Innovation costs time and money; a business like MS won't invest if they don't have to, and as far as they are concerned IE has achieved its goal of browser dominance.
Once this is sufficiently threatened, they will react. Once, say, Firebird picks up 50% of the market, they will react. It's just-in-time management. Until then there are other fish to fry.
You might think that poking the sleeping dragon isn't such a smart idea, though.
I didn't say MS invented any of that, merely that it was new in XP, compared to 2000; the original complaint was that the only thing changing was the UI.
And furthermore, you might find MS had a lot to do with UPnP, including chairing the UPnP forum.
5 years ago Linux was still commandline, KDE came around, now we have KDE 3.2 coming out. Look at the difference, you can look down the list and see hundreds maybe thousands of new features.
Bullshit. 7 years ago, I was running a multiuser linux system (Debian 0.96 on a 486DX4-100) and X ran quite nicely on my S3-968 video card with 2MB VRAM, thank you very much. These days, things don't seem to have moved on, or run that much faster on my P3-1GHz and P4-2GHz.
People like you said the same stuff about WindowsXP, WindowsXP turned out to be Windows2000 with a new skin/theme.
Yeah, and: uPNP, IEEE1394, USB2, fast user switching, RDP, soft firewall, P4 optimisations... a lot of these were incremental, but just because MS aren't rewriting their VM or IDE subsystems every 6 months doesn't mean that nothing ever changes. Linux is very much still in the experimental "fast-growth" stages, particularly on non-X86 architectures.
Also known as Inky, Dinky, Pinky and Clyde; Clyde is the red one, and has a different game logic to the others. He's still predictable, though - Bazo's Breaker is one pattern which basically guarantees you can beat the game.
The big advantage of Linux installation over Windows installation, in my experience, is that I don't have to dork around with all these brain-dead installers and web pages all of which are full of misinformation and poorly-crafted installers. If I want a linux soundcard to work, it's generally just 'alsaconf' and I'm done. One command, one action, one minute.
I agree that the quality of OEM drivers from many manufacturers (sound cards and video cards are the worst) is pretty pathetic. You have to remember that this is NOT due to Microsoft - drivers can be as simple as a.SYS file and a.INF file, and maybe a.DLL if the driver has a userland interface. If Creative feel the need to ship a 25 MB behemoth with a DVD-Audio sample clip, two Flash presentations, multimedia speaker "calibrator", garishly-skinned mixer application (application skins, what IS the point?) and a stupid start-up splash screen that runs on every boot, it's THEIR fault.
Microsoft have a strategy of shipping Windows with approved minimal-functionality drivers, which does mean that MOST hardware works out of the box, for some value of "works". Both OSes have problems with bleeding edge stuff. Building my new machine, I found Windows XP needed OEM drivers to support the S-ATA controller and the Gigabit Ethernet controller on Asus's new Granite Bay boards, for example. This wasn't exactly an earth-shattering discovery, and I'd already collected the latest versions from the net; I'd have to do EXACTLY the same for Linux, with the additional step of integrating the driver source into the kernel build, or worse, installing binary-only modules for a specific kernel version whether I wanted that version or not.
On that basis, I think the Windows driver architecture (WDM) is better than Linux. Factor in power management and wake-from-sleep support and WDM is streets ahead.
Have you actually TRIED installing them? The only reason they want you to use the installation CD is so you get all the other junk that's on it. Now, I suspect like me the parent poster could give a shit less about having Creative Keytar & some talking parrot installed, thankyouverymuch. Every machine I build has a Creative sound card installed, and that's how I do my driver installs; I never use the fucking installation CD, precisely because it's usually only the drivers that are needed, and drivers on CDs are so out of date as to be worthless.
What Microsoft could usefully do is acquire Copernic and integrate it into Windows. For the uninitiated, Copernic is a meta search engine; it contacts other search engines (including Google as of Copernic Agent) to do the heavy lifting, then collects the results.
That sort of functionality wouldn't be out of place in a modern OS.
I think you're referring to the EPIA-ME6000 (600 MHz Eden CPU). Yes, it is fanless and supports USB2; I didn't know it existed - it wasn't available when we built ours; the M9000 (933 MHz C3 CPU) had only just been announced. Might consider upgrading - thanks.
...set up as a portable demo box for our software (which needs an FPGA PCI card, so laptops aren't possible). It's the cheapest Epia board (533 MHz C3 chip). To my suprise, it's been rock-solid stable and pretty functional (although we've only got Windows 2000 on it).
However, the lack of L2 cache (and maybe not even any L1?) absolutely cripples performance on some things; a Logitech USB web cam struggles to get 3 FPS, because it needs the CPU to do decompression of the video stream. USB-1 isn't fast enough to stream 640x480 uncompressed video, and this board doesn't support USB-2 (the newest ones do, but they also NEED a CPU fan).
