Well, it is a Microsoft product, so it's not without its flaws (The Vista dependency for one), but over all it's a good API for taking advantage of modern hardware without all the legacy crud that plagues OpenGL.
If you've used D3D8 or older, you'll find it a massive improvement.
Thing is it's not even close to being easy to use anymore. Especially not if you're interested in performance.
Because of the two decades of crud that has accumulated, there are so many ways of falling off the fast path in OGL, and it's next to impossible to know beforehand what will and will not work. Drivers are also a bitch to develop and maintain because of the size of the thing, which makes things even worse since what works on Nvidia may not work on ATI and vice versa.
The only way to fix this is a good cleanup to bring it in line with modern hardware. What they did was add even more crud.
That certainly isn't my experience. Most people on the OGL discussion boards were very much looking forward to the changes to the API. The previews Khronos posted in the Pipeline newsletter looked bloody amazing.
But when those previews are followed by almost a year of complete silence and then finally an API which is nothing at all like the one they promised, but rather some more spit and polish on the mess that is OGL 2.1 (much like OGL 2.0 was really just 1.6 with a new name), people got pissed off. And rightfully so.
The only ones pleased with this change as far as I've been able to gather are the CAD people wanting to continue to run their old, stale OpenGL bases code until the end of time. For new development, using OpenGL is a pain in the back side, which is why I just began bringing my renderer up in D3D10.
I can mention lots, but they don't have extremely awesome HDR graphics and player counts in the millions.
Their developers still can make a comfortable living though, and that's really the point. When you're a small team you don't need EA-scale success to make a profit.
It's time to drop this old complaint. In my experience this hasn't been the case since around the time the Radeon 9700 was king (in Windows). In fact, with the problems Nvidia has been having on Vista I'd say the opposite is closer to the truth. Driver stability just isn't a problem for ATI/AMD any more.
It's not even a hypothesis. For something to be a scientific hypothesis it needs to be testable. We must be able to prove it right or wrong.
The basic premise of creationism is that god did it, and that the existence of god cannot be proven. You have to believe, meaning it's simply not science.
ATI have open specs. At least for a lot of their hardware. They are releasing more and more documentation as it gets cleaned up and cleared by legal. Open Source ATI drivers are coming on in leaps and bounds as a result.
fglrx has seen massive improvement lately. It is supposed to be mostly in sync with the Windows Catalyst drivers these days. It's still a bit off perfect of course, but a lot better than it was.
Neither KDE nor Gnome are just window managers (that's Metacity and Kwin). Desktop environment is a more fitting term for them. They both aim to include most of what you need for basic day-to-day use of your computer. They also make sure everything they include is nicely consistent, which makes for a good user experience.
As for your speed concerns, I don't see how inclusion of a few new apps will slow down anything? It will take a bit more disk space probably, but it won't slow anything down unless you use these new apps. You're also free to uninstall anything you feel is redundant.
Yeah, I'd be fine with that, but that is unfortunately not how it works. IE8 will default to IE7 mode, and will only be standards compliant if you add the tag telling it you want it to be. That's what the uproar is about. This solution basically makes sure IE7 will be around (and targeted by developers not knowing any better) forever.
What's not to like? This tag allows the people that are writing horrible IE-only HTML/CSS in Visual Studio today to continue their business as usual. There is absolutely no incentive for these people to adhere to standards since the IE7 engine will be kept around for "a couple of web lifetimes" apparently (source). This effectively locks browsers other than IE out of these sites since they won't have this IE7 compatibilty.
I hear people use that excuse all the time, but I really don't buy it. Does Firefox, Opera or Safari break all these old IE6/7 pages? I know they don't render all of them pixel perfect, but it's been a real good long time since I've seen them break a page completely (making it unreadable).
IE8 in standards mode should render pages almost exactly like these browsers, so what's the problem with making that the default again?
Microsoft is afraid of a situation like when they introduced IE7, but that only happened because IE7 wasn't completely standards compliant and instead introduced a new set of quirks and bugs people had to develop hacks to get around. A completely standards compliant browser shouldn't have that much of a problem since people have been targeting all these other browsers for yonks already. The few pages that actually are affected could also easily be updated to force IE into IE7 mode through this new meta tag.
Compared to having to add this tag forever and ever until the end of time to stay out of IE7 mode, that's a tiny price to pay.
Thanks a lot mate. Been experimenting with this a bit now and it seems to work like a charm. I'm able to grow the array by both adding drives and exchanging them all for larger ones. The bad news is that I'm now about to throw $1500 away on a new server with way too much storage for my needs.:)
Heh, I think I'll stick with one completely unfamiliar concept at a time. My experience with Linux RAID stops with once setting it up through OpenFiler on VMWare.:)
Out of interest, how would this work if you switched all the 250GB drives for 500GB drives (one at a time to keep the array alive of course), and wanted to take advantage of the extra space on each new drive? That's a more likely scenario for me and my home server as I don't want to keep adding drive after drive, but I would like to add more space occasionally. Does either Linux RAID or ZFS support that without too much hassle, or would I have to connect all drives (old and new) and do a copy to a new array on the new disks?
Slashdot Mirror
Well, it is a Microsoft product, so it's not without its flaws (The Vista dependency for one), but over all it's a good API for taking advantage of modern hardware without all the legacy crud that plagues OpenGL.
If you've used D3D8 or older, you'll find it a massive improvement.
Thing is it's not even close to being easy to use anymore. Especially not if you're interested in performance.
Because of the two decades of crud that has accumulated, there are so many ways of falling off the fast path in OGL, and it's next to impossible to know beforehand what will and will not work. Drivers are also a bitch to develop and maintain because of the size of the thing, which makes things even worse since what works on Nvidia may not work on ATI and vice versa.
The only way to fix this is a good cleanup to bring it in line with modern hardware. What they did was add even more crud.
You don't fork a spec. You create a new one and try to get it accepted by the industry (ATI, Nvidia and Intel in this case).
Good luck with that.
That certainly isn't my experience. Most people on the OGL discussion boards were very much looking forward to the changes to the API. The previews Khronos posted in the Pipeline newsletter looked bloody amazing.
But when those previews are followed by almost a year of complete silence and then finally an API which is nothing at all like the one they promised, but rather some more spit and polