Hell reported to be experiencing record-breaking temperature lows. Predictions show that Hell will freeze over just about when Duke Nukem Forever goes gold.
As much as I'd like to see more commercial competition for MS on x86, it aint gonna happen. OSX/x86, if released, will go the way of Solaris/x86. Solaris/SPARC is much better supported than Solaris/x86, both on the software side and hardware compatibility side.
The only compelling reason to switch might be the GUI. Personally I'm quite happy with what I can do in Linux, with all the choices around. OSX on x86 might find a niche, possibly for office machines - cheap x86 boxes running user friendly software might appeal to many businesses who don't need much more than word processing, spreadsheets, web browsing, and email.
Done it many a time. I actually had a public Descent 3 game going on Linux on... a 486/66MHz laptop with 12MB of RAM, 256-color VGA display, 1GB HDD (out of an old Mac laptop) and a PCMCIA netcard. The funny part its it actually worked, with a 2-player games pings were decent (100) although they spiked to 200-500 with a 6 player game. The people in the game went "WTF?" when I told them what they were playing on:-)
I tried to look at the article but it's already slashdotted. Looks like the site is hosted on one of those spare-laptop servers... *ducks*
The RIAA doesn't usually go after movie swappers. RIAA=music, MPAA=movies.
I say someone frames Osama (IP/DNS spoof) and uses P2P to share a boatload of music and movies on a Linux server, so that the RIAA, MPAA, and SCO all go after him. He won't last 12 hours before getting royally pwned by 3 predatory legal teams!
If I recall correctly, MS-DOS was reverse engineered from IBM's PC-DOS (which legend has it evolved from QDOS - Quick and Dirty Operating System).
If I remember correctly, you are wrong. Didn't MS buy the rights to QDOS, modify it, and re-name it to MS-DOS? And then PC-DOS was a knockoff of MS-DOS, right?
Simple. Equip every pilot with a pair of Joo-Janta 2000 Peril-Sensitive Sunglasses, that turn totally black at the first indication of danger! Joo-Janta 2000 Peril-Sensitive Sunglasses: Another fine product of the Sirius Cybernetics Corporation!
This shouldn't really be modded as funny. It's quite a valid solution. What do you think welders use? They use LCD helmets that go opaque as soon as the torch strikes an arc. That way you can actually see what you're welding, instead of putting on you helmet and having everything go black. Not terribly expensive either. Seems to me this wouldn't be difficult to convert for use by pilots or in the plane's windshield.
I still have a 200MHz Pentium Pro system in service that is now around eight years old serving as DSL/NAT router, database server, and web server.
Well, at work we just replaced a Sun SparcStation 10 which was the only webserver for a university department. ~500MB SCSI hard drive. 30MHz TMS390 Sparc CPU. Continuous uptime since it was switched on, when bought new. It never failed, not even the hard drive, and the fans are pretty clogged with dust.
Sun hardware rocks:-)
At home I use a lot of old stuff, I'm posting this from a 5 year old Sun Ultra 10 @333MHz running Gentoo and kept up to date software-wise (firefox 1.0PR, etc). Also have a 6 year old dual P2-333, a 3 year old P4 1.4, and a <1 year old P4 3.06 laptop for when just pure power is needed. The laptop (most powerful) sees the least use of everything, and the U10 (least powerful) sees the most.
Modern software is just getting insanely bloated, and that's all there is to it.
Let's see... I can take a 100Hp subaru from 1994 and rip out the drivetrain, put in parts from a STi, crank it up to 600 Hp and drive on down the road. What can't I change on this thing?
Your analogy is kind of flawed, because that's roughly equivalent to ripping out the entire guts of a PC - motherboard, CPU, RAM, disks, expansion cards, everything - and building a new one inside the old case.
