I would recommend @Home instead. I am currently using Rogers@Home in Toronto and it works great. I regularly get speeds of over 300 kb/s (that's kilo*bytes* not bits), it's always fast and fairly reliable. A few of my friends also use @Home and have the same results. On the other hand, I heard nothing but complains from people who use Sympatico. First of all the PPPoE protocol is a piece of shit. Windows clients are flaky and it's a pain in the ass to get it to work on Linux. Then there is a speed issue. Despite all the garbage they spit out in their commercials, Sympatico DSL is much slower than cable. And reliability sucks. One guy's connection was cut off for 2 days until the repair guy from Bell showed up. Sure, it may be better in some areas, but I heard too many complains about the quality to even try it now.
Besides, the little-known fact is that you can actually set up Rogers cable with a static IP. Even though they do use DHCP, they apparently never expire your IP address, so you can basically get a static IP! That's nice if you are running www/mail/ftp/whatever server.
I was wondering the same thing myself, so I checked the spec benchmarks. Turns out that Sun UltraSPARC II is about the same speed on SPECint as Intel at the same clock speed (even about 10% slower). However, it is about 80% faster on SPECfp. Too bad it is stuck at 450MHz while both Intel and AMD already have 1GHz+ (ehh, ok ok one of them does;-). Check out spec.org for more details.
UltraSPARC III looks like a major improvement. We'll see how it stacks up against competition. My guess is at 900MHz it should kick major ass.
The reason why memory bandwidth is not a a problem right now is precisely because of caching.
Only very few applications (like large databases) will get a significant speed boost from greater memory bandwidth. 99% of other applications (even games) will get only a marginal increase.
We are NOT, repeat NOT limited by memory bandwidth. This is why Rambus performs so crappy. Rambus has twice the bandwidth of 100MHz SDRAM but that does not mean computers with Rambus have twice the performance of the computers with SDRAM. In fact, in some cases computers with 800MHz Rambus are *slower* then the ones with 100MHz SDRAM because Rambus has higher latency.
DDR does not have latency problem. It is basically the same as current SDR, but has double the bandwidth (hence Double Data Rate). So, computers with DDR will get a slight speed increase without incurring latency penalty. DDR machines will be faster than both SDR and Rambus, but still *nowhere near* twice as fast! ___
It's not nearly impossible. It's totally impossible. The guy who wrote it has no clue. Black holes suck in everything, including light waves. That's why they are called black holes. You can only observe black hole's effect on other objects. That way you can figure out that there must be a black hole of certain mass at some location. Repeat: you cannot see black holes. ___
I tried M17 and it was really fast for me. The windows version at least. Now, I'm not running on p166. I've got p3-600 with 128MB RAM. But still, it's as fast if not faster than IE. The Linux version is a different story... But perhaps that's because they have all the debug code compiled in? ___
There is nothing an average user would need this kind of CPU for. Not even games. Currently the 3d games are NOT CPU-limited. They are video card limited. Just check out the benchmarks. At 32 bit color and resolution of 1024x768 and above, the CPU becomes irrelevant. Even a Celeron is able to completely saturate the video card with data. Of course when faster video cards come out, then the CPUs will need to get a little faster too, but nowhere near 2GHz. Now, considering that Microsoft is falling behind in the bloatware department (i.e. you don't need the latest CPU just to run the newest version of Windoze and Word any more), there is nothing else that would require this kind of power.
What I would like to see is faster buses (regular PCI is too slow for say a nice SCSI controller), better HD interface (IDE sucks), and (gasp!) DDR memory. But that is not Intel's agenda. ___
But try to explain that to the ignorant masses. I'm willing to bet that 90% of all the Joe Shmoes who are buying a computer this year will care about the 1GHz thingy. It'll speed up the internet. It'll fix my fridge. It'll make my runaway cat come back. etc.
While I agree with most of your post, I disagree about Pentium Pro being a success. It wasn't. It was so expensive that it never made it to the masses. And it had almost the same performace as plain pentium. It had only 2 things that pentium didn't:
1. Non-castrated motherboards. As you may or may not know, Intel had to limit the amount of memory VX and TX boards can cache to 64Mb, just to promote ppro.
