Linux is faster then the win32 port on voodoo 2.
on
Q3T on Mac First
·
· Score: 1
I had a IIfx for a long time. There have been X servers for the mac for a long time (from Apple and others), so an old version should run OK on System 6.08.
You could also install A/UX (Apple 68K Unix) on that machine, which includes an XServer. (But using it precludes using the MacOS interface.) --
Don't forget that 99% of the game market is running Windows 9x and MacOS, which means no SMP support in the OS. Adding SMP support for the 1% running Linux or NT might not have been worth the cost. --
It's actually more similar to the comparison between OS/2 and Win3.1.
The Win16 subsystem in NT (WOW - Windows on Windows) works by mapping Win16 and DOS API calls to the Win32 equivalent. Thus Win16 programs get Preemptive MT and Memory Protection under NT. I think this is similar to how WINE works - by mapping Win* calls to the Unix/X equivs.
MacOS X runs the MacOS subsystem (Blue Box) by virtualizing the hardware and "booting" the authentic MacOS. Thus, no modern features between MacOS programs, although the Blue Box probably couldn't take down the OS. Similar to how OS/2 ran DOS and Win16 programs.
The announced Carbon subsystem (subset of MacOS APIs) will work in a manner similar to WOW, if I understand correctly.
My understanding was that the Blue Box (MacOS subsystem) really wasn't officially supported under OS X Server. One reason for this is that they have not merged the networking yet, which means you need two IP addresses, one each for Blue and Yellow.
It would also be difficult to see how VirtualPC would be able to work under OS/X at all because all of the hardware is virtualized in the Blue Box. (Virtual PC appears to talk directly to the SCSI and Network hardware using emulated drivers.) I think this is why there's been talk of a "Red Box", which would provide emulation services under OS X. --
Following myself up, Allison does have a message for those who take the Mindcraft bench at FUD value alone:
The study has also shown that the knowledge of how to tune filesystem performance in Linux is equally obscure. We need to do a better job of educating Linux administrators about how to get the most out of their systems.
I'm sure there'll be more benchmark disappointments in store for us. After all, how else do we learn what we need to fix? But the strength of open source is that we can face the errors without trying to deny them. We just fix 'em and move on.
I just took a look at the linked article written by Jeremy Allison of Samba.
A few interesting points -
* In the often referred-to ZD Samba versus NT benchmarks (where Linux+Samba wins), the Samba/Linux configuration was tuned by a Samba team member. Objectively, this makes the ZD benchmark actually less valid as the Mindcraft study, because as far as we know, a Microsoft-employeed SMB developer wasn't actually there tuning the server.
* Tuning Linux properly involves cryptic commands such as:
While I'm sure these commands are documented somewhere, this sort of tuning makes the NT Registry Editor look like a model user interface. Low level tuning like this really needs a nicer front end, or preferably, a daemon which monitors system activity and dynamically tunes these settings.
It sounds like the Mindcraft study has been a kick in the pants for the Linux community to get some high performance documentation together. I'd like to see a nice How-To which lays out some of the more obscurantist tricks such as echoing strings to the/proc filesystem. --
If I understand correctly, those numbers are public webservers only. MS IIS's market strength has been internal Intranet solutions (where there's probably an existing NT file+print setup). IIIS's intranet market is probably going to being going up, not down, as things like the Office 2000 server get deployed.
A few years ago, nobody in their right mind would have proposed any x86 (Novell, OS/2, NT, Linux)solution for anything other than workgroup filesharing or a ccMail postoffice. The fact that you can now seriously consider Linux or NT as a contender at the low end of the midrange market is primarily due to the advances in Intel hardware.
In a few years, Solaris, Tru64, HP/UX, Linux, and NT will all be running on essentially the same Intel IA64 hardware. At this point, the appeal of NT's one-size-fits-all design is going to start breaking down. But on the other hand, hardware equality is going to get Microsoft's salesmen in the door for midrange solutions that were previously above their heads. And Microsoft is more price competitive than commercial Unix, so NT deployment is probably going to increase in this market, not decrease. (Same argument for Linux.)
The 'Rush Limbaugh' principle is a very valid point, especially in this context. Don't forget the target market for this study is Microsoft partners and WinNT-based shops.
Aside from all the meaninless numbers (who cares if your web server can saturate a 100BT line with static pages!), the study drives home an important point to NT Administrators - If you've invested in a high end IIS system, and you've got it tuned, there's probably no good reason to switch that box over to Linux. If the Linux box was tuned correctly, I doubt the difference would be that great performance-wise.
