By: philt (ptay1685.delete@this.bigpond.net.au), June 29, 2007 8:09 pm
Room: Moderated Discussions
Andi Kleen (ak-rwt@muc.de) on 6/29/07 wrote:
---------------------------
>philt (ptay1685@bigpond.net.au) on 6/29/07 wrote:
>---------------------------
>>Linus Torvalds (torvalds@osdl.org) on 6/28/07 wrote:
>>---------------------------
>>>philt (ptay1685@bigpond.net.au) on 6/28/07 wrote:
>>>>
>>>>FYI just had a look at the Intel pdf for Core 2 errata.
>>>>Amazed at the number of errata listed.
>>>
>>>Read that as:
>>>"Amazed that Intel actually tells the world, rather than
>>>hiding it under the rug".
>>>
>>
>>Is it really so amazing? The bugs exist in reality, whether they admit them or
>>not.
>
>Depends. A lot are in very contrived circumstances
>that are unlikely to be hit in any real situation
>(usually found then with randomized tests)
>
That word "unlikely" is rather worrying. I prefer "never".
>>Sooner or later the bugs will cause problems for someone, >either software developers
>>or end users,
>
>Unlikely for most of them I would say.
>
>
>>As I said - they were kind of forced to admit the bugs, or completely lose their
>>reputation. Not really something to praise them for. Like naughty children they
>>only came clean after they were "found out".
>
>The point was that most other chip companies are still
>naughty children in your terminology (tried to get
>an errata sheet for the ARM CPU in your cellphone
>recently?)
I think that a company should publicise to its customers any shortcomings of their product that may affect them. It is surely the right thing to do. I just dont think you should pat people on the back for doing the right thing as if it is so wonderful - it is to be expected, in deed it is required.
If someone passes me in the street without pulling out a gun and shooting me, I dont approach them and say "gee thanks for not killing me - your a real pal - let me buy you a drink".
If Intel have done the right thing, thats good, but we shouldnt applaud them for it as if they are doing us a big favour by not lying to us. Thats my point.
>
>>Nothing is perfect, but the IT industry as a whole seems to scrape the bottom of
>>the barrel. I guess this is inevtiable with the pace of change. I for one would
>>be happier if it all got slowed down.
>
>At some point Moore's law will run out of steam
>(as in chips cannot be shrunken anymore; ok
>they might add more layers then but at some
>point that will also run into some bottleneck
>and transistor limits will not raise anymore)
>Just nobody knows when it will be. But likely
>(unless you're already quite old) you'll live
>to see it. Perhaps you'll be happy then.
>
I said slowed down, not stopped. Suppose one day they will be able to design and build new cpus on a weekly basis - would you like that? Pace of change can be both too fast or too slow. I like it when its just right (like in the Goldilocks story). I think a new generation of cpu's almost every year is too fast. But Im not advocating, as you seem to be suggesting, that progress is brought to a complete halt.
>>It would be nice for once to buy something
>>that didnt become outmoded junk three weeks after it was purchased.
>
>As long as it works for you it's not outmodded junk.
>Most of the PCs bought 5 years ago are still perfectly
>usable even with most modern software (yes there
>are exceptions, but it's still generally true)
>
Except when you want to add in some kind of upgrade, like new memory to a motherboard, and find out its no longer available (as with my Pentium 4 m/b that uses RDRAM). Or when you want to move your ultra expensive AGP graphics card over to your new motherboard but find there is no AGP slot.
Sure if you never do an upgrade, then thats ok. But how often is that the real world situation? It never works for me.
>>and then
>>the industry can put efforts into bug eradication rather than continual peformance
>>enhancement, and the things will become truly reliable.
>
>Typically each new stepping fixes a few bugs. But
>if they cannot fix the problem without a larger
>redesign they won't do that because the redesign
>would be likely to add even more bugs. That's a common
>problem in complex systems, unlikely to be solved generally.
>
>>One can only dream.
>
>Your expectations are somewhat unrealistic.