I plan to play with emulation (I think it'd be amusing to turn one into a every-obsolete-computer-you-ever-owned box) but the lack of cache might kill that idea. It ought to be able to emulate a 2MHz 6502 though...
Nope, the correct term is "sweat" as in "sweat soldering" - the plumbers trick of coating two ends of copper pipe in solder, positioning and then blowtorching them to make the final joint.
Cooling is all fine and dandy, but increasing the voltage increases electric field strength across the transistors, as well as increasing the power as the square of the voltage. As I understand it, this should cause many more early failures due to electro-migration and thinning of the aluminium / copper traces in the metal layers.
However, I'm guessing electro-migration is thermally activated, or at least sensitive to the temperature; does the extreme cooling mitigate the increased field strengh?
I do know what a true virtual machine, and when I label the MS dos emulator as a virtual machine, it's only because that's what I've read. Either way, it emulates DOS like other dos emulators do.
In that case, you don't know what "CMD.EXE" is; here's a free clue, it's NOT the same as COMMAND.COM. It's a real 32 bit process, which JUST HAPPENS to emulate some DOS commands.
CMD.EXE is a real CLI to Windows, in the same way that "bash" is a CLI to Linux; it's just that bash is more configurable and more powerful. CMD.EXE is most certainly NOT a virtual machine; start one up and compare the running processes to COMMAND.COM, which will start up an NTVDM.EXE process (which is actually the NT Virtual DOS Machine you have heard about).
NTVDM is also commonly found hanging out with WOWEXEC.EXE - this is the 16 bit Windows-on-Windows process which runs Win16 applications. Win64 on Itanium will use a similar mechanism to suport Win32, probably called WOWEXEC32 or similar.
Maxtor DiamondMax something-740s (40 GB, 7200 rpm) on Intel i845 boards (ICH4 south bridge I think)
Do your drives have 8MB of cache on them ?
No, neither do the SCSI disks. I am contemplating replacing the IDE drives with WD special editions, but I think I might wait and see how SATA pans out. The SCSI disks are 2 year-old IBM (probably 2nd generation) 10K RPM disks - possibly the access time is what makes the difference for my work load. Certainly the head rate from them is nothing special these days.
Possibly it's all perception, but a SCSI disk subsystem seems to make a overall better balanced system which doesn't lock up solid thrashing the disks. That said, if SATA / SAS takes off I might well go for SATA RAID on my next system; SCSI IS expensive and the heat and noise problems I've had with one of my systems might swing it.
Maybe I'm just one of those "rare" power users, but I commonly run up against the I/O limits of IDE on the desktop (Windows XP); with IDE, things just bog down under I/O intensive workloads (like, running Adobe Acrobat & Illustrator, MS Visual Studio.NET compile, maybe a few RDP / terminal server sessions).
My other Windows system is entirely SCSI based and never bogs down in the same way (and before you ask, yes, they have similar RAM, M/B, CPU, software load...). In fact, it's usually CPU bound.
Partly it's due to the fact that Windows pages to disk, even with large amounts of RAM (both machines have 512 MB, about to be upgraded); that just kills IDE stone dead; under heavy paging it's totally unusable without multiple drives on dedidicated channels.
Sure, I get no advantage by having SCSI optical drives except for reclaiming the IDE interrupts. Annoyingly, mine are severely obsolete now that very few people make SCSI CD-R drives or DVDROM drives. For the disk subsystem though, I do see a real difference. I think the randomised I/O you mention actually happens most of the time with a real working system (certainly my systems work for their living;-). I've yet to see a benchmark that actually reflects the reality of life as a software developer.
We're still waiting for a solid 1.4 release on FreeBSD because of the nonsense requiring X, or at least Xvfb, to be running to support Java graphics calls
Er, that's exactly what Sun (finally) fixed in 1.4; the ability to run AWT applications "headless" (without X). Finally your app can call Toolkit.getToolkit().beep() and not crash horribly trying to connect to:0...
In a straight port of code highly optimized for x86-32, Counter-Strike dedicated server tests with both 32- and 64-bit versions revealed a 30% clock-for-clock gain, and is expected to show further performance gains in future upgrades.
Sounds like they are simply re-compiling with a new tool chain; nothing about actually changing the code base to take specific advantage of Opteron features. Still, kudos to their coders if their code base just works on 64 bit platforms; there'll be plenty out there that won't, despite availability of the SDKs and programming guides like this and this
So, are there any independent review sites out there? How do they get their hands on pre-release hardware? Just how close to payola is the whole thing, anyway?
Enquiring minds want to know (before they blow ${WEEKS_WAGES} on new toys...)
You mean to tell me that MS has disable the copy-and-paste, too? Seriously. Why couldn't I just copy-and-paste my secret memo into a text file and then forward it to FuckedCompany or AssWipeMemos or whatever Pud's pimping.