2. 4-way SMP. Not that too many people used it.
Back then Ppro was about the equivalent of today's Xeon. In fact Xeon is a direct successor of ppro. Just like it's older brother, it offers no performance gains over P3 and costs an arm and a leg. Oh, and not all Xeons can even do 4-way SMP. ___
Good point, but remember that this modern day processor costs far less than the Cisco stuff. Since you are not paying for the Cisco label, you can actually spend more money on the hardware. If we compare a Cisco router to an equivalently priced x86 Linux box, I have no doubt that Linux would beat Cisco to the punch. ___
Now this is a definite conflict. How do you go about atomic operations in a fully-preemptive kernel? Does atomicity even exist in real time OSes? If not, is that why all real time apps work only on uniprocessor? Actually, the lack of automicity would preclude proper thread synchronization, meaning that you would not be able to write any multi-threaded applications, uniprocessor or not. Am I missing something? ___
Yeah, like the other poster said, try Debian. Its package management system is simply amazing. No other Linux distribution (well, except the ones based on Debian;-) come even close. Heck, no other OS has anything like it, period!
Imagine a system, similar to FreeBSD's ports, that:
1. Works with binaries. You can download and install binaries instead of having to wait for it to compile. You can also request to get source though, and compile it (just like in FreeBSD).
2. Figures out dependencies auto-magically and downloads all the required libraries. For example, if you say "apt-get install ssh", apt will figure out that ssh depends on libssl, libwrap, libz, etc. It will also figure out that ssh *conflicts* with ssh2, ssh-nonfree, ssh-socks. That is, apt will determine not only dependencies but also conflicts. So, (unless you force it), it will not let you install sendmail, qmail, smail, and exim at the same time.
3. *Handles upgrades*. "apt-get upgrade" will upgrade the packages you have installed. And no, you don't need to do deinstall and reinstall everything. Apt is smarter than that. I guess it's called apt for a reason;-)
(btw, could any Debian developer tell me what apt really stands for?)
Most professional game programmers have been happy with this approach for a long time now, and adopted DirectX years ago.
Tell that to John Carmack. Quake 1/2/3 uses OpenGL. And he personally said that DirectX sucks. Come to think of it, Unreal uses OpenGL too. And so do all the games based on the Quake or Unreal engine.
I don't know what the hell you are talking about. It took me about 2 minutes to set up my workstation to print to HP LaserJet something or other that has the JetDirect card in it. Just follow the HOWTO that is available at www.linuxdoc.org. You can use either a remote lpd queue (which is what I did) or print over smb. The whole setup was a piece of cake. Now, printer drivers are a different story altogether. The printer in the office happened to support postscript (most high-end lasers do), so there were no compatibility issues. However, most of the lowly inkjets and el cheapo lasers don't speak postscript and drivers for them are not always available. That is something that needs to be addressed.
___
Hope some astrophysicist will answer these (clueless?) questions.
It is my understanding that a black hole is an enourmous amount of matter compressed into some tiny space. It has a gigantic mass, and, due to the small size, its density is effectively infinite (or is it *absolutely* infinite?)
1. How do the black holes form? (i.e. Where do babies come from?;-)
2. Once a black hole is formed, it sucks all nearby matter in. Will it keep growing indefinitely or will it somehow stop at some point?
3. What exactly is meant by the size of the black hole? You can't just go there and measure its size. Heck, you can't even see it. You can only infer its mass by looking at the effects on nearby objects. Right? So what's this talk about small/big black holes?
