IBM Power Family Slide - Power 6 65nm

Posted:
in Future Apple Hardware edited January 2014
via the Inquirer:



UNFORTUNATELY our Japanese is virtually non-existent here at the INQ, but there appears to be some information about IBM Power CPUs up on a site on the islands.



Apart from the Power 5+, a 90 nanometer shrink, there's also info about the 2006 Power 6, shrunk to 65 nanometres.



Click here for slide image

Comments

  • Reply 1 of 15
    marcukmarcuk Posts: 4,442member
    I wonder if the chip industry is going to shoot itself in the foot regarding die shrinks, Prescott is the living proof that Intel couldn't do 90nm. AMD hasn't done it yet, Freescale, hasn't done it yet, and IBM has had some major difficulties.



    What if 65nm,45nm,32nm really doesn't work at all? No one thought 90nm would be a problem, but it has some major ones. They need to discover why 90nm was such a problem, and fix it before they start 65nm. Hate to think IBM built a $4billion 65nm fab that didn't work. LMAO if Intel did thou.
  • Reply 2 of 15
    emig647emig647 Posts: 2,455member
    HAHA Marc!!!



    It's interesting that IBM is the first MAJOR processor producer to really start shipping 90nm consistantly. Aren't there 90nm graphic card chips out there?



    I believe once they really get consistant with 90nm they won't have a problem moving to 65nm... any projected date when they were thinkg about this?



    Does that mean the 980 will be 45nm? Seems like the g5's get the Power+ side of things, each + is a smaller die shrink.
  • Reply 3 of 15
    onlookeronlooker Posts: 5,252member
    90nm was a learning process like anything else. Once they figure out why it gave them problems they will make estimations based on those facts, and in turn prototype 65nm. Again based on findings with the prototyped 65nm they will move to refinement, and hopefully more prototypes. Then eventually come out ahead of the bunch. They are still ahead of the race in shrinking those little suckers so far. I have no idea if they have all the bugs out of the 90nm G5's yet, but they said they were doing better recently.



    I think the powerMac looks better every time I look at it, but I still cant do sh*t with that video card, nor do I have Maya Unlimited on the Mac, but If I had the video card options I need right now I'd say the dual 2.5 GHz G5 would be the computer I bought after WWDC. Although I'd still want to see independent testing results from a fair non Mac'ish testing group before I committed to that.
  • Reply 4 of 15
    emig647emig647 Posts: 2,455member
    I never thought I'd see you say that,



    Its weird like with all apple products, we get caught up in technology before they come and out, we get pissed when they don't come out with FUTURE technologies... a few weeks go by and they start to grow on us.



    I do this with every product they release, even if I don't have any intention of buying the product.



    After reading those articles you posted on the graphics cards, I definitely have to agree that they could definitely use more graphics power.



    And be careful when you look at benches from "non-macish" people. Usually the only benches that work are ones that you do yourself. Personally I use Cinebench... but that depends on graphics cards. But it also relies a lot on cpu power, This will be the bench marking program I will use to see how my dual 2.5 really performs... sucks I have to wait for another 4 weeks for it ;(
  • Reply 5 of 15
    emig647emig647 Posts: 2,455member
    I also wanted to mention, a lot of companies had problems going from 180nm to 130nm... although not as big of problems... hopefully it is going to get exponentionally harder from here.
  • Reply 6 of 15
    marcukmarcuk Posts: 4,442member
    I also remember someone from IBM recently saying "that scaling died somewhere between 130nm and 90nm. So i think I might have a point. I also remember hearing that below 130nm "Cosmic Rays" affect the chip really badly!.
  • Reply 7 of 15
    emig647emig647 Posts: 2,455member
    Can I ask what "Cosmic Rays" are?
  • Reply 8 of 15
    Quote:

    Originally posted by emig647

    Can I ask what "Cosmic Rays" are?



    Alpha particals and crap like that. They can just come shooting in from space. They come from solar flares and supernovas, and such.



    I think.
  • Reply 9 of 15
    onlookeronlooker Posts: 5,252member
    Quote:

    Originally posted by Transcendental Octothorpe

    Alpha particals and crap like that. They can just come shooting in from space. They come from solar flares and supernovas, and such.