>
So it would seem.
But perhaps if they slowed things down, they might get closer to it. Who knows?
---------------------------
>philt (ptay1685@bigpond.net.au) on 6/29/07 wrote:
>---------------------------
>>Linus Torvalds (torvalds@osdl.org) on 6/28/07 wrote:
>>---------------------------
>>>philt (ptay1685@bigpond.net.au) on 6/28/07 wrote:
>>>>
>>>>FYI just had a look at the Intel pdf for Core 2 errata.
>>>>Amazed at the number of errata listed.
>>>
>>>Read that as:
>>>"Amazed that Intel actually tells the world, rather than
>>>hiding it under the rug".
>>>
>>
>>Is it really so amazing? The bugs exist in reality, whether they admit them or
>>not.
>
>Depends. A lot are in very contrived circumstances
>that are unlikely to be hit in any real situation
>(usually found then with randomized tests)
>
That word "unlikely" is rather worrying. I prefer "never".
>>Sooner or later the bugs will cause problems for someone, >either software developers
>>or end users,
>
>Unlikely for most of them I would say.
>
>
>>As I said - they were kind of forced to admit the bugs, or completely lose their
>>reputation. Not really something to praise them for. Like naughty children they
>>only came clean after they were "found out".
>
>The point was that most other chip companies are still
>naughty children in your terminology (tried to get
>an errata sheet for the ARM CPU in your cellphone
>recently?)
I think that a company should publicise to its customers any shortcomings of their product that may affect them. It is surely the right thing to do. I just dont think you should pat people on the back for doing the right thing as if it is so wonderful - it is to be expected, in deed it is required.
If someone passes me in the street without pulling out a gun and shooting me, I dont approach them and say "gee thanks for not killing me - your a real pal - let me buy you a drink".
If Intel have done the right thing, thats good, but we shouldnt applaud them for it as if they are doing us a big favour by not lying to us. Thats my point.
>
>>Nothing is perfect, but the IT industry as a whole seems to scrape the bottom of
>>the barrel. I guess this is inevtiable with the pace of change. I for one would
>>be happier if it all got slowed down.
>
>At some point Moore's law will run out of steam
>(as in chips cannot be shrunken anymore; ok
>they might add more layers then but at some
>point that will also run into some bottleneck
>and transistor limits will not raise anymore)
>Just nobody knows when it will be. But likely
>(unless you're already quite old) you'll live
>to see it. Perhaps you'll be happy then.
>
I said slowed down, not stopped. Suppose one day they will be able to design and build new cpus on a weekly basis - would you like that? Pace of change can be both too fast or too slow. I like it when its just right (like in the Goldilocks story). I think a new generation of cpu's almost every year is too fast. But Im not advocating, as you seem to be suggesting, that progress is brought to a complete halt.
>>It would be nice for once to buy something
>>that didnt become outmoded junk three weeks after it was purchased.
>
>As long as it works for you it's not outmodded junk.
>Most of the PCs bought 5 years ago are still perfectly
>usable even with most modern software (yes there
>are exceptions, but it's still generally true)
>
Except when you want to add in some kind of upgrade, like new memory to a motherboard, and find out its no longer available (as with my Pentium 4 m/b that uses RDRAM). Or when you want to move your ultra expensive AGP graphics card over to your new motherboard but find there is no AGP slot.
Sure if you never do an upgrade, then thats ok. But how often is that the real world situation? It never works for me.
>>and then
>>the industry can put efforts into bug eradication rather than continual peformance
>>enhancement, and the things will become truly reliable.
>
>Typically each new stepping fixes a few bugs. But
>if they cannot fix the problem without a larger
>redesign they won't do that because the redesign
>would be likely to add even more bugs. That's a common
>problem in complex systems, unlikely to be solved generally.
>
>>One can only dream.
>
>Your expectations are somewhat unrealistic.
>
So it would seem.
But perhaps if they slowed things down, they might get closer to it. Who knows?
