Heh, not to mention this quote from the above article:
His 1977 role as Obi-Wan Kenobi introduced him to a new generation of filmgoers and made him financially secure. ``I might never have been heard of again if it hadn't been for `Star Wars','' he said.
But he detested the ``Star Wars'' phenonenom, and the fans that went along with it. He once described the dialogue as ``frightful rubbish'' and said he felt like a ``caged animal'' on the set.
LONDON, England (AP) - Actor Sir Alec Guinness, whose roles in a 66-year career ranged from Hamlet to Obi-Wan Kenobi in ``Star Wars,'' has died, a hospital spokesman said Monday. He was 86.
Guinness became ill at his home near Petersfield, southern England and was taken by ambulance to the King Edward VII Hospital where he died Saturday, said hospital spokeswoman Jenny Masding.
Some newspapers reported that Guinness died from liver cancer, but the hospital would not confirm the cause of death.
...
But with ``The Bridge on the River Kwai'' in 1957 he established that his versatility had nothing to do with disguise. He won an Oscar for his performance as the disciplined, inflexible Col. Nicholson in a World War II Japanese prison camp.
Three years later, he played Nicholson's opposite - the boorish, hard-drinking Scottish Lieut.-Col. Jock Sinclair in ``Tunes of Glory.''
He once described it as his favorite film role - ``perhaps the best thing I've done.''
That last paragraph should excuse him from the Star Wars section, although I'm sure the/. crowd knows him best from those flicks.
You're right on with the technical details, but I'm going to have to say that Cray is still Cray.:) Ironically, Cray's new SV2 is ccNUMA,also. However, NUMA isn't an SGI- or Cray-only thing these days, IBM (Sequent) is also doing it.
As an aside for those who don't care to read whitepapers and didn't already know this: cc = Cache Coherent. The memory interface logic keeps track of what cache has what data cached so the data doesn't get corrupted. NUMA = Non-Uniform Memory Access. There's no one point of access to core memory. Each node or board (or processor, let's just call them system blocks) has its own memory inteface, and if other system blocks need some data from another system block's memory, it must ask for it. This removes bottlenecks to relatively slow (insert-favorite-flavor-of-)DRAM. This allows for a greater theoretical system-bandwidth peak through parallelization of slow storage pipes. Writing 16 bytes, 1 byte at a time will always be 16 times slower than writing 16 bytes to 16 banks of DRAM all at once.
VPro is better than commodity parts? Actually, no. NVidia made SGI's "VPro" graphics chipsets as part of SGI's "we'll drop our lawsuit against you if you sell us some chips really really cheap". The VPro, depending on the model, is either the Geforce or the Geforce2 with a SGI-customized bios and a slight overclock. Very slight.
Yes, there's a decent market for these machines. Given SGI's situation, however (they've restructured every quarter for the past 2 years) and the fact that the (non-embedded) MIPS processor line is a few generations behind similar offerings from IBM and Sun, I've a feeling that many customers with just-fat-enough wallets will take a wait and see on these machines, or just look at similar offerings from more stable companies.
As an aside: "lost energy" has something to do with the power factor in the supply as well. Motors are inductive, ie the peak of the voltage wave leads the peak of the current wave (as viewed in the time domain). Many industrial customers whose loads are extremely inductive (pf of about.85 or less) will either have power company-imposed capacitor banks at their shops/buildings, or are charged a higher fee for the inductive loads. Power companies are in the business of selling Watts, not Volt-Amps.
IBM's fabs in Burlington do contract work for several companies. That's the point of operating a general facility like that: you can give priority to your parent company, while having backup business they can run when their parent company has downtime in part runs. If the Transmeta business went away, there would be more business to take its place. IBM wins either way.
Chilling it to negative temps isn't the answer. Supercooled items, when they come in contact with air, condense. That's why the Kryotech attempted to keep the chip closer to room temperature, which in silicon-world, is a pretty cool temperature anyway.
The Cray 1 was freon cooled, Cray 2 was fluorinert, C90 was air (I think), XMP/YMP were air...
Then the T90s, introduced in ~1994, used a big ol' fluorinert tank. Highend, high density Cray boxen (T90) use this cooling, smaller machines (J90, SV1) are air cooled in temperature controlled environments, and some, like the T3E can be water cooled or air cooled, depending on the configuration.
Power:
Minimum 30W (processor speed and SDRAM size dependent) 5V 12-18A, 12V.5A Integrated power supplies: 3.3V and VCore are generated on-board, power is drawn from host system power supply.
My take on the text I highlighted is that power isn't taken from the pci slot, but from the power distribution wiring harness...