Were the old 350Mhz celerons considered i686 or only i586? I can't remember, but I think they were all i686. But in the unlikely event they were i586-based, that is why it crashed and burned for you. Too bad. I was hoping to get some impression of how it would run on my old 200 MHz Pentium Pro. Anybody else try on a slower machine like that?
Celerons are all i686 class as are Pentium Pros and Pentium IIs. Pentiums and Pentium-MMXs are i586.
I had Slackware 9.0 running on a P2-233 with 64M RAM a couple years ago and it was reasonably fast, even running Mozilla 1.4. Expect a PPro-200 to be the same or slightly better because the PPro's L2 cache is clocked twice as fast as on the P2. Slack 9.0 is mostly optimized from i386 to i586 depending on the packages, so expect Yoper to be _much_ faster.
I'd say it would be manageable for email, web browsing, and that kind of thing but not much more. It'd make a real nice X terminal if you have some bigger boxes on a 100mbit network.
One of the main reasons I was looking forward to the new GCC versions was faster compile times for C++. Yet it seems that for povray (the only c++ source in the benchmark), the compile times consistently get worse with newer versions of gcc. Does no-one care about people who actually compile kde?
Moore's Law says that the speed of a computer will roughly double every 18 months. I don't see GCC's compile time doubling anywhere near every 18 months, so the compile time is actually improving with newer hardware.
Although Gentoo is indeed fast as well, the main differentiating factor with Gentoo is that you build most of your system from source, which has other benefits (disadvantages) than simply execution speed.
One of the other reasons, portability to other architectures, is what made me start using Gentoo in the first place. Since it's compiled from source, it's not too bad to port to other architectures and maintain packages for all of them. Gentoo runs on x86, x86-64, sparc/sparc64, mips, ppc, and alpha too if I'm not mistaken.
I run Gentoo on sparc64 since it is the only distro that provides recent sparc packages except for Debian, which doesn't mix well with my taste. Redhat, Suse, Mandrake, and Splack (which is Slackware ported to sparc) used to support sparc but seem to have quit, so you can't get any recent packages for them.
I think even deleting system programs is way too far. Suppose you have a user (an eye dee ten tee) who has his only copy of his master's thesis in his home directory... you can imagine what happens next. You can make the argument that he deserved it, but it doesn't justify wiping out his thesis.
If the program instead followed your suggestion (never minding the permission issues - it would have to be run as root) and deleted system software, what would happen if the program was run on a production server? Sure, it really ticks off the user, but a lot of things on the server for all users would grind to a halt until the deleted files are replaced. You just can't justify doing something this drastic.
And besides... if you want real security, just do double encryption. use two common, off-the-shelf encryption methods, one encrypting the other's data and your data is now as safe as it can get. further encryptions in encryptions will only add to bloat and time to decrypt.
That works fine until you try double encrypting with ROT13:-)
BTW, doubling the number of layers of encryption really doesn't make a whole lot of difference: at most it increases the complexity as much as if you had increased a single cipher by one bit. 8 levels of encryption at 128-bit each, you have at most a cipher of strength equivalent to 131 bits.
Can't we just do something like OpenBoot on Sparc does - maybe ide0,0 for pri master, ide0,1 for pri slave, ide1,0 for sec master, and ide1,1 for sec slave? It sure makes sense to me: ide[channel number],[drive number] . It would also make terminology for a third IDE channel a little more obvious.
Vintage benchmarks from vintage boxes... my wife's athlon 600 gets better IO with her single 7200rpm maxtor drive.
The benchmarks are recent (within the last month or so), and yes the boxes are old. An Athlon 600 may be faster, but you're comparing a 600MHz CPU to a pair of 333MHz CPUs and a 200MHz CPU of a different architecture. If I wanted to talk about pure speed, I'd bring out the big guns (P4 3.06GHz with ATA100 drives, etc). I was just making a point that even on old, weak systems software RAID-0 is still pretty nice.
One thing I'd like to add is that the chunk size of the RAIDs is 4kb on the PII and 8kb on the SPARC, much smaller than the 32-256KB the article suggests.