Of course, the study didn't address stability, which is the number one problem with IIS. --
A Quad Xeon is the highest end box that NT runs on (barring Alpha).
Note that WinNT is driving the "high-end" x86 hardware market. Vendors like Dell and Compaq make boxes with only 4 CPUs because that's all vanilla NT will support. When Win2000 comes out, it will support 8? processors, which means the hardware companies will immedeatly follow with 8 CPU iron. (Implictly making this hardware available to some Linux folks.)
Of course a better benchmark would be the $50,000 NT/Dell box versus the $50,000 Sun/HP/DEC box, etc. --
Some people actually do run core business services on NT, although not many. And if you cut through the hype, most people don't run their core business services on Linux, *BSD, SCO, or any other x86 platform either.
There's a number of reasons for this other than "NT Sux" - x86 server hardware only has started to approach the price/performance of midrange hardware within the last couple years. Also, early versions of Oracle/NT were not very stable, and MS didn't come out with an enterprise competitive DBMS until a couple months ago (SQL 7). Likewise with Linux - the pieces are just now appearing.
Not to mention that most "core business" applications undergo years of planning and development, and have a self life of 10 years or more. VMS is still alive for this reason, and there's no doubt that a certain part of the Unix/S390/AS400 sales pitch is legacy compatibility. There has not been enough time for most shops to build for NT or Linux or migrate what they've got to newer platforms.
So, now that all the pieces are there, expect x86 to start getting midrange marketshare, but it's going to be slow going for either Linux or NT. (As was mentioned in the NT vs Linux debate, at $50K - $100K price point, there's a few more options than x86+Linux or x86+NT.) And, yes 390 will live forever.
How is this going to be any different than AT+T or other private data networks (other than IPv6)?
It seems the discussion is going like this:
1) Private networks offer security and QOS - let's use them. 2) No, the Internet is cool and cheap - let's use it. 3) Hey - the Internet has low security and QOS - let's use a private network! Aren't we smart! --
While you make valuable points, don't forget that differing Unixes only offer source compatiblity and no binary compatiblity at all. So while it might be nice that you can recompile an OpenWindows program and run it on Linux/Gnome system, that doesn't do an iota for someone whose StarOffice install is failing due to strangely placed files or a library conflict.
Furthermore, while many of the base APIs in Unix are old and stable, the modern, more concerted efforts to create an integrated environment are not. Ten years from now, it may be technically possible to run your legacy Gnome 1.0 application on your Pentium64 SuperPro, but it could involve finding and trying to compile all 30 legacy Gnome libraries. The cost/benefit of porting starts to fall of really quickly.
There's a few more dependancies involved other than just "Unix is Unix". When GNU refused to port gcc to Apple A/UX (System V-based with MacOS GUI), it crippled the platform. All the wonderful Unix source code available is hardly any good when you can't compile it. Future changes to gcc could render quite a bit of existing Linux code obsolete.
Sure the Windows API is a moving target. On the other hand, most Win16 and non-game DOS programs run just fine on WinNT4 and probably also on Win2000. This binary compatiblity has given Windows quite an advantage over Unix's source compatiblity on the desktop (at the cost of the WinTel duopoly).
Linux standards are an opportunity to change all of this. Let's hope RedHat, Debian and so on can play nice and avoid the small picture thinking and proprietary maneuvers typical of other Unix vendors. --
Actually, the Presentation Manager came out in about 1989, so it took IBM six years to get it "right".
Unfortunately for OS/2, most people got their OS/2 lumps with version 2.x, where despite the "power" of PM it was nearly unusable out of the box. As crappy as the Windows 3.x GUI was, at least it had program icons people could find.
The guy does have a certain point. I saw the first USB ports in 1996 or so, covered with a "Pending Software Support" sticker. It's only been within the last few months (since the iMac and Win98 came out) that USB has been everywhere.
So, I wouldn't say that the iMac created USB support, but having a USB-only computer certainly helped. --
Right on. I'm using the IBM AT click-clack keyboard with the red trackpoint clitty. Certainly the best PC keyboard ever made (and still retails for $200 or so).
$10 PC keyboards seem to lead to carpal tunnel syndrome. It's my secret hope that the companies foisting these will get hit with a class action suit. --
As far as I know, Darwin isn't booting yet (someone correct me if I'm wrong), so no one yet knows if Darwin is operationally better or worse than Linux or BSD.
