You can't trademark numbers. AMD, Cyrix, NexGen, and several others had released souped-up 486es and labelled them as "586". Intel didn't like this, so they named their "586" as Pentium, trademarked it, then refused to license the trademark to their competitors. Thus, you could buy an AMD 586, IBM 586, Cyrix 586, NexGen 5x86, or Intel Pentium.
Later, when Intel licensed the Pentium bus or chipset or whatever to AMD and company and they started to produce Socket 7 compatible CPUs, things got even more confusing. Cyrix had the 6x86, AMD had the 586 and K6, NexGen had the 5x86 and the K5, Intel had the Pentium, IBM had the 586 and some other chip. WinChip had the C6. Gone were the days of everything being named the same. No longer could you say "I'm running a 486" and not care about who made it.
You are a virtual organisation with telecommuters, yet you don't have the infrasctucture in place to support them?
Why are you relying on an ISP for mail services? Why are you using existing IM networks? You should be running these yourself. Get a domain. Run a DNS server. Run an SMTP/IMAP server. Manage it all in-house. Install a messaging server and keep your business IM off the public networks.
E-mail isn't broken. Your e-mail configuration/setup/infrastructure is.
Don't blame the tools when the house collapses. Blame the carpenters who put up the framing.
You can be as innacurate as you want with any post relating to Linux or Windows. When it comes to FreeBSD, though, it better be 100% correct.:) Gotta keep the balance, you know.:D
You've never heard of drive imaging software, have you?
Install the OS. Install the patches and service packs. Install the apps you need. Install the patches and service packs. Image the drive to CD, DVD, HD, network, wherever.
Next time you need to do a re-install, just backup your data, wipe the drive, and re-install from the image. Voila. You have a working system with all your apps all ready to go from the start.
Why do repetitious, pointless work? You're suypposed to be a geek -- that means lazy.
There's no equivalent to su, but there is an equivalent to sudo... the RunAs Service, available in 32-bit Windows since Windows 2000.
Enable it via Administrative Tools > Services. Then you can run any program on the system as a different user just by creating a shortcut to it and editing the properties of that shortcut. Select "Run as different user" and a dialog will pop up asking for the user and password to run the program as.
I use this on my Win2K Pro system at home. I have an admin account and a power user account. Everyone else in the house gets normal user. I use the power user account for everything, and use RunAs to install software. The admin account never gets logged into.
This can also be done in WindowsXP and I'd hazard a guess that it's also available in Windows 2003 Server.
You've got your terms all confused. Monolithic kernels can be modular (like the Linux or BSD kernels). Once a module is loaded, it is actually part of the kernel, running in kernel space. There is no difference between a Linux kernel with all the drivers compiled into it, and a Linux kernel that loads all the drivers via modules.
A microkernel has a teeny tiny kernel, then all the drivers are loaded as separate processes, all the programs are loaded as separate processes, and the kernel just passes messages around between those processes. Very different from a monolithic kernel. Nothing runs in kernel space except for the tiny little kernel itself.
My first desktop (IBM something or other) was a 486/100 running Windows 95. Suspend and sleep worked great.
My first laptop (Fujitsu Lifebook 765 Dx) was a P166 running Windows 95 OSR2.1. Suspend, sleep, and hibernate (suspend to disk) worked flawlessly. It's now running DragonflyBSD; suspend and sleep still work.
My work systems running Windows 98, 2000, and XP all suspend and sleep without problems (Seanix, Compaq, and Dell systems). My laptop running FreeBSD 5.x has working suspend and sleep as well.
You have to look beyond MHz and to overall CPU performance. Sure, the Pentium-M is only around 1.7 GHz right now. But that translates into a *lot* better performance than a Pentium4 at 2.0 GHz.
Intel has finally realised they can't compete on pure MHz anymore. Look at what that got them: a super-fast Pentium4 that requires more electricity and generates more heat than the electrical heating system in my apartment.
The Pentium-M architecture won't scale into the 10's of GHz. But the performance will be much better than a Pentium4 that does scale into the 10's of GHz.
Which would you rather have: a super-high GHz CPU that maybe does 3 or 4 instructions per cycle, or a slower, less power-hungry CPU that can do 6 or 8 instructions per cycle? Do the math, and the low-speed, high IPC system will (always) beat the high-speed, low IPC system.
Windows didn't skip version 4. Windows 95 has the internal version number 4, as does Windows 98 and ME. There's also Windows NT 4.
The DOS 4 that was released was horrible. The DOS 4 that didn't was simply amazing (multi-tasking and everything). Only a few European customers got to see that one.:(
And the Pentium4 isn't the fourth iteration of anything. It's nothing more than a bizarre marketing moniker. Especially considering the Pentium-II was actually the third CPU to have the Pentium name (fourth if you include the Pentium/MMX).