4. (I totally don't get this). When something is being sucked into a black hole it starts accelerating due to the gravitational pull of the black hole. At some point the speed of the sucking object (did I just invent this term?;-) will approach the speed of light, at which point the time is supposed to slow down. This effectively means that at some point the speed of the object will start to decrease, and, eventually, it will stop moving. Therefore, the object will, in fact, never reach the black hole. Did I just pull this out of my ass or is it at least partially true? ___
I completely disagree with that. Let's look at Windows. Windows 95 was a big step forward since Windows 3.1. But that was the last evolutionary step Windows has made. Ever since then, the changes were cosmetic. Windows 98 is just a repackaged Win95 with IE "integrated". Same with Win98SE and WinME. There have been absolutely NO major changes to the OS, despite the fact that MS is still a big company with thousands of employees.
Now let's look at Office. Is there any difference between Office 2000 and Office 97? (ok, the clippy and incompatible file formats don't count). Is there any difference between Office 97 and Office 95? For that matter, is there any difference between Office 95 and Office 4.2? I've used all of them and, again, the changes are cosmetic. One major improvement Office 95 has is long file names (it was released after Windows 95). But other than that, there are NO major changes between the releases.
Now let's take a look at IE. I haven't seen IE 1.0 but I did see IE 2.0. It was a complete joke. I can only assume IE 1.0 was even worse. IE 3.0 was much much better. IE 4.0 was better still (but more bloated). And IE 5.0 is the best browser currently available (I'm not counting Mozilla as it's not out yet). But they gave away IE for free. Why did it become so much better? Because MS actually had to *compete* against Netscape (well, they also "cut their air supply", but that's a different story altogether;-).
The point I'm trying to make is that the software companies will improve their software only if they are forced to by the competition. If there is no competition, they'll produce such amazing "innovations" as MS Bob. ___
Intel is stumbling blindly with the P4 and the Itanium.. Even joe average knows that the P3 does not give anyone a 'better internet experience', and the difference between a 600MHz Celeron and a 1.3GHz P3 is unnoticeable during day to day tasks..
Actually, I wouldn't go that far. My guess is that average joe is just as excited about the 1GHz thingy as Intel VP of Marketing is;-)
The Itanium will probably be outperformed by whatever the latest Celeron is when it finally gets released
Rumour has it, that this already is happening;-)
Clock-for-clock, i doubt the P3 is any significantly faster than the P2, and i doubt that
clock-for-clock, the P4 will be either.
That depends on which P3 you are talking about. The slotted P3 is identical to P2 except for the SSE instructions and the infamous id number. The socketed P3's have full-speed on-die cache, a la Celeron, which does make them a little faster (SSE and serial number still apply). As for P4, it has been shown that clock-for-clock it will actually be *slower* than P3. It will go up to significantly higher clock speeds but it will need that clock increase just to break even with P3. ___
Having used RedHat's 5.x series (that was my first exposure to Linux), I got so sick of it that I decided to never use it again. Switched to SuSE, and later to Mandrake.
If there is a one-sentence description of Mandrake, it would be this: better RedHat than RedHat. (Or, if you prefer, a RedHat that actually works). ___
I've been running for over 2 years. Debian is one of my favorite ditro's. I consider it to be absolutely the best distribution for a server for exactly the reasons people mentioned above in this thread: 1) long testing cycles, 2) amazing package management. I can describe my experience with Debian in one sentence: it's a pain in the ass to install, but a pleasure to maintain. Note that none of the ideological reasons do it for me. My reasons for preferring Debian are very practical. Also, I don't find Debian very well suited for a desktop/workstation. I'm running Mandrake on mine. ___
First of all, why the hell was it moderated up to 3? This post is absolutely clueless.
What you mention (oh, and it's ctrl alt + / ctrl alt - NOT just alt + / alt -) does not change the resolution. The virtual desktop is still stuck at whatever your highest resolution setting in XF86Config is. Well, I guess technically you are right -- the screen resolution does change, but that it not nearly good enough. Even in XFree 4.0 there is NO way to change the resolution AND the virtual desktop size at the same time. There is NO way to change the color depth on the fly either. ___
Microsoft is Microsoft. Don't forget that chief Windows NT architect is Dave Cutler, who used to work on VMS at Digital. Look at NT now, err I mean Windows 2000.