The SPARC should be quite capable of burning a CD even at something like 48X, it has very fast I/O. I just don't have a SCSI drive that fast to test it with. It isn't even measurably loaded burning a CD at 8X.
I use software RAID-0 a lot at home and the performance increase is awesome. All my arrays are with two drives. Thing is, most of the time it's actually to increase the space on a single partition - the performance jump is a bonus. I'd rather set up two 30GB drives I already have in a RAID-0 than go out and buy another 60GB drive which wouldn't perform as well.
I have a dual Pentium-II 333 on a board which supports ATA33 and has two 5400RPM 30GB Maxtor drives. I've seen burst speeds right up at the theoretical max, ~70MB/sec. Continuous is lower - 50MB/sec. Compare to 28MB/sec off one of those drives. They're on seperate IDE controllers of course.
Right now I'm on a Sun Ultra 1-Enterprise @200MHZ with a pair of Seagate Barracuda 4LP drives (7200RPM, fast+wide SCSI, SCA connectors). These drives absolutely scream every way you look at them - they're faster at random access than the Maxtor 30GB drives, have command queuing (sp? sorry grammar nazis), and noisy bearings. A RAID-0 across the two drives sees burst ~25MB/sec, continuous 20MB/sec, under random access 1-5MB/sec. The drives in the PII get 0-2MB/sec under random load. Again, I really built the RAID to get the space, and the performance is extra.
Now about the reliability of a RAID-0. I think RAID-0 is a really bad misnomer, since there is no redundancy in a RAID-0. In fact, while other RAID levels increase the reliability of the entire array compared to a single drive, RAID-0 decreases the reliability. If you have two drives instead of one, and you need every piece of data on each to be stable, the array is twice as likely to fail than a single drive. A 3-disk RAID-0, similarly, would be 1/3 the stability of any of the three drives alone.
I'm not terribly worried about my drives failing - the Maxtor drives haven't been heavily used (and came with a 3 year warranty), and Seagate drives have proven very reliable for me. Besides, I do a complete rsync to another box once in a while too.
I would not by any means use RAID-0 on a production server - this is RAID-5 land for small setups, and RAID 0+1/1+0 for bigger setups.
Bash me about my setups, fine, but that's how I've got things configured and it works for me.
It's responsible for the entirety of the entire game code
You must work for the Department of Redundancy Department.
On a more serious note, that is amazing. You're telling me practically the entire game is written in a scripting language?! And it's no slouch for performance, either - it must be compiled into CPU-native instructions on load (self-modifying code anyone?), otherwise why is it so FAST?
Slashdot Mirror
Hell reported to be experiencing record-breaking temperature lows. Predictions show that Hell will freeze over just about when Duke Nukem Forever goes gold.
As much as I'd like to see more commercial competition for MS on x86, it aint gonna happen. OSX/x86, if released, will go the way of Solaris/x86. Solaris/SPARC is much better supported than Solaris/x86, both on the software side and hardware compatibility side.
The only compelling reason to switch might be the GUI. Personally I'm quite happy with what I can do in Linux, with all the choices around. OSX on x86 might find a niche, possibly for office machines - cheap x86 boxes running user friendly software might appeal to many businesses who don't need much more than word processing, spreadsheets, web browsing, and email.
look at his hit counter! 350K+ views... now I know why slashdotting is so effective.
Done it many a time. I actually had a public Descent 3 game going on Linux on... a 486/66MHz laptop with 12MB of RAM, 256-color VGA display, 1GB HDD (out of an old Mac laptop) and a PCMCIA netcard. The funny part its it actually worked, with a 2-player games pings were decent (100) although they spiked to 200-500 with a 6 player game. The people in the game went "WTF?" when I told them what they were playing on :-)
I tried to look at the article but it's already slashdotted. Looks like the site is hosted on one of those spare-laptop servers... *ducks*
The RIAA doesn't usually go after movie swappers. RIAA=music, MPAA=movies.