    I think.




    D000d. I've never laughed so hard.

    Thanks for that.
  • Reply 10 of 15
    I aim to please.
  • Reply 11 of 15
    the cool gutthe cool gut Posts: 1,714member
    Quote:

    Originally posted by emig647

    Can I ask what "Cosmic Rays" are?



    Alpha particles cause bit rot, cosmic rays donot (except occasionally in spaceborne computers). Intel could not explain random bit drops in their early chips, and one hypothesis was cosmic rays. So they created the World's Largest Lead Safe, using 25 tons of the stuff, and used two identical boards for testing. One was placed in the safe, one outside. The hypothesis was that if cosmic rays were causing the bit drops, they should see a statistically significant difference between the error rates on the two boards. They did not observe such a difference. Further investigation demonstrated conclusively that the bit drops were due to alpha particle emissions from thorium (and to a much lesser degree uranium) in the encapsulation material. Since it is impossible to eliminate these radioactives (they are uniformly distributed through the earth's crust, with the statistically insignificant exception of uranium lodes) it became obvious that one has to design memories to withstand these hits.
  • Reply 12 of 15
    pbpb Posts: 4,255member
    Quote:

    Originally posted by Transcendental Octothorpe

    Alpha particals and crap like that. They can just come shooting in from space. They come from solar flares and supernovas, and such.



    I think.




    Cool, where is my crap helmet? Some of you guys here have true inspiration .
  • Reply 13 of 15
    wmfwmf Posts: 1,164member
    Cosmic rays are not a joke. The whole point of ECC RAM, for example, is that it corrects errors caused by cosmic rays. It's even worse at high elevation (like Los Alamos), because there's less atmosphere to protect you. As transistors get smaller, it is easier for a cosmic ray to flip a bit.
  • Reply 14 of 15
    airslufairsluf Posts: 1,861member
    Kickaha and Amorph couldn't moderate themselves out of a paper bag. Abdicate responsibility and succumb to idiocy. Two years of letting a member make personal attacks against others, then stepping aside when someone won't put up with it. Not only that but go ahead and shut down my posting priviledges but not the one making the attacks. Not even the common decency to abide by their warning (afer three days of absorbing personal attacks with no mods in sight), just shut my posting down and then say it might happen later if a certian line is crossed. Bullshit flag is flying, I won't abide by lying and coddling of liars who go off-site, create accounts differing in a single letter from my handle with the express purpose to decieve and then claim here that I did it. Everyone be warned, kim kap sol is a lying, deceitful poster.



    Now I guess they should have banned me rather than just shut off posting priviledges, because kickaha and Amorph definitely aren't going to like being called to task when they thought they had it all ignored *cough* *cough* I mean under control. Just a couple o' tools.



    Don't worry, as soon as my work resetting my posts is done I'll disappear forever.
  • Reply 15 of 15
    programmerprogrammer Posts: 3,457member
    Quote:

    Originally posted by AirSluf

    I guess you haven't read anything about the avionics development for the 777. Memory densities are finally high enough in avionics rated hardware that engineers expect 4-6 bit flips per pacific crossing flight above 32K ft due to cosmic rays and other solar generated radiation. The software and hardware has been heavily infused with error detection and correction techniqies. 747s flying with older digital hardware (which had much lower densities) would average one bit flip per month--vice multiples per Pacific crossing. Those flips still happen at lower altitudes, just nowhere near as often.



    I don't know about that Intel study, but the boys at Boeing flew their equipment in 747s as ride-alongs for a couple years. I think the extra $$ they decided they had to spend after that for the error detection and correction facilities says quite a lot about the real threat of cosmic radiation to digital data in hardware.




    The altitude makes all the difference. It is the atmosphere that is protecting us from all that radiation, so sitting 30,000 feet farther down really drops the likelyhood of having a bit flip. I wouldn't trust your financials in Excel while on a plane across the Pacific though (err... assuming you'd trust Excel at the best of times ).



    The other factor is that EEC memory doesn't really buy you a lot of security -- it can miss errors fairly often.
Sign In or Register to comment.