longshot

Free Play Account
  • Content count

    349
  • Joined

  • Last visited

Posts posted by longshot


  1. i still think the GPU should be easily upgradeable. it wouldnt be too difficult to have a modled 'piece' that fits into a slot (think laptop pcmcia cards but with a mold that conforms to and matches the iMac exterior).

    The problem isn't matching the exterior but tying the thermal management of an upgrade card into the internal airflow for cooling. You either:

    1) leave an over-sized thermal management (better-gpu= more power)system in the enclosure and hope that your thermal contact between the slide-in gpu (sans heatsink) and the thermal management system doesn't leave your gpu temperature too high

    2) allow the system to account for the on-card thermal management system (say goodbye to thin systems and hello to a much larger blower to overcome additional airflow impedance and competing flow directions). The current airflow is balanced for the finstacks/impedances that it is designed for. To try and account for all market-potential finstack impedances/designs on one conduit section requires at least one airflow sensor added (and calibrated) into each unit and increasing the blower height (bad for thin enclosure). TBH, it probably couldn't be done without 2 sensors (one for each leg of the conduit and good deal of smu code to interpret the output of the sesnsors) and still requires increasing the size and noise of the existing blower.

    3) disconnect the gpu thermal management system off from the cpu thermal management system (currently joined by heatpipes and common finstack). Assuming that the customer doesn't make a mistake in thermally attaching the gpu card to the Apple-designed gpu thermal-management system and that none of the baffles are dislodged/damaged by sliding the thermal control system into/out of the conduit, you still have added in an impedance from the additional GPU only cooling finstack.

    The problem is compounded by the historically bad power-management code shipped by graphics companies especially with new products. The graphics companies get kudoes for performance so their code keeps the gpu spun up almost all the time (and often ignores system calls to reduce power). When the blower to cool the extra power is sitting 18" in front of your face, this may not be exactly what you want in a customer experience.

    Please note that this is a personal response and in no way an official statement by Apple, Inc., or its Product Design group.


  2. Perhaps I am not making myself clear. In what you are calling the "Game Loop" does that include the Physics Engine (which calculates all the motion/acceleration of objects in the game)? Wouldn't a physics engine update of 100 times per second versus 1 per second result in wildly different physical outcomes? An example:

    A bullet fired from a gun at time 0...

    100 updates per second system - the bullet's velocity is recalculated at 10msec intervals so drag is applied correctly, steadily increasing deceleration.

    1 update per second system - the bullet's speed is re-calculated after 1 full second during which the bullet moves at full muzzle velocity for the entire second. The system would then apply a huge amount of deceleration for the next second.

    The first bullet might have travelled 500m while the second one might have travelled 750m.


  3. Tautu:

    Be careful with this. The ATI chips came in hot (i.e. burning more power than they should) which is why they were down-clocked until ATI got better control on power. The GPU also ignores calls to lower it's power-consumption when the CPU needs more power. Running with the gpu wide-open increases the chances that too much power (from both CPU and GPU) will go into the thermal solution, causing local dry-out of the heatpipe. The CPU has over-temperature protection so nothing too bad should happen to it; however, the GPU does not. If the dry-out occurs at the GPU, you could be looking at a dead-unit.


  4. Voyager2 and Ivertti:

    Seriously, thank you guys for being customers and for being understanding when things don't come out exactly as desired in mass-production. The company usually does a good job of making it right for the customer when something unforeseen happens. Innovation and perfection have never played particularly well together.

    Voyager2, I'll trade you in-depth walk-throughs of the design for helicopter flying lessons. I do volunteer work for emergency response when I am in CA, and I have to admit, it's a huge rush to ride the ANG's Blackhawks to be dropped off to a remote spot in the mountains. We all race to respond to incidents in the Wilderness areas because "the early bird gets the worm" as it were. The latecomers usually have to hike or 4x4 to their (closer) assignments. ;-)

    Xgidds:

    I am not a bad guy, and in re-reading I regret some of the more direct statements towards you. However, you lashed out pretty hard at me and people I work with every day.


  5. Btw' date=' Apple are a**holes, I used to work for the feckers. They dont give a toss about their customers, unless high end. Hence the release of the intel Macs.[/quote']

    Nope, we uncaring "feckers and a**holes" don't care at all. Nope, we struggled to deliver the Intel products six months before industry expectations because we knew that the improved performance didn't matter at all to our customers. We could have just waited the six months (incidentally about when rumor-sites claim that Intel should be delivering a lower power chipset) rather than struggling to cool almost 90 watts of power in a thin portable enclosure. We forced lower power-saving modes compared to every other manufacturer to extend battery life just so customers using the portables had to suffer longer while using our products. We delayed the MacBooks and the Minis as long as we possibly could to force more people to buy the more expesnsive MacBook Pro product.