Slashdot Mirror
I would recommend @Home instead. I am currently using Rogers@Home in Toronto and it works great. I regularly get speeds of over 300 kb/s (that's kilo*bytes* not bits), it's always fast and fairly reliable. A few of my friends also use @Home and have the same results. On the other hand, I heard nothing but complains from people who use Sympatico. First of all the PPPoE protocol is a piece of shit. Windows clients are flaky and it's a pain in the ass to get it to work on Linux. Then there is a speed issue. Despite all the garbage they spit out in their commercials, Sympatico DSL is much slower than cable. And reliability sucks. One guy's connection was cut off for 2 days until the repair guy from Bell showed up. Sure, it may be better in some areas, but I heard too many complains about the quality to even try it now.
Besides, the little-known fact is that you can actually set up Rogers cable with a static IP. Even though they do use DHCP, they apparently never expire your IP address, so you can basically get a static IP! That's nice if you are running www/mail/ftp/whatever server.
___
I was wondering the same thing myself, so I checked the spec benchmarks. Turns out that Sun UltraSPARC II is about the same speed on SPECint as Intel at the same clock speed (even about 10% slower). However, it is about 80% faster on SPECfp. Too bad it is stuck at 450MHz while both Intel and AMD already have 1GHz+ (ehh, ok ok one of them does ;-). Check out spec.org for more details.
UltraSPARC III looks like a major improvement. We'll see how it stacks up against competition. My guess is at 900MHz it should kick major ass.
___
The reason why memory bandwidth is not a a problem right now is precisely because of caching.
Only very few applications (like large databases) will get a significant speed boost from greater memory bandwidth. 99% of other applications (even games) will get only a marginal increase.
___
We are NOT, repeat NOT limited by memory bandwidth. This is why Rambus performs so crappy. Rambus has twice the bandwidth of 100MHz SDRAM but that does not mean computers with Rambus have twice the performance of the computers with SDRAM. In fact, in some cases computers with 800MHz Rambus are *slower* then the ones with 100MHz SDRAM because Rambus has higher latency.
DDR does not have latency problem. It is basically the same as current SDR, but has double the bandwidth (hence Double Data Rate). So, computers with DDR will get a slight speed increase without incurring latency penalty. DDR machines will be faster than both SDR and Rambus, but still *nowhere near* twice as fast!
___
It's not nearly impossible. It's totally impossible. The guy who wrote it has no clue. Black holes suck in everything, including light waves. That's why they are called black holes. You can only observe black hole's effect on other objects. That way you can figure out that there must be a black hole of certain mass at some location. Repeat: you cannot see black holes.
___
I tried M17 and it was really fast for me. The windows version at least. Now, I'm not running on p166. I've got p3-600 with 128MB RAM. But still, it's as fast if not faster than IE. The Linux version is a different story... But perhaps that's because they have all the debug code compiled in?
___
I think only Dell, the Intel's bitch, got it. Never seen it anywhere else.
___
There is nothing an average user would need this kind of CPU for. Not even games. Currently the 3d games are NOT CPU-limited. They are video card limited. Just check out the benchmarks. At 32 bit color and resolution of 1024x768 and above, the CPU becomes irrelevant. Even a Celeron is able to completely saturate the video card with data. Of course when faster video cards come out, then the CPUs will need to get a little faster too, but nowhere near 2GHz. Now, considering that Microsoft is falling behind in the bloatware department (i.e. you don't need the latest CPU just to run the newest version of Windoze and Word any more), there is nothing else that would require this kind of power.
What I would like to see is faster buses (regular PCI is too slow for say a nice SCSI controller), better HD interface (IDE sucks), and (gasp!) DDR memory. But that is not Intel's agenda.
___
Dunno. I use mine as a cup coaster. I got their platinum version right now, but I really liked the gold one. It blends in with the furniture.
;-)
___
But try to explain that to the ignorant masses. I'm willing to bet that 90% of all the Joe Shmoes who are buying a computer this year will care about the 1GHz thingy. It'll speed up the internet. It'll fix my fridge. It'll make my runaway cat come back. etc.
