatibility. As such it was a fat mistake.
  • kindwarrior, on 05/22/2008, -0/+1Great article: Really the RISC vs CISC debate on technical merit has been long over (RISC makes more sense) the real battle is between the value of Legacy instructions vs better architectures that are not compatible with the x86 legacy.

    @dwhitbeck: I don't fault intel for it's early approach to x86 architecture; until many of the look ahead technologies and predictive cache analysis tools came together to allow RISC imbeding comonly called instructions on chip made a lot of sense. The mystery is why they clung to the x86 instruction set throughout the '90s (the answer to the Mystery is "pressure from Microsoft" -- so yet another example of how Microsoft has been a blight to progress in the computer industry).
  •  

    Login or register to add a comment

    Create a new account or login to join in the conversation on Digg. You'll also be able to Digg stories to help promote things you like.


© Digg Inc. 2008 — Content posted by Digg users is dedicated to the public domain.
DIGG, DIGG IT, DUGG, DIGG THIS, Digg graphics, logos, designs, page headers, button icons, scripts, and other service names are the trademarks of Digg Inc.