We run a modified version of LTSP 2.x on dual-proc P3 1 GHz systems with 3 or 4 GB RAM. We have no problem running 30 clients with OpenOffice, Mozilla, QCad, and Cycas running. However, we also don't run pure-thin client setups.
Instead, we have our client stations (P 166 through P3 450) configured to run as diskless stations (mount everything via NFS) and run as many applications as possible using the local CPU/RAM. If the client can't handle a certain piece of software, then only that piece of software is run on the server as a remote X app.
So far, this setup has allowed us to run labs of 30 without problems (low load and RAM/CPU usage on server) and two schools are running 60 clients without taxing the server too much. All using bog-standard X11R6 protocol. We also run TightVNC for remote administration of the clients, as it's a little faster than pure-X over switched-56K and ADSL connections.
Have a look at NX from NoMachine. Haven't had a chance to use it yet, but reading through the available docs, it works better than VNC/RDP/ICA, using the standard X protocol. It just compresses things, combines commands, and caches a lot of regularly-used info/commands on the client computer. It can also be used with VNC/RDP to make them run even faster.
Their goal, and supposedly they've achieved it, it to be able to use standard X apps across a 9600 bps modem.
They've also got a tesdrive server you can connect to to try things out.
Geo Thermal isn't all that expensive up here in Canada. We have an entire subdivision with over 200 houses (plus golf course and restaurant) completely heated via geo thermal here in Kamloops, BC, Canada. Additional price for house, less than $2000 CDN.
There are also plans to incorporate geo thermal into other subdivisions around town, although I don't think any have gone beyond the planning stage just yet.
You're probably thinking of Theo deRaadt's posting to deadly.org (now undeadly.org) about modifying OpenBSD to use the hardware crypto instructions in the VIA C3.
That's what I never liked about D&D, AD&Dv1 and AD&Dv2... it was all hack & slash. Hell, you got more experience for finding treasure than you did for killing demi-gods. There was no skill involved. You just gave yourself the biggest, baddest weapons, armour, and spells, and killed anything that moved. Sure, it was an okay way to spend an evening... but it wasn't really fun.
I always preferred the RoleMaster systems by Iron Crown Entertainment. These were systems where you actually had skillsets, where you could develop a character you actually cared about, where you could actually go on adventures where the point was not hack & slash. There was an entire book devoted to fighting with tables of damages of almost 50 weapons against various types of armour, with extremely detailed critical hit tables (which we always embellished even more). These systems let your imagination soar such that you actually tried to keep your character alive and not kill everything just for the hell of it.
The White Wolf and Palladium gaming systems were also like this. There was more detail involved, allowing the players to immerse themselves in the game. Yes, you could also run a pure hack & slash game with these (we had one DM who always ran the games like AD&D games), but you didn't have to. It was too much work to try and get an AD&D game to be more than just hack & slash.
That's the main problem with video RPGs: they're just hack & slash.
You're forgetting something very crucial here... the Athlon64 is clocked almost 1 GHz slower than the P4... yet the performance difference is virtually nil. That says a lot more about the performance of the Athlon64 than anything.
That's not a "ho-hum" benchmark to me. That's an "Intel has royally fubar'd themselves. Here's hoping their Pentium-M strategy brings them back on track."
It has to do with speed and re-usability, mainly. Sure, if you go into the Advanced tab, or Options area, you can get a GUI program to do just about everything you could do at the CLI, but it takes time to flip through all those tabs, click all those checkboxes, and fill in all those text boxes. In the time it takes you to read the available options, a CLI user has already gotten the results and moved on.
With history features and auto-completion, using a CLI is much faster... for those who know how to use it.
I much prefer small utilities that do one thing, and one thing only, but do it well. String a bunch of those utilities together, and you get one powerful little toolbox. Much easier (for me) than trying to fire up some huge monolith of a GUI and flip through pages of options trying to get it to do what I want.
The CLI is not for everyone, though. Neither is a GUI. Limiting a user to one or the other is just that... limiting. Give the user the option of both, with CLI and GUI tools that can do everything, and you'll have very happy users. (This is why I like having config info in text files -- GUI tools can be used to edit them, and any old CLI text manipulation tools can as well, if the GUI does down for any reason.)
Slashdot Mirror
Uh, SFU is based on Interix, which uses OpenBSD 3.0. Where do you get the reference to Linux??
The last desktop CPU was a 266 MHz CPU with MMX.
The last non-MMX CPU topped out around 200 MHz, but was very hard to find.
Don't know about the server or mobile CPUs.
You can't trademark numbers. AMD, Cyrix, NexGen, and several others had released souped-up 486es and labelled them as "586". Intel didn't like this, so they named their "586" as Pentium, trademarked it, then refused to license the trademark to their competitors. Thus, you could buy an AMD 586, IBM 586, Cyrix 586, NexGen 5x86, or Intel Pentium.