I would imagine that the AFP services in Darwin would be much better than Linux/BSD, for example.
Open Source is only part of the reason to run Linux, many people run it only because it's a good OS.
I don't know of any video editing solutions based on Real or NetShow. Real/Netshow are broadcast solutions which QuickTime does not even compete with (at least until next week). --
In a previous thread, someone made the point that you'd probably only buy $50,000 x86 solution to run WinNT. If you were going to run Unix, there's many other options at that price range.
That being said, if RedHat and others are trying to provide commercial-quality support, they probably need a commercial-quality PR department that will assist with benchmark engineering. --
Considering that the survey was probably aimed at NT admins who might be considering Linux, the fact that Linux support and tuning information is harder to find is significant.
Basically the survey said (below the numbers) - "Do you have a high end NT box? Have you tuned it? If so, it's probably not worth the bother to try to figure out how to replace it with a tuned Linux system"
This is much deeper level FUD, than the numbers. Would you bet your job on something that may or may not be well documented and easy to understand? --
There are more NT servers than Linux server (by an order of magnitude, probably), therefore nothing can be concluded.
I think you're making a common Slashdot error - assuming that public webservers listed in Mindcraft is some sort of benchmark for all server usage. NT has greater than 50% of the file+print market, a good chunk (more than 50%?) of the internal mail server market, and a good chunk of the "intranet" web server market which is not measured by Mindcraft.
I don't think anyone really knows how many production Linux boxes there are out there, because unlike commercial operating system, you can't count the paid licence base. --
Very good summary. These types of NT exploits (as opposed to IIS buffer stuff), don't get much press because they are actually really really old. NT 3.1 and Lan Manager servers were comprismised by the same strategy as L0pht, and really not much has changed. I would imagine that this stuff is such old news that it's not even as 31331 as hacking a RedHat 5.1 box.
A year ago, I would have guessed that 75% of NT Internet servers were running with the Lanman service and Ports 137-9 wide open. Recently, things have tightened up a bit, but I'm sure there's quite a few NT admins repeating "C2 Security" over and over again.
Slashdot Mirror
I had a IIfx for a long time. There have been X servers for the mac for a long time (from Apple and others), so an old version should run OK on System 6.08.
You could also install A/UX (Apple 68K Unix) on that machine, which includes an XServer. (But using it precludes using the MacOS interface.)
--
Don't forget that 99% of the game market is running Windows 9x and MacOS, which means no SMP support in the OS. Adding SMP support for the 1% running Linux or NT might not have been worth the cost.
--
It's actually more similar to the comparison between OS/2 and Win3.1.
The Win16 subsystem in NT (WOW - Windows on Windows) works by mapping Win16 and DOS API calls to the Win32 equivalent. Thus Win16 programs get Preemptive MT and Memory Protection under NT. I think this is similar to how WINE works - by mapping Win* calls to the Unix/X equivs.
MacOS X runs the MacOS subsystem (Blue Box) by virtualizing the hardware and "booting" the authentic MacOS. Thus, no modern features between MacOS programs, although the Blue Box probably couldn't take down the OS. Similar to how OS/2 ran DOS and Win16 programs.
The announced Carbon subsystem (subset of MacOS APIs) will work in a manner similar to WOW, if I understand correctly.
--
My understanding was that the Blue Box (MacOS subsystem) really wasn't officially supported under OS X Server. One reason for this is that they have not merged the networking yet, which means you need two IP addresses, one each for Blue and Yellow.
It would also be difficult to see how VirtualPC would be able to work under OS/X at all because all of the hardware is virtualized in the Blue Box. (Virtual PC appears to talk directly to the SCSI and Network hardware using emulated drivers.) I think this is why there's been talk of a "Red Box", which would provide emulation services under OS X.
--
Following myself up, Allison does have a message for those who take the Mindcraft bench at FUD value alone:
The study has also shown that the knowledge of how to tune filesystem performance in Linux is equally obscure. We need to do a better job of educating Linux administrators about how to get the most out of their systems.
I'm sure there'll be more benchmark disappointments in store for us. After all, how else do we learn what we need to fix? But the strength of open source is that we can face the errors without trying to deny them. We just fix 'em and move on.
--
I just took a look at the linked article written by Jeremy Allison of Samba.
/proc filesystem.
A few interesting points -
* In the often referred-to ZD Samba versus NT benchmarks (where Linux+Samba wins), the Samba/Linux configuration was tuned by a Samba team member. Objectively, this makes the ZD benchmark actually less valid as the Mindcraft study, because as far as we know, a Microsoft-employeed SMB developer wasn't actually there tuning the server.