Go google for "mandatory access control" vs "discretionary access control". Basically, if you have clearance to create top-secret documents, you CAN'T (the OS won't let you) create documents at a lower clearance level; sure, you can cut and paste, but only into another top-secret document.
Having recently worked for a Big Japanese A/V manufacturer, I can most certainly assure you that uITRON (micro-iTRON) ain't going away. Instead, they layer an uITRON compatibility layer on top of the Linux kernel (modifying the kernel as necessary, mentioning the GPL is a big no-no) and then run everything (including a Java VM of all things) on top of the uiTRON layer.
I can't actually think of a worse thing to do; lose all the flexibility of Linux and the real-time behaviour of uiTRON. Gah.
At least one embedded OS (eCOS) already HAS an iTRON API but Redhat seem to have killed it off.
ACTUALLY, it predates the Internet by a quite a few years; it's from the Molesworth books (English public skool humour) by Willans and Searle. See here for details. They are very funny, as any fule kno.
Wrong, wrong wrong! You don't know what you're talking about. MS have been transitioning the API to being 64 bit clean for at least a year, probably two; They already supplied a 64 bit SDK/DDK with the MSDN so developers can check their applications. FFS, Nvidia are SHIPPING 64 bit video drivers for their cards for WinXP-64.
It's already happening, you just haven't noticed it yet.
Jon.
Re:So where's the SVG authoring apps?
on
SVG On the Rise
·
· Score: 1
Have you tried Adobe Illustrator 10? What doesn't it do that you need?
So is this designed to be used with Palladium products? It sounds like it's a mixed mode CD with some control information in the data tracks that is read by Palladium-enabled applications or OSes to control what the user can do with it.
If that's all it is, it's not going to stop anyone from ripping it on pre-Palladium systems, nor from CD players with digital I/O (although that'll only work at single speed).
And what does the article mean by "layered"? Surely not an actual multilayered disk like a DVD? Is that backwards compatible?
Slashdot Mirror
Once this is sufficiently threatened, they will react. Once, say, Firebird picks up 50% of the market, they will react. It's just-in-time management. Until then there are other fish to fry.
You might think that poking the sleeping dragon isn't such a smart idea, though.
Jon.
And furthermore, you might find MS had a lot to do with UPnP, including chairing the UPnP forum.
Jon.
Bullshit. 7 years ago, I was running a multiuser linux system (Debian 0.96 on a 486DX4-100) and X ran quite nicely on my S3-968 video card with 2MB VRAM, thank you very much. These days, things don't seem to have moved on, or run that much faster on my P3-1GHz and P4-2GHz.
People like you said the same stuff about WindowsXP, WindowsXP turned out to be Windows2000 with a new skin/theme.
Yeah, and: uPNP, IEEE1394, USB2, fast user switching, RDP, soft firewall, P4 optimisations... a lot of these were incremental, but just because MS aren't rewriting their VM or IDE subsystems every 6 months doesn't mean that nothing ever changes. Linux is very much still in the experimental "fast-growth" stages, particularly on non-X86 architectures.
Jon.
Jon.
Jon
I agree that the quality of OEM drivers from many manufacturers (sound cards and video cards are the worst) is pretty pathetic. You have to remember that this is NOT due to Microsoft - drivers can be as simple as a .SYS file and a .INF file, and maybe a .DLL if the driver has a userland interface. If Creative feel the need to ship a 25 MB behemoth with a DVD-Audio sample clip, two Flash presentations, multimedia speaker "calibrator", garishly-skinned mixer application (application skins, what IS the point?) and a stupid start-up splash screen that runs on every boot, it's THEIR fault.
Microsoft have a strategy of shipping Windows with approved minimal-functionality drivers, which does mean that MOST hardware works out of the box, for some value of "works". Both OSes have problems with bleeding edge stuff. Building my new machine, I found Windows XP needed OEM drivers to support the S-ATA controller and the Gigabit Ethernet controller on Asus's new Granite Bay boards, for example. This wasn't exactly an earth-shattering discovery, and I'd already collected the latest versions from the net; I'd have to do EXACTLY the same for Linux, with the additional step of integrating the driver source into the kernel build, or worse, installing binary-only modules for a specific kernel version whether I wanted that version or not.
On that basis, I think the Windows driver architecture (WDM) is better than Linux. Factor in power management and wake-from-sleep support and WDM is streets ahead.
Jon
Here is your free clue for the day.
Have you actually TRIED installing them? The only reason they want you to use the installation CD is so you get all the other junk that's on it. Now, I suspect like me the parent poster could give a shit less about having Creative Keytar & some talking parrot installed, thankyouverymuch. Every machine I build has a Creative sound card installed, and that's how I do my driver installs; I never use the fucking installation CD, precisely because it's usually only the drivers that are needed, and drivers on CDs are so out of date as to be worthless.