Later, when Intel licensed the Pentium bus or chipset or whatever to AMD and company and they started to produce Socket 7 compatible CPUs, things got even more confusing. Cyrix had the 6x86, AMD had the 586 and K6, NexGen had the 5x86 and the K5, Intel had the Pentium, IBM had the 586 and some other chip. WinChip had the C6. Gone were the days of everything being named the same. No longer could you say "I'm running a 486" and not care about who made it.
Matt Dillon put RC2 on the download page.
It's also available via BitTorrent.
You are a virtual organisation with telecommuters, yet you don't have the infrasctucture in place to support them?
Why are you relying on an ISP for mail services? Why are you using existing IM networks? You should be running these yourself. Get a domain. Run a DNS server. Run an SMTP/IMAP server. Manage it all in-house. Install a messaging server and keep your business IM off the public networks.
E-mail isn't broken. Your e-mail configuration/setup/infrastructure is.
Don't blame the tools when the house collapses. Blame the carpenters who put up the framing.
You can be as innacurate as you want with any post relating to Linux or Windows. When it comes to FreeBSD, though, it better be 100% correct. :) Gotta keep the balance, you know. :D
The first line of the second paragraph should read: ...
The inefficiency in the SMP in FreeBSD 4 is due to the fact that
Yeah, it was mostly driven by people wanting to prevent men from having sex, alcohol, drugs, guns, money, fun, and so forth. :)
Hmmm, yeah, you're right. Close enough, though. :)
You've never heard of drive imaging software, have you?
Install the OS. Install the patches and service packs. Install the apps you need. Install the patches and service packs. Image the drive to CD, DVD, HD, network, wherever.
Next time you need to do a re-install, just backup your data, wipe the drive, and re-install from the image. Voila. You have a working system with all your apps all ready to go from the start.
Why do repetitious, pointless work? You're suypposed to be a geek -- that means lazy.
There's no equivalent to su, but there is an equivalent to sudo ... the RunAs Service, available in 32-bit Windows since Windows 2000.
Enable it via Administrative Tools > Services. Then you can run any program on the system as a different user just by creating a shortcut to it and editing the properties of that shortcut. Select "Run as different user" and a dialog will pop up asking for the user and password to run the program as.
I use this on my Win2K Pro system at home. I have an admin account and a power user account. Everyone else in the house gets normal user. I use the power user account for everything, and use RunAs to install software. The admin account never gets logged into.
This can also be done in WindowsXP and I'd hazard a guess that it's also available in Windows 2003 Server.
You've got your terms all confused. Monolithic kernels can be modular (like the Linux or BSD kernels). Once a module is loaded, it is actually part of the kernel, running in kernel space. There is no difference between a Linux kernel with all the drivers compiled into it, and a Linux kernel that loads all the drivers via modules.
A microkernel has a teeny tiny kernel, then all the drivers are loaded as separate processes, all the programs are loaded as separate processes, and the kernel just passes messages around between those processes. Very different from a monolithic kernel. Nothing runs in kernel space except for the tiny little kernel itself.
You've had bad hardware.
My first desktop (IBM something or other) was a 486/100 running Windows 95. Suspend and sleep worked great.
My first laptop (Fujitsu Lifebook 765 Dx) was a P166 running Windows 95 OSR2.1. Suspend, sleep, and hibernate (suspend to disk) worked flawlessly. It's now running DragonflyBSD; suspend and sleep still work.
My work systems running Windows 98, 2000, and XP all suspend and sleep without problems (Seanix, Compaq, and Dell systems). My laptop running FreeBSD 5.x has working suspend and sleep as well.
You have to look beyond MHz and to overall CPU performance. Sure, the Pentium-M is only around 1.7 GHz right now. But that translates into a *lot* better performance than a Pentium4 at 2.0 GHz.
Intel has finally realised they can't compete on pure MHz anymore. Look at what that got them: a super-fast Pentium4 that requires more electricity and generates more heat than the electrical heating system in my apartment.
The Pentium-M architecture won't scale into the 10's of GHz. But the performance will be much better than a Pentium4 that does scale into the 10's of GHz.
Which would you rather have: a super-high GHz CPU that maybe does 3 or 4 instructions per cycle, or a slower, less power-hungry CPU that can do 6 or 8 instructions per cycle? Do the math, and the low-speed, high IPC system will (always) beat the high-speed, low IPC system.
Windows didn't skip version 4. Windows 95 has the internal version number 4, as does Windows 98 and ME. There's also Windows NT 4.
:(
The DOS 4 that was released was horrible. The DOS 4 that didn't was simply amazing (multi-tasking and everything). Only a few European customers got to see that one.
And the Pentium4 isn't the fourth iteration of anything. It's nothing more than a bizarre marketing moniker. Especially considering the Pentium-II was actually the third CPU to have the Pentium name (fourth if you include the Pentium/MMX).