___
While I agree with most of your post, I disagree about Pentium Pro being a success. It wasn't. It was so expensive that it never made it to the masses. And it had almost the same performace as plain pentium. It had only 2 things that pentium didn't:
1. Non-castrated motherboards. As you may or may not know, Intel had to limit the amount of memory VX and TX boards can cache to 64Mb, just to promote ppro.
2. 4-way SMP. Not that too many people used it.
Back then Ppro was about the equivalent of today's Xeon. In fact Xeon is a direct successor of ppro. Just like it's older brother, it offers no performance gains over P3 and costs an arm and a leg. Oh, and not all Xeons can even do 4-way SMP.
___
Good point, but remember that this modern day processor costs far less than the Cisco stuff. Since you are not paying for the Cisco label, you can actually spend more money on the hardware. If we compare a Cisco router to an equivalently priced x86 Linux box, I have no doubt that Linux would beat Cisco to the punch.
___
Now this is a definite conflict. How do you go about atomic operations in a fully-preemptive kernel? Does atomicity even exist in real time OSes? If not, is that why all real time apps work only on uniprocessor? Actually, the lack of automicity would preclude proper thread synchronization, meaning that you would not be able to write any multi-threaded applications, uniprocessor or not. Am I missing something?
___
err, its scratching the itch. That's a very important difference.
;-)
___
Yeah, like the other poster said, try Debian. Its package management system is simply amazing. No other Linux distribution (well, except the ones based on Debian ;-) come even close. Heck, no other OS has anything like it, period!
;-)
Imagine a system, similar to FreeBSD's ports, that:
1. Works with binaries. You can download and install binaries instead of having to wait for it to compile. You can also request to get source though, and compile it (just like in FreeBSD).
2. Figures out dependencies auto-magically and downloads all the required libraries. For example, if you say "apt-get install ssh", apt will figure out that ssh depends on libssl, libwrap, libz, etc. It will also figure out that ssh *conflicts* with ssh2, ssh-nonfree, ssh-socks. That is, apt will determine not only dependencies but also conflicts. So, (unless you force it), it will not let you install sendmail, qmail, smail, and exim at the same time.
3. *Handles upgrades*. "apt-get upgrade" will upgrade the packages you have installed. And no, you don't need to do deinstall and reinstall everything. Apt is smarter than that. I guess it's called apt for a reason
(btw, could any Debian developer tell me what apt really stands for?)
___
Tell that to John Carmack. Quake 1/2/3 uses OpenGL. And he personally said that DirectX sucks. Come to think of it, Unreal uses OpenGL too. And so do all the games based on the Quake or Unreal engine.
___
I don't know what the hell you are talking about. It took me about 2 minutes to set up my workstation to print to HP LaserJet something or other that has the JetDirect card in it. Just follow the HOWTO that is available at www.linuxdoc.org. You can use either a remote lpd queue (which is what I did) or print over smb. The whole setup was a piece of cake. Now, printer drivers are a different story altogether. The printer in the office happened to support postscript (most high-end lasers do), so there were no compatibility issues. However, most of the lowly inkjets and el cheapo lasers don't speak postscript and drivers for them are not always available. That is something that needs to be addressed.
___
Hope some astrophysicist will answer these (clueless?) questions.
;-)
;-) will approach the speed of light, at which point the time is supposed to slow down. This effectively means that at some point the speed of the object will start to decrease, and, eventually, it will stop moving. Therefore, the object will, in fact, never reach the black hole. Did I just pull this out of my ass or is it at least partially true?
It is my understanding that a black hole is an enourmous amount of matter compressed into some tiny space. It has a gigantic mass, and, due to the small size, its density is effectively infinite (or is it *absolutely* infinite?)
1. How do the black holes form? (i.e. Where do babies come from?
2. Once a black hole is formed, it sucks all nearby matter in. Will it keep growing indefinitely or will it somehow stop at some point?
3. What exactly is meant by the size of the black hole? You can't just go there and measure its size. Heck, you can't even see it. You can only infer its mass by looking at the effects on nearby objects. Right? So what's this talk about small/big black holes?