Jon.
Your Name is SatanicPuppy, and you're an idiot.
Jon.
That sort of functionality wouldn't be out of place in a modern OS.
Jon
Jon.
However, the lack of L2 cache (and maybe not even any L1?) absolutely cripples performance on some things; a Logitech USB web cam struggles to get 3 FPS, because it needs the CPU to do decompression of the video stream. USB-1 isn't fast enough to stream 640x480 uncompressed video, and this board doesn't support USB-2 (the newest ones do, but they also NEED a CPU fan).
I plan to play with emulation (I think it'd be amusing to turn one into a every-obsolete-computer-you-ever-owned box) but the lack of cache might kill that idea. It ought to be able to emulate a 2MHz 6502 though...
Jon.
Jon
However, I'm guessing electro-migration is thermally activated, or at least sensitive to the temperature; does the extreme cooling mitigate the increased field strengh?
Jon.
In that case, you don't know what "CMD.EXE" is; here's a free clue, it's NOT the same as COMMAND.COM. It's a real 32 bit process, which JUST HAPPENS to emulate some DOS commands.
CMD.EXE is a real CLI to Windows, in the same way that "bash" is a CLI to Linux; it's just that bash is more configurable and more powerful. CMD.EXE is most certainly NOT a virtual machine; start one up and compare the running processes to COMMAND.COM, which will start up an NTVDM.EXE process (which is actually the NT Virtual DOS Machine you have heard about).
NTVDM is also commonly found hanging out with WOWEXEC.EXE - this is the 16 bit Windows-on-Windows process which runs Win16 applications. Win64 on Itanium will use a similar mechanism to suport Win32, probably called WOWEXEC32 or similar.
Jon.
Maxtor DiamondMax something-740s (40 GB, 7200 rpm) on Intel i845 boards (ICH4 south bridge I think)
Do your drives have 8MB of cache on them ?
No, neither do the SCSI disks. I am contemplating replacing the IDE drives with WD special editions, but I think I might wait and see how SATA pans out. The SCSI disks are 2 year-old IBM (probably 2nd generation) 10K RPM disks - possibly the access time is what makes the difference for my work load. Certainly the head rate from them is nothing special these days.
Possibly it's all perception, but a SCSI disk subsystem seems to make a overall better balanced system which doesn't lock up solid thrashing the disks. That said, if SATA / SAS takes off I might well go for SATA RAID on my next system; SCSI IS expensive and the heat and noise problems I've had with one of my systems might swing it.
Jon
My other Windows system is entirely SCSI based and never bogs down in the same way (and before you ask, yes, they have similar RAM, M/B, CPU, software load...). In fact, it's usually CPU bound.
Partly it's due to the fact that Windows pages to disk, even with large amounts of RAM (both machines have 512 MB, about to be upgraded); that just kills IDE stone dead; under heavy paging it's totally unusable without multiple drives on dedidicated channels.
Sure, I get no advantage by having SCSI optical drives except for reclaiming the IDE interrupts. Annoyingly, mine are severely obsolete now that very few people make SCSI CD-R drives or DVDROM drives. For the disk subsystem though, I do see a real difference. I think the randomised I/O you mention actually happens most of the time with a real working system (certainly my systems work for their living
Jon.
Er, that's exactly what Sun (finally) fixed in 1.4; the ability to run AWT applications "headless" (without X). Finally your app can call Toolkit.getToolkit().beep() and not crash horribly trying to connect to :0...
Jon.
Sounds like they are simply re-compiling with a new tool chain; nothing about actually changing the code base to take specific advantage of Opteron features. Still, kudos to their coders if their code base just works on 64 bit platforms; there'll be plenty out there that won't, despite availability of the SDKs and programming guides like this and this
Jon.
Enquiring minds want to know (before they blow ${WEEKS_WAGES} on new toys...)
Jon.
Go google for "mandatory access control" vs "discretionary access control". Basically, if you have clearance to create top-secret documents, you CAN'T (the OS won't let you) create documents at a lower clearance level; sure, you can cut and paste, but only into another top-secret document.
Jon.
I can't actually think of a worse thing to do; lose all the flexibility of Linux and the real-time behaviour of uiTRON. Gah.
At least one embedded OS (eCOS) already HAS an iTRON API but Redhat seem to have killed it off.
Jon
Jon.
It's already happening, you just haven't noticed it yet.
Jon.
Jon.
If that's all it is, it's not going to stop anyone from ripping it on pre-Palladium systems, nor from CD players with digital I/O (although that'll only work at single speed).
And what does the article mean by "layered"? Surely not an actual multilayered disk like a DVD? Is that backwards compatible?
More details anyone?
Jon