I say someone frames Osama (IP/DNS spoof) and uses P2P to share a boatload of music and movies on a Linux server, so that the RIAA, MPAA, and SCO all go after him. He won't last 12 hours before getting royally pwned by 3 predatory legal teams!
If I remember correctly, you are wrong. Didn't MS buy the rights to QDOS, modify it, and re-name it to MS-DOS? And then PC-DOS was a knockoff of MS-DOS, right?
Doom on a PSP? Dude! That would absolutely ROCK! Quick frag-fest on the way to work... yeaaaah!
This shouldn't really be modded as funny. It's quite a valid solution. What do you think welders use? They use LCD helmets that go opaque as soon as the torch strikes an arc. That way you can actually see what you're welding, instead of putting on you helmet and having everything go black. Not terribly expensive either. Seems to me this wouldn't be difficult to convert for use by pilots or in the plane's windshield.
Run Longhorn?
Well, at work we just replaced a Sun SparcStation 10 which was the only webserver for a university department. ~500MB SCSI hard drive. 30MHz TMS390 Sparc CPU. Continuous uptime since it was switched on, when bought new. It never failed, not even the hard drive, and the fans are pretty clogged with dust.
Sun hardware rocks
At home I use a lot of old stuff, I'm posting this from a 5 year old Sun Ultra 10 @333MHz running Gentoo and kept up to date software-wise (firefox 1.0PR, etc). Also have a 6 year old dual P2-333, a 3 year old P4 1.4, and a <1 year old P4 3.06 laptop for when just pure power is needed. The laptop (most powerful) sees the least use of everything, and the U10 (least powerful) sees the most.
Modern software is just getting insanely bloated, and that's all there is to it.
Your analogy is kind of flawed, because that's roughly equivalent to ripping out the entire guts of a PC - motherboard, CPU, RAM, disks, expansion cards, everything - and building a new one inside the old case.
The geek is strong in this one...
Add 802.11b support and an ARM CPU (pun intended) and now you're talkin.
Celerons are all i686 class as are Pentium Pros and Pentium IIs. Pentiums and Pentium-MMXs are i586.
I had Slackware 9.0 running on a P2-233 with 64M RAM a couple years ago and it was reasonably fast, even running Mozilla 1.4. Expect a PPro-200 to be the same or slightly better because the PPro's L2 cache is clocked twice as fast as on the P2. Slack 9.0 is mostly optimized from i386 to i586 depending on the packages, so expect Yoper to be _much_ faster.
I'd say it would be manageable for email, web browsing, and that kind of thing but not much more. It'd make a real nice X terminal if you have some bigger boxes on a 100mbit network.
Moore's Law says that the speed of a computer will roughly double every 18 months. I don't see GCC's compile time doubling anywhere near every 18 months, so the compile time is actually improving with newer hardware.
One of the other reasons, portability to other architectures, is what made me start using Gentoo in the first place. Since it's compiled from source, it's not too bad to port to other architectures and maintain packages for all of them. Gentoo runs on x86, x86-64, sparc/sparc64, mips, ppc, and alpha too if I'm not mistaken.
I run Gentoo on sparc64 since it is the only distro that provides recent sparc packages except for Debian, which doesn't mix well with my taste. Redhat, Suse, Mandrake, and Splack (which is Slackware ported to sparc) used to support sparc but seem to have quit, so you can't get any recent packages for them.
I think even deleting system programs is way too far. Suppose you have a user (an eye dee ten tee) who has his only copy of his master's thesis in his home directory... you can imagine what happens next. You can make the argument that he deserved it, but it doesn't justify wiping out his thesis.
If the program instead followed your suggestion (never minding the permission issues - it would have to be run as root) and deleted system software, what would happen if the program was run on a production server? Sure, it really ticks off the user, but a lot of things on the server for all users would grind to a halt until the deleted files are replaced. You just can't justify doing something this drastic.