4. (I totally don't get this). When something is being sucked into a black hole it starts accelerating due to the gravitational pull of the black hole. At some point the speed of the sucking object (did I just invent this term?
___
I completely disagree with that. Let's look at Windows. Windows 95 was a big step forward since Windows 3.1. But that was the last evolutionary step Windows has made. Ever since then, the changes were cosmetic. Windows 98 is just a repackaged Win95 with IE "integrated". Same with Win98SE and WinME. There have been absolutely NO major changes to the OS, despite the fact that MS is still a big company with thousands of employees.
;-).
Now let's look at Office. Is there any difference between Office 2000 and Office 97? (ok, the clippy and incompatible file formats don't count). Is there any difference between Office 97 and Office 95? For that matter, is there any difference between Office 95 and Office 4.2? I've used all of them and, again, the changes are cosmetic. One major improvement Office 95 has is long file names (it was released after Windows 95). But other than that, there are NO major changes between the releases.
Now let's take a look at IE. I haven't seen IE 1.0 but I did see IE 2.0. It was a complete joke. I can only assume IE 1.0 was even worse. IE 3.0 was much much better. IE 4.0 was better still (but more bloated). And IE 5.0 is the best browser currently available (I'm not counting Mozilla as it's not out yet). But they gave away IE for free. Why did it become so much better? Because MS actually had to *compete* against Netscape (well, they also "cut their air supply", but that's a different story altogether
The point I'm trying to make is that the software companies will improve their software only if they are forced to by the competition. If there is no competition, they'll produce such amazing "innovations" as MS Bob.
___
Actually, I wouldn't go that far. My guess is that average joe is just as excited about the 1GHz thingy as Intel VP of Marketing is ;-)
The Itanium will probably be outperformed by whatever the latest Celeron is when it finally gets released
Rumour has it, that this already is happening ;-)
Clock-for-clock, i doubt the P3 is any significantly faster than the P2, and i doubt that clock-for-clock, the P4 will be either.
That depends on which P3 you are talking about. The slotted P3 is identical to P2 except for the SSE instructions and the infamous id number. The socketed P3's have full-speed on-die cache, a la Celeron, which does make them a little faster (SSE and serial number still apply). As for P4, it has been shown that clock-for-clock it will actually be *slower* than P3. It will go up to significantly higher clock speeds but it will need that clock increase just to break even with P3.
___
some of us are already in Canada! Bad, Hemos, bad!
___
So does Mandrake.
Having used RedHat's 5.x series (that was my first exposure to Linux), I got so sick of it that I decided to never use it again. Switched to SuSE, and later to Mandrake.
If there is a one-sentence description of Mandrake, it would be this: better RedHat than RedHat. (Or, if you prefer, a RedHat that actually works).
___
I've been running for over 2 years. Debian is one of my favorite ditro's. I consider it to be absolutely the best distribution for a server for exactly the reasons people mentioned above in this thread: 1) long testing cycles, 2) amazing package management. I can describe my experience with Debian in one sentence: it's a pain in the ass to install, but a pleasure to maintain. Note that none of the ideological reasons do it for me. My reasons for preferring Debian are very practical. Also, I don't find Debian very well suited for a desktop/workstation. I'm running Mandrake on mine.
___
First of all, why the hell was it moderated up to 3? This post is absolutely clueless.
What you mention (oh, and it's ctrl alt + / ctrl alt - NOT just alt + / alt -) does not change the resolution. The virtual desktop is still stuck at whatever your highest resolution setting in XF86Config is. Well, I guess technically you are right -- the screen resolution does change, but that it not nearly good enough. Even in XFree 4.0 there is NO way to change the resolution AND the virtual desktop size at the same time. There is NO way to change the color depth on the fly either.
___
Microsoft is Microsoft. Don't forget that chief Windows NT architect is Dave Cutler, who used to work on VMS at Digital. Look at NT now, err I mean Windows 2000.
___