Rambus RDRAM is a spec. There's more than 1 company making Rambus ASIC Cells (RACs) for inclusion into memory access chipsets. There's also a few different fabs making the actual memory parts. I'm pretty sure I wouldn't be making such a grand leap as to say that most/. readers are in the position to buy parts containing Rambus technology rather than to develop using Rambus' technology.
Everybody's favorite hardware source around here appears to be Tom's Hardware... my guess is that his whole experience with RDRAM is using Intel chipsets. Is it quite possible that Intel's way of implementing a RAC wasn't the best way? Much like Cyrix's attempts at x86 domination didn't pan out the way they would have hoped. Was x86 flawed, or was it just Cyrix's implementation of it? In the end, it's just a high-performance (if done right) spec that is going to make RMBS shareholders a lot of money.
I'm not so sure about that equation. Better speakers will only show you the shortcomings of MP3 compression. I believe the advantage to MP3 is the fact that they sound the same on computer speakers, which have decidedly poorer audio quality and response. Usually your PC speakers are the weak link in the Total Audio Quality flowchart (Audio Source == CD). MP3s give you smaller sizes without becoming a weaker link than your PC speakers. The sound is only as good as the weakest component.
Better speakers will give you better output, but unless you're getting better input than a compressed MP3 file, you're wasting your money as the Audio Source (MP3 File) is now the weak link.
Now instead of Windows informally getting application features (like IE, Office integration, etc.), we'll be inundated with large amounts of "Microsoft Windows, Inc. has partnered with Microsoft, Inc. to leverage interoperability technology, blah blah" press releases. Yay.
Short term: It won't be better, just different.
Long term: Can GNU/Linux, or another 'alternative' OS make more gains into the PC/Workstation arena? </AssumeSplitActuallyHappens>
There's reasons for everything, from "because it's cool" to "because I sense a great need for this". On which end of the spectrum is Havenco? I can see the need for more privacy in the days of increasing legislation, but what do you say to those who assume you're doing what you're doing to help the shady side of the world? What are you hoping to accomplish, and due to the nature of your company, can/will you be doing anything to keep your service legitimate without becoming the entity that you appear to be positioning yourself against?
Right. From what I've read about the X-Box, MS isn't concerned about the hardware at all. They'll leave that up to Dell or whichever company wants it. These console boxes lose money upfront, but you've got to figure that over the course of its lifetime the cost per unit tends to get driven down after shipping a few hundred million of them. So if the PSX2 goes out and is cloned, it's possible that the competition that is created could drive prices down a bit more, creating a business barrier for the X-Box.
If it weren't for M$'s market dominance, Visual Basic Scripting wouldn't be this popular, and millions of people around the globe wouldn't be sharing and running open source.vbs files. M$ is bringing the world together by giving us the technology required to share the joy of open source software.
I fear the day when the MP3 variant hits: instead of deleting.mp3 files, it attaches them along with itself and sends them to your entire address book...
We've got 3 quad xeon 550 boxes running RH6.1 here at work being used as very small compute servers. When all 4 processors are at full utilization, the overall efficiency for the sparse-memory (ie the 2MB processor cache doesn't help much) computations (verilog simulations) that are being done is terrible.. something on the order of 63%. Benchmarks we got were something like... 1 job = 100%, 2 jobs = 83%, 3 jobs = 75%, 4 jobs = 63%. (i.e. for 4 jobs, each processor is running at 63% of what the single processor ran the same job at)
Now, of course these are shared-bus (100 MHz) i440GX chipset motherboards with 2GB ram. I really couldn't tell you what impact the kernel has on these numbers, but because the caches are constantly missing, the memory bandwidth is pretty much pegged past 2 processors.
I don't have numbers for our SMP processor Alpha Linux box, but it's only 2 processors. Memory bandwidth (I don't believe) is getting swamped with just 2 processors on that box.
It's frustrating because our 32 processor Origin2k (Crossbar processor/memory interface, not shared-bus) can handle more jobs without the falloff of the linux boxes, but if you're just running 1-2 simulations, the intel processors absolutely smoke the Solaris/Irix boxes. Intel's Shared bus just doesn't scale past a couple of processors.
Herman Miller chairs in conference rooms? Please. My employer has the right formula... make the conference room chairs *really* uncomfortable so as to keep meetings short and sweet, while being nice enough to put the H-M Aeron chairs in all of our offices. Now *that's* the secret of success.
