But with a 130W power consumption for the cpu alone, it'd last about 5 minutes on battery alone. But I guess that wouldn't matter, it'd self destruct in 2 minutes due to all that heat in such a small space.
Wouldn't the "ability to support more than one user simultaneously" account for more than 20% of what the 11/780 could do over an apple 2. Besides, a vax it a lot more fun than a little old apple 2:)
Judging by your post, you already know about the following, but this is for those who have old nubus powermacs and don't like being stuck with macos < X. http://nubus-pmac.sf.net/
Unfortunately the last update of the news on there was January 2002 and the last patch was against 2.4.5. Ignoring that, my experience was that it worked quite well on a 6100/60 (although my iBook kicks it's arse)
The Itanic draws somewhere around 130W, the P4 ~70W and (for the r14k I believe) mips processors draw ~20W. What's the point of some high speed if you draw soo much power that the lights dim with an add operation and the street goes black during a sub operation.
You're thinking too small, the largest "off the shelf" configuration you can get from sgi is 512 processors, if you're willing to "custom order", they'll happily fit together a several thousand cpu configuration (I hope you have enough space for the racks).
Re:faster than anything you have used.
on
SGI launches R16000
·
· Score: 2, Insightful
How about with memory throughput, I believe the cray C90 is still up around 4-5, and it's around 10 years old. What's the point of massive compute power if it has to sit idle most of the time waiting for memory access.
How about we bring in the {mips|flops|whatever) per watt, then we start to get even better. The USIII runs off what, around 25-30W, the p4 draws around 70W. Using this, we have the Sun box beating the P4 quite easily, as we can scale up the number of processors to a higher number before we have problems due to dissipating heat and drawing soo much power that we blow the breakers in the data center.
As you can guess from the above post, I don't like the x86 architecture, ugly_hack(){ ugly_hack(); }. There's something to be said about elegance in the design of a processor.
Just remember, the O2's were great for working with large textures (due to the UMA architecture), but not much else (I don't think they had things like Geometry Engines, etc. which made other SGI's great for high end 3D work).
Dump works well, except if you accidently created a filesystem using reiserfs before you deciede on the backup method...I don't like using tar for backups, it just doesn't work quite as nicely (and takes too many arguments to get a clean backup). Dump works low level enough to make everything soo simple.
But with a 130W power consumption for the cpu alone, it'd last about 5 minutes on battery alone. But I guess that wouldn't matter, it'd self destruct in 2 minutes due to all that heat in such a small space.
You also seem to be forgettting that hp already had their 64 bit pa8x00's.
Wouldn't the "ability to support more than one user simultaneously" account for more than 20% of what the 11/780 could do over an apple 2. Besides, a vax it a lot more fun than a little old apple 2 :)
Then we'll have to wait that long again for them to port it to itanic.
Even though that 7.8TB/s is the bandwidth within the supercomputer, I'd love to see someone just *try* to slashdot that.
I'll be the first to say it. MHz don't mean jack.
Judging by your post, you already know about the following, but this is for those who have old nubus powermacs and don't like being stuck with macos < X.
http://nubus-pmac.sf.net/
Unfortunately the last update of the news on there was January 2002 and the last patch was against 2.4.5. Ignoring that, my experience was that it worked quite well on a 6100/60 (although my iBook kicks it's arse)
The Itanic draws somewhere around 130W, the P4 ~70W and (for the r14k I believe) mips processors draw ~20W. What's the point of some high speed if you draw soo much power that the lights dim with an add operation and the street goes black during a sub operation.
Why make a beowulf out of something that already scales above the size of any beowulf out there.
How about with memory throughput, I believe the cray C90 is still up around 4-5, and it's around 10 years old. What's the point of massive compute power if it has to sit idle most of the time waiting for memory access.
How about we bring in the {mips|flops|whatever) per watt, then we start to get even better. The USIII runs off what, around 25-30W, the p4 draws around 70W. Using this, we have the Sun box beating the P4 quite easily, as we can scale up the number of processors to a higher number before we have problems due to dissipating heat and drawing soo much power that we blow the breakers in the data center.
As you can guess from the above post, I don't like the x86 architecture, ugly_hack(){ ugly_hack(); }. There's something to be said about elegance in the design of a processor.
Shut up, you're giving them ideas.
s/32/31/ Damn off by one errors
*cough* 10.*.*.*, 172.16-32.*.*, 192.168.*.* are the private network spaces
Assuming you didn't drown first...
Nah, if only they'd get it working on the sparc port, then I'd have a reason to go out and find another SM51 processor for my SS10...
Then again, it is a PI 4D/20, wouldn't expect it to have something more recent than that on it (12.5MHz of R3000A goodness).
Then there's the Indy running 5.3, but I don't even have any boxes with 6.? on them yet (damn it, I want my octane).
Nah, we got in early. 2.4.11 was THE broken release. It was released and couldn't even be compiled, a few hours later Linus released 2.4.12.
Just remember, the O2's were great for working with large textures (due to the UMA architecture), but not much else (I don't think they had things like Geometry Engines, etc. which made other SGI's great for high end 3D work).
Don't you realise why it's called Duke Nukem Forever? It's going to take them FOREVER to finish it.
Dump works well, except if you accidently created a filesystem using reiserfs before you deciede on the backup method...I don't like using tar for backups, it just doesn't work quite as nicely (and takes too many arguments to get a clean backup). Dump works low level enough to make everything soo simple.
Why does it sound like we're going the way of running Linux/390 under hercules on Linux/390 on hercules on some other linux supported platform?