We run a modified version of LTSP 2.x on dual-proc P3 1 GHz systems with 3 or 4 GB RAM. We have no problem running 30 clients with OpenOffice, Mozilla, QCad, and Cycas running. However, we also don't run pure-thin client setups.
Instead, we have our client stations (P 166 through P3 450) configured to run as diskless stations (mount everything via NFS) and run as many applications as possible using the local CPU/RAM. If the client can't handle a certain piece of software, then only that piece of software is run on the server as a remote X app.
So far, this setup has allowed us to run labs of 30 without problems (low load and RAM/CPU usage on server) and two schools are running 60 clients without taxing the server too much. All using bog-standard X11R6 protocol. We also run TightVNC for remote administration of the clients, as it's a little faster than pure-X over switched-56K and ADSL connections.
Have a look at NX from NoMachine. Haven't had a chance to use it yet, but reading through the available docs, it works better than VNC/RDP/ICA, using the standard X protocol. It just compresses things, combines commands, and caches a lot of regularly-used info/commands on the client computer. It can also be used with VNC/RDP to make them run even faster.
Their goal, and supposedly they've achieved it, it to be able to use standard X apps across a 9600 bps modem.
They've also got a tesdrive server you can connect to to try things out.
Geo Thermal isn't all that expensive up here in Canada. We have an entire subdivision with over 200 houses (plus golf course and restaurant) completely heated via geo thermal here in Kamloops, BC, Canada. Additional price for house, less than $2000 CDN.
There are also plans to incorporate geo thermal into other subdivisions around town, although I don't think any have gone beyond the planning stage just yet.
You're probably thinking of Theo deRaadt's posting to deadly.org (now undeadly.org) about modifying OpenBSD to use the hardware crypto instructions in the VIA C3.
That's what I never liked about D&D, AD&Dv1 and AD&Dv2 ... it was all hack & slash. Hell, you got more experience for finding treasure than you did for killing demi-gods. There was no skill involved. You just gave yourself the biggest, baddest weapons, armour, and spells, and killed anything that moved. Sure, it was an okay way to spend an evening ... but it wasn't really fun.
I always preferred the RoleMaster systems by Iron Crown Entertainment. These were systems where you actually had skillsets, where you could develop a character you actually cared about, where you could actually go on adventures where the point was not hack & slash. There was an entire book devoted to fighting with tables of damages of almost 50 weapons against various types of armour, with extremely detailed critical hit tables (which we always embellished even more). These systems let your imagination soar such that you actually tried to keep your character alive and not kill everything just for the hell of it.
The White Wolf and Palladium gaming systems were also like this. There was more detail involved, allowing the players to immerse themselves in the game. Yes, you could also run a pure hack & slash game with these (we had one DM who always ran the games like AD&D games), but you didn't have to. It was too much work to try and get an AD&D game to be more than just hack & slash.
That's the main problem with video RPGs: they're just hack & slash.
Eh, checksum, CRC, close enough. You get the gist of it, right. :)
You lazy fool, using a pre-built application.
:)
Real geeks open the socket directly and parse the raw packets. If you can't perform a CRC in your head, you shouldn't be surfing the 'Net.
If you're not going read the article, or the links, then at least read the entire post. This is not a release notice.
They're just saying that Firefox 1.0, when it is released, will be based on Mozilla 1.7. They aren't saying Firefox 1.0 is available.
Yeesh!
You're forgetting something very crucial here ... the Athlon64 is clocked almost 1 GHz slower than the P4 ... yet the performance difference is virtually nil. That says a lot more about the performance of the Athlon64 than anything.
That's not a "ho-hum" benchmark to me. That's an "Intel has royally fubar'd themselves. Here's hoping their Pentium-M strategy brings them back on track."
It has to do with speed and re-usability, mainly. Sure, if you go into the Advanced tab, or Options area, you can get a GUI program to do just about everything you could do at the CLI, but it takes time to flip through all those tabs, click all those checkboxes, and fill in all those text boxes. In the time it takes you to read the available options, a CLI user has already gotten the results and moved on.
... for those who know how to use it.
... limiting. Give the user the option of both, with CLI and GUI tools that can do everything, and you'll have very happy users. (This is why I like having config info in text files -- GUI tools can be used to edit them, and any old CLI text manipulation tools can as well, if the GUI does down for any reason.)
With history features and auto-completion, using a CLI is much faster
I much prefer small utilities that do one thing, and one thing only, but do it well. String a bunch of those utilities together, and you get one powerful little toolbox. Much easier (for me) than trying to fire up some huge monolith of a GUI and flip through pages of options trying to get it to do what I want.
The CLI is not for everyone, though. Neither is a GUI. Limiting a user to one or the other is just that