    We travel constantly to Asia because we just love China's industrial zones and the fine cuisine (seasoned with the delicate aroma of leaded fuel) that can only be experienced there. In fact, we spend extra hours every week just plotting ways to screw over the customers (we originally intended that the Ipod would give electric shocks whenever the customer touched it, but we screwed that design up too.). We generally sign over our patents to the company willingly (rather than hiding the idea and starting our own companies as engineers tend to do elsewhere) because we are too stupid to succeed anywhere else. We sweat tolerances to 0.05mm because bigger numbers are too hard for us. We do our own design in-house rather than relying on engineers in Taiwan/China because we prefer having conference calls in foreign languages until 10 PM to ensure that the design intent is understood.

    No, Apple isn't perfect - largely because we must work with the same vendor base as the rest of the industry while trying to control cosmetic reveals and tolerance stackups much tighter. Imagine trying to get a separate company to TRULY internalize quality control and cosmetic impact when every other computer company just screams at them about quantity shoved out the door per hour. We also have such an understanding customer base... they only want lighter, faster, smaller, full-featured, easier-to-use, cheaper products with bigger displays that run cooler while having no fan noise. Oh, and of course, they must double as works of art. We must be incompetent that we keep missing such an easy target.

    Since you used the word "feckers"and the phrase "give a toss", are you from the UK/Ireland? I doubt that you are old enough to have worked in the last production plant we had there. So exactly what did you do for Apple there or anywhere else for that matter that gives you such keen insight into the corporate culture and our lack of concern for customers at Apple? Were you a sales clerk who got sacked from a UK Apple Store and has an axe to grind without really knowing anything about Apple? If you worked for Apple Computer, Inc., when were you employed? In what department and what was the name of your manager?

    As I doubt you'll have the stones to reply meaningfully, maybe I should just deposit my latest paycheck and laugh at you personally for continuing to buy when you know how little we care about your purchases. That would be what an uncaring a**hole would do, eh?


  6. Keep me in mind for testing. I have the 17" MacBook Pro. Oh, I can answer your general mechanical questions about the portables via PM- I am one of the engineers who designed the MacBook and the MacBook Pro (still unsure about the name-change). Software, driver, and applications questions are likely to be met with a blank stare.


  7. Seri Thai was one part of it. Apparently, one of the other politicians (Pridi) went into hiding in Thailand so that his signature was forged on the document declaring war. He then went on to form a resistance movement within Thailand that eventually linked up with the external movements (Seri Thai in the US and a weaker one in UK. I forget the UK leader's name but he was less successful because he was perceived as a Royalist seeking to return the absolute monarchy that they had thrown off a few years before). The whole episode is confusing, but it seems like a lot of collusion on the part of the Thai politicoes to comply with Japan's demands to the letter of the law while executing them so poorly that their actual intent was clear to the Allies.

    They must have done something right because the world community had no objection when the "dictator" responsible for leading Thailand during the war was voted back to power in 1947.


  8. I'll go ahead and close it out. The Thai ambassador to the US refused to deliver the war declaration to the US Secretary of State, Cordell Hull. He told Hull what was in his pocket and said he felt it did not represent the desire of the Thai people, so he instead resigned his position and volunteered to help the US to create a Thai resistance movement called Seri Thai.


  9. Voyager2, the internal numbers are between 2x-4x better for the new hardware. Running WW2OL is still problematic on the pre-prod hardware as the software is still maturing so I can't do a direct compare yet. I will be given a personal unit within a week of the beginning of production. At that point, I can do a direct compare to my current Powerbook and post numbers. The only concern that I might have is whether any part of WW2OL would hit the emulators which will bork performance. PM me at intervals to keep this testing in my mind... as you might imagine, I have a fair few things keeping me busy until we get into production. :-)

    They really are that great. I would buy one if I wasn't addicted to the bigger screen size. The marketing commit lowballs to guaranteed performance of 1.83GHz. These babies with run-of-the-mill chips are carrying 2.17GHz with ease in normal room temps. Unless you live in Arizona, you'll be exceeding specs. (please note that the previous comments are my own and do not represent any official company guarantee of performance)