* Tuning Linux properly involves cryptic commands such as:
echo "80 500 64 64 80 6000 6000 1884 2" >/proc/sys/vm/bdflush
echo "60 80 80" >/proc/sys/vm/buffermem
While I'm sure these commands are documented somewhere, this sort of tuning makes the NT Registry Editor look like a model user interface. Low level tuning like this really needs a nicer front end, or preferably, a daemon which monitors system activity and dynamically tunes these settings.
It sounds like the Mindcraft study has been a kick in the pants for the Linux community to get some high performance documentation together. I'd like to see a nice How-To which lays out some of the more obscurantist tricks such as echoing strings to the
--
If I understand correctly, those numbers are public webservers only. MS IIS's market strength has been internal Intranet solutions (where there's probably an existing NT file+print setup). IIIS's intranet market is probably going to being going up, not down, as things like the Office 2000 server get deployed.
--
A few years ago, nobody in their right mind would have proposed any x86 (Novell, OS/2, NT, Linux)solution for anything other than workgroup filesharing or a ccMail postoffice. The fact that you can now seriously consider Linux or NT as a contender at the low end of the midrange market is primarily due to the advances in Intel hardware.
In a few years, Solaris, Tru64, HP/UX, Linux, and NT will all be running on essentially the same Intel IA64 hardware. At this point, the appeal of NT's one-size-fits-all design is going to start breaking down. But on the other hand, hardware equality is going to get Microsoft's salesmen in the door for midrange solutions that were previously above their heads. And Microsoft is more price competitive than commercial Unix, so NT deployment is probably going to increase in this market, not decrease. (Same argument for Linux.)
--
The 'Rush Limbaugh' principle is a very valid point, especially in this context. Don't forget the target market for this study is Microsoft partners and WinNT-based shops.
Aside from all the meaninless numbers (who cares if your web server can saturate a 100BT line with static pages!), the study drives home an important point to NT Administrators - If you've invested in a high end IIS system, and you've got it tuned, there's probably no good reason to switch that box over to Linux. If the Linux box was tuned correctly, I doubt the difference would be that great performance-wise.
Of course, the study didn't address stability, which is the number one problem with IIS.
--
A Quad Xeon is the highest end box that NT runs on (barring Alpha).
Note that WinNT is driving the "high-end" x86 hardware market. Vendors like Dell and Compaq make boxes with only 4 CPUs because that's all vanilla NT will support. When Win2000 comes out, it will support 8? processors, which means the hardware companies will immedeatly follow with 8 CPU iron. (Implictly making this hardware available to some Linux folks.)
Of course a better benchmark would be the $50,000 NT/Dell box versus the $50,000 Sun/HP/DEC box, etc.
--
Some people actually do run core business services on NT, although not many. And if you cut through the hype, most people don't run their core business services on Linux, *BSD, SCO, or any other x86 platform either.
There's a number of reasons for this other than "NT Sux" - x86 server hardware only has started to approach the price/performance of midrange hardware within the last couple years. Also, early versions of Oracle/NT were not very stable, and MS didn't come out with an enterprise competitive DBMS until a couple months ago (SQL 7). Likewise with Linux - the pieces are just now appearing.
Not to mention that most "core business" applications undergo years of planning and development, and have a self life of 10 years or more. VMS is still alive for this reason, and there's no doubt that a certain part of the Unix/S390/AS400 sales pitch is legacy compatibility. There has not been enough time for most shops to build for NT or Linux or migrate what they've got to newer platforms.
So, now that all the pieces are there, expect x86 to start getting midrange marketshare, but it's going to be slow going for either Linux or NT. (As was mentioned in the NT vs Linux debate, at $50K - $100K price point, there's a few more options than x86+Linux or x86+NT.) And, yes 390 will live forever.
--
You've never heard about the worldwide WINS server?
--
How is this going to be any different than AT+T or other private data networks (other than IPv6)?
It seems the discussion is going like this:
1) Private networks offer security and QOS - let's use them.
2) No, the Internet is cool and cheap - let's use it.
3) Hey - the Internet has low security and QOS - let's use a private network! Aren't we smart!
--
While you make valuable points, don't forget that differing Unixes only offer source compatiblity and no binary compatiblity at all. So while it might be nice that you can recompile an OpenWindows program and run it on Linux/Gnome system, that doesn't do an iota for someone whose StarOffice install is failing due to strangely placed files or a library conflict.