That works fine until you try double encrypting with ROT13
BTW, doubling the number of layers of encryption really doesn't make a whole lot of difference: at most it increases the complexity as much as if you had increased a single cipher by one bit. 8 levels of encryption at 128-bit each, you have at most a cipher of strength equivalent to 131 bits.
I find it ironic that the Google context-sensitive ad for this article is about making your website a "revenue generator"...
Can't we just do something like OpenBoot on Sparc does - maybe ide0,0 for pri master, ide0,1 for pri slave, ide1,0 for sec master, and ide1,1 for sec slave? It sure makes sense to me: ide[channel number],[drive number] . It would also make terminology for a third IDE channel a little more obvious.
The benchmarks are recent (within the last month or so), and yes the boxes are old. An Athlon 600 may be faster, but you're comparing a 600MHz CPU to a pair of 333MHz CPUs and a 200MHz CPU of a different architecture. If I wanted to talk about pure speed, I'd bring out the big guns (P4 3.06GHz with ATA100 drives, etc). I was just making a point that even on old, weak systems software RAID-0 is still pretty nice.
One thing I'd like to add is that the chunk size of the RAIDs is 4kb on the PII and 8kb on the SPARC, much smaller than the 32-256KB the article suggests.
The SPARC should be quite capable of burning a CD even at something like 48X, it has very fast I/O. I just don't have a SCSI drive that fast to test it with. It isn't even measurably loaded burning a CD at 8X.
I use software RAID-0 a lot at home and the performance increase is awesome. All my arrays are with two drives. Thing is, most of the time it's actually to increase the space on a single partition - the performance jump is a bonus. I'd rather set up two 30GB drives I already have in a RAID-0 than go out and buy another 60GB drive which wouldn't perform as well.
I have a dual Pentium-II 333 on a board which supports ATA33 and has two 5400RPM 30GB Maxtor drives. I've seen burst speeds right up at the theoretical max, ~70MB/sec. Continuous is lower - 50MB/sec. Compare to 28MB/sec off one of those drives. They're on seperate IDE controllers of course.
Right now I'm on a Sun Ultra 1-Enterprise @200MHZ with a pair of Seagate Barracuda 4LP drives (7200RPM, fast+wide SCSI, SCA connectors). These drives absolutely scream every way you look at them - they're faster at random access than the Maxtor 30GB drives, have command queuing (sp? sorry grammar nazis), and noisy bearings. A RAID-0 across the two drives sees burst ~25MB/sec, continuous 20MB/sec, under random access 1-5MB/sec. The drives in the PII get 0-2MB/sec under random load. Again, I really built the RAID to get the space, and the performance is extra.
Now about the reliability of a RAID-0. I think RAID-0 is a really bad misnomer, since there is no redundancy in a RAID-0. In fact, while other RAID levels increase the reliability of the entire array compared to a single drive, RAID-0 decreases the reliability. If you have two drives instead of one, and you need every piece of data on each to be stable, the array is twice as likely to fail than a single drive. A 3-disk RAID-0, similarly, would be 1/3 the stability of any of the three drives alone.
I'm not terribly worried about my drives failing - the Maxtor drives haven't been heavily used (and came with a 3 year warranty), and Seagate drives have proven very reliable for me. Besides, I do a complete rsync to another box once in a while too.
I would not by any means use RAID-0 on a production server - this is RAID-5 land for small setups, and RAID 0+1/1+0 for bigger setups.
Bash me about my setups, fine, but that's how I've got things configured and it works for me.
You must work for the Department of Redundancy Department.
On a more serious note, that is amazing. You're telling me practically the entire game is written in a scripting language?! And it's no slouch for performance, either - it must be compiled into CPU-native instructions on load (self-modifying code anyone?), otherwise why is it so FAST?