Growing up in central Wisconsin, all I had access to was the Stevens Point Journal. I'd say that over 80% of articles were all pulled off the AP wire, but there was the obligatory local section which talked about high school football scores, and who won the local Ice Show. For communities that are not as expansive as Minneapolis/St. Paul (to use a local example) and do not generate enough local news, that's what newspapers are reduced to... local print versions of national stories. Yet at the same time, these are the communities that are slightly behind the whole internet revolution as well. I feel you can replace newspapers right now with a little work. Heck, I even got my parents back home finding what they wanted to see via various websites rather than being dependent upon print versions.:) So until VCRs and TVs are fully web-enabled, I can't see how newspapers, in their present forms, will go away. They're just too convenient, too portable, and after 8 hours of staring at a 21" monitor at work, a nice change of pace.
Slashdot Mirror
Index of /hotmail/users/l/littlejohnny/ w indows98_machine.txt _ not_spam.txt
Updated: Thu Mar 9 05:00:05 2000
From-Mom--Subject-Hotmail_is_pretty_cool.txt
From-Dad--Subject-mom_wont_quit_using_the_damn_
From-YourFriend--Subject-Get_Rich_Quick_this_is
I found a page on sgi.com giving a bit on the difference between VPro/Intel and VPro/Irix.
As an aside for those who don't care to read whitepapers and didn't already know this:
cc = Cache Coherent. The memory interface logic keeps track of what cache has what data cached so the data doesn't get corrupted.
NUMA = Non-Uniform Memory Access. There's no one point of access to core memory. Each node or board (or processor, let's just call them system blocks) has its own memory inteface, and if other system blocks need some data from another system block's memory, it must ask for it. This removes bottlenecks to relatively slow (insert-favorite-flavor-of-)DRAM. This allows for a greater theoretical system-bandwidth peak through parallelization of slow storage pipes. Writing 16 bytes, 1 byte at a time will always be 16 times slower than writing 16 bytes to 16 banks of DRAM all at once.
VPro is better than commodity parts? Actually, no. NVidia made SGI's "VPro" graphics chipsets as part of SGI's "we'll drop our lawsuit against you if you sell us some chips really really cheap". The VPro, depending on the model, is either the Geforce or the Geforce2 with a SGI-customized bios and a slight overclock. Very slight.
Yes, there's a decent market for these machines. Given SGI's situation, however (they've restructured every quarter for the past 2 years) and the fact that the (non-embedded) MIPS processor line is a few generations behind similar offerings from IBM and Sun, I've a feeling that many customers with just-fat-enough wallets will take a wait and see on these machines, or just look at similar offerings from more stable companies.
As an aside: "lost energy" has something to do with the power factor in the supply as well. Motors are inductive, ie the peak of the voltage wave leads the peak of the current wave (as viewed in the time domain). Many industrial customers whose loads are extremely inductive (pf of about .85 or less) will either have power company-imposed capacitor banks at their shops/buildings, or are charged a higher fee for the inductive loads. Power companies are in the business of selling Watts, not Volt-Amps.
IBM's fabs in Burlington do contract work for several companies. That's the point of operating a general facility like that: you can give priority to your parent company, while having backup business they can run when their parent company has downtime in part runs. If the Transmeta business went away, there would be more business to take its place. IBM wins either way.
Chilling it to negative temps isn't the answer. Supercooled items, when they come in contact with air, condense. That's why the Kryotech attempted to keep the chip closer to room temperature, which in silicon-world, is a pretty cool temperature anyway.
Ah. I wasn't sure what the liquid cooled T3E used so I asked one of the guys who used to be in STCO, and he thought water. Doh! Oh well.
The Cray 1 was freon cooled, Cray 2 was fluorinert, C90 was air (I think), XMP/YMP were air...
Then the T90s, introduced in ~1994, used a big ol' fluorinert tank. Highend, high density Cray boxen (T90) use this cooling, smaller machines (J90, SV1) are air cooled in temperature controlled environments, and some, like the T3E can be water cooled or air cooled, depending on the configuration.
Everybody's favorite hardware source around here appears to be Tom's Hardware... my guess is that his whole experience with RDRAM is using Intel chipsets. Is it quite possible that Intel's way of implementing a RAC wasn't the best way? Much like Cyrix's attempts at x86 domination didn't pan out the way they would have hoped. Was x86 flawed, or was it just Cyrix's implementation of it? In the end, it's just a high-performance (if done right) spec that is going to make RMBS shareholders a lot of money.
Better speakers will give you better output, but unless you're getting better input than a compressed MP3 file, you're wasting your money as the Audio Source (MP3 File) is now the weak link.
Q: So, can you wipe your ass?
Mr. Cooper: Are you talking about is it possible to wipe his ass, or if he's ever done it?