Furthermore, while many of the base APIs in Unix are old and stable, the modern, more concerted efforts to create an integrated environment are not. Ten years from now, it may be technically possible to run your legacy Gnome 1.0 application on your Pentium64 SuperPro, but it could involve finding and trying to compile all 30 legacy Gnome libraries. The cost/benefit of porting starts to fall of really quickly.
There's a few more dependancies involved other than just "Unix is Unix". When GNU refused to port gcc to Apple A/UX (System V-based with MacOS GUI), it crippled the platform. All the wonderful Unix source code available is hardly any good when you can't compile it. Future changes to gcc could render quite a bit of existing Linux code obsolete.
Sure the Windows API is a moving target. On the other hand, most Win16 and non-game DOS programs run just fine on WinNT4 and probably also on Win2000. This binary compatiblity has given Windows quite an advantage over Unix's source compatiblity on the desktop (at the cost of the WinTel duopoly).
Linux standards are an opportunity to change all of this. Let's hope RedHat, Debian and so on can play nice and avoid the small picture thinking and proprietary maneuvers typical of other Unix vendors.
--
Newer Macs also support firmware drivers.
--
Actually, the Presentation Manager came out in about 1989, so it took IBM six years to get it "right".
Unfortunately for OS/2, most people got their OS/2 lumps with version 2.x, where despite the "power" of PM it was nearly unusable out of the box. As crappy as the Windows 3.x GUI was, at least it had program icons people could find.
--
The guy does have a certain point. I saw the first USB ports in 1996 or so, covered with a "Pending Software Support" sticker. It's only been within the last few months (since the iMac and Win98 came out) that USB has been everywhere.
So, I wouldn't say that the iMac created USB support, but having a USB-only computer certainly helped.
--
Paul Allen, the self-proclaimed father of the backslash.
--
Right on. I'm using the IBM AT click-clack keyboard with the red trackpoint clitty. Certainly the best PC keyboard ever made (and still retails for $200 or so).
$10 PC keyboards seem to lead to carpal tunnel syndrome. It's my secret hope that the companies foisting these will get hit with a class action suit.
--
As far as I know, Darwin isn't booting yet (someone correct me if I'm wrong), so no one yet knows if Darwin is operationally better or worse than Linux or BSD.
I would imagine that the AFP services in Darwin would be much better than Linux/BSD, for example.
Open Source is only part of the reason to run Linux, many people run it only because it's a good OS.
--
I don't know of any video editing solutions based on Real or NetShow. Real/Netshow are broadcast solutions which QuickTime does not even compete with (at least until next week).
--
In a previous thread, someone made the point that you'd probably only buy $50,000 x86 solution to run WinNT. If you were going to run Unix, there's many other options at that price range.
That being said, if RedHat and others are trying to provide commercial-quality support, they probably need a commercial-quality PR department that will assist with benchmark engineering.
--
Considering that the survey was probably aimed at NT admins who might be considering Linux, the fact that Linux support and tuning information is harder to find is significant.
Basically the survey said (below the numbers) - "Do you have a high end NT box? Have you tuned it? If so, it's probably not worth the bother to try to figure out how to replace it with a tuned Linux system"
This is much deeper level FUD, than the numbers. Would you bet your job on something that may or may not be well documented and easy to understand?
--
The statistic actually is:
There are more NT servers than Linux server (by an order of magnitude, probably), therefore nothing can be concluded.
I think you're making a common Slashdot error - assuming that public webservers listed in Mindcraft is some sort of benchmark for all server usage. NT has greater than 50% of the file+print market, a good chunk (more than 50%?) of the internal mail server market, and a good chunk of the "intranet" web server market which is not measured by Mindcraft.
I don't think anyone really knows how many production Linux boxes there are out there, because unlike commercial operating system, you can't count the paid licence base.
--
Very good summary. These types of NT exploits (as opposed to IIS buffer stuff), don't get much press because they are actually really really old. NT 3.1 and Lan Manager servers were comprismised by the same strategy as L0pht, and really not much has changed. I would imagine that this stuff is such old news that it's not even as 31331 as hacking a RedHat 5.1 box.
A year ago, I would have guessed that 75% of NT Internet servers were running with the Lanman service and Ports 137-9 wide open. Recently, things have tightened up a bit, but I'm sure there's quite a few NT admins repeating "C2 Security" over and over again.
--