Final-Fan said:
Squilliam said:
The issue with the Xbox 360 is mainly a GPU issue.
The issue with the Nvidia GPUs is a GPU issue.
The reasons why I bought it up is because of the parellels between the two issues with regards to solder/packaging issues of the GPUS brought about by partially by an increase in heat output of GPUs over time.
1. The Nvidia GPUs were tested by competent engineers by both Nvidia and the companies which installed them in the laptops. Since it passed at least two phases of competent testing and was installed in envirionments which were inside the specifications for the parts. The issues only became aparant once the devices were released into the wild. So saying that systematic quality control errors was the cause of the issue is moot if the problems can show up with several different companies who all tested the design and passed it through their own quality control procedures.
2. The issues related to the Nvidia GPU would have required a complete respin of the chip, so if the same were true of the Xbox 360 they would have had to remove the console from the market for a minimum period of 6 months and more likely 9-12 months of testing and design work. For that reason had an issue become aparant in the PS3 GPU they would have likely faced a further 9 months of redisign and testing and the PS3 would have been forced to release in 2007 and thats a huge issue, not to mention the cost of the already manufactured PS3s. Could you really say that the executives at Sony would have held back the PS3 for a further 6-9months to fix a GPU problem and let the Xbox 360 gain an unassailable lead?
3. This is an issue that was always going to creep into the computer industry at some point or another. The problem is that thermals are increasing, not decreasing with time and newer process shrinks are not helping the problem at all. This is all especially true of GPUs which have increased their thermal output over time quite considerably. The GTX 280 uses more power than the Xbox 360 or PS3 as a whole. Its just one chip!
|
1. Are you so sure that those parts were tested as rigorously as console components are tested? The articles I saw made it sound like the laptop manufacturers just trusted that NVidia would deliver to spec and then it came up with a hotter GPU and fudged the numbers to hide it.
In any case "both GPU problems" is an oversimplification. In the laptops the GPU was supposed to run at a certain temp range and it came in over that range, unless I misunderstand, and it began to fry itself. In the RRoD case, from what I can tell, no one has said that the GPU was delivered from ATI in any way out of spec; but the heatsink was totally inadequate for the job, so the GPU was actually warping the motherboard and that's why the solder broke.
"Life expectancy is all over the map because the design has very little margin for most of the important parameters. That means it's not a fault tolerant design."
They designed the system to within the specifications given by Nvidia. The laptops were within the specifications but the parts themselves and specifically the packaging weren't themselves up to the specifications given by Nvidia. They were tested at all stages of the manufacturing process. We are only dealing specifically with the RROD issue caused by the GPU but even in spite of that there were multiple points of failure. The Xbox 360 without a GPU problem would have still had a failure rate which was too high. Theres your systematic failure. The heatsink uses was likely inside the spec given by ATI - GPUs can run extremely hot; The Rv770 has a die temperature of 80 degrees Celsius 176F. Like with the Laptop GPUS the packaging was not as tough as the GPU itself.
2. http://kotaku.com/gaming/sony/official-sony-re+confirms-spring-ps3-launch-155800.php
http://news.bbc.co.uk/2/hi/business/4807858.stm
"Technical hitches related to the console's Blu-ray disc drive had forced the delay, Sony said."
OH SNAP! Sony DID delay 6-9 months due to a hardware issue! If the GPU was faulty it could have been fixed in the same timeframe; fault detection doesn't have to be consecutive instead of concurrent you know.
I knew about the delay, but would they have delayed a further 6-9 months if given pressure by the top executives. Theres a point where they would have to release anyway.
3. This is an issue that responsible engineers would not allow to get to the point of chip destruction.
You can come up with all the comparisons you want, but the fact is that hardware failure on this epic level has not been seen for decades if EVER in the game industry. Moreover, you have completely FAILED to either give sources on your claim of RRoD-comparable NVidia failure rates or admit you don't have any reputable ones; etc. etc.
I'm perfectly willing to admit that all companies are at some small risk for some kind of stealth defect. But the RRoD was not such a beast. "MS was so focused on beating Sony this cycle that the 360 was rushed to market when all indications were that it had serious flaws." It was totally preventable and would have been had MS quality control not been incompetent and/or hogtied. Again, you must either agree with this analysis or call my interview source a liar. Which is it, Squilliam?
I know that the failure rate exists, its hard enough to find the failure rate on Xbox 360s actually its practically impossible to get recent and accurate numbers. The problem is a recent problem with Nvidia so its impossible to quantify. Read above to answer the question.
|