Q: Yes, is it possible to wipe his ass
Mr. Cooper: Ambiguous. Please reread the question
(reporter rereads the question)
Mr. Cooper: Excluding conversations with your counsel
A: I'm not sure.
Q: You're not sure?
Mr. Cooper: Please re-read the questions (reporter rereads the question)
A: I guess I don't know.
Q: You guess? You can't say?
(confidential)
A: I don't know.
Now instead of Windows informally getting application features (like IE, Office integration, etc.), we'll be inundated with large amounts of "Microsoft Windows, Inc. has partnered with Microsoft, Inc. to leverage interoperability technology, blah blah" press releases. Yay.
Short term: It won't be better, just different.
Long term: Can GNU/Linux, or another 'alternative' OS make more gains into the PC/Workstation arena?
</AssumeSplitActuallyHappens>
There's reasons for everything, from "because it's cool" to "because I sense a great need for this". On which end of the spectrum is Havenco?
I can see the need for more privacy in the days of increasing legislation, but what do you say to those who assume you're doing what you're doing to help the shady side of the world? What are you hoping to accomplish, and due to the nature of your company, can/will you be doing anything to keep your service legitimate without becoming the entity that you appear to be positioning yourself against?
But if Sony is the sole supplier of the chips, and suddenly there's a 3x rush towards those chips...
Seems to me that there could be a much greater bottleneck, unless other companies are allowed to fab the chips as well.
Right. From what I've read about the X-Box, MS isn't concerned about the hardware at all. They'll leave that up to Dell or whichever company wants it.
These console boxes lose money upfront, but you've got to figure that over the course of its lifetime the cost per unit tends to get driven down after shipping a few hundred million of them.
So if the PSX2 goes out and is cloned, it's possible that the competition that is created could drive prices down a bit more, creating a business barrier for the X-Box.
If it weren't for M$'s market dominance, Visual Basic Scripting wouldn't be this popular, and millions of people around the globe wouldn't be sharing and running open source .vbs files. M$ is bringing the world together by giving us the technology required to share the joy of open source software.
.mp3 files, it attaches them along with itself and sends them to your entire address book...
I fear the day when the MP3 variant hits: instead of deleting
We've got 3 quad xeon 550 boxes running RH6.1 here at work being used as very small compute servers. When all 4 processors are at full utilization, the overall efficiency for the sparse-memory (ie the 2MB processor cache doesn't help much) computations (verilog simulations) that are being done is terrible.. something on the order of 63%. Benchmarks we got were something like... 1 job = 100%, 2 jobs = 83%, 3 jobs = 75%, 4 jobs = 63%. (i.e. for 4 jobs, each processor is running at 63% of what the single processor ran the same job at)
Now, of course these are shared-bus (100 MHz) i440GX chipset motherboards with 2GB ram. I really couldn't tell you what impact the kernel has on these numbers, but because the caches are constantly missing, the memory bandwidth is pretty much pegged past 2 processors.
I don't have numbers for our SMP processor Alpha Linux box, but it's only 2 processors. Memory bandwidth (I don't believe) is getting swamped with just 2 processors on that box.
It's frustrating because our 32 processor Origin2k (Crossbar processor/memory interface, not shared-bus) can handle more jobs without the falloff of the linux boxes, but if you're just running 1-2 simulations, the intel processors absolutely smoke the Solaris/Irix boxes. Intel's Shared bus just doesn't scale past a couple of processors.
Herman Miller chairs in conference rooms? Please. My employer has the right formula... make the conference room chairs *really* uncomfortable so as to keep meetings short and sweet, while being nice enough to put the H-M Aeron chairs in all of our offices. Now *that's* the secret of success.
If you had the right combination of $$, you could see if the Chippewa Falls technology museum would sell theirs. =o)
Oh, and speaking as a Dec `99 Madtown graduate.. Go Badgers!
Growing up in central Wisconsin, all I had access to was the Stevens Point Journal. I'd say that over 80% of articles were all pulled off the AP wire, but there was the obligatory local section which talked about high school football scores, and who won the local Ice Show. :)
For communities that are not as expansive as Minneapolis/St. Paul (to use a local example) and do not generate enough local news, that's what newspapers are reduced to... local print versions of national stories. Yet at the same time, these are the communities that are slightly behind the whole internet revolution as well.
I feel you can replace newspapers right now with a little work. Heck, I even got my parents back home finding what they wanted to see via various websites rather than being dependent upon print versions.
So until VCRs and TVs are fully web-enabled, I can't see how newspapers, in their present forms, will go away. They're just too convenient, too portable, and after 8 hours of staring at a 21" monitor at work, a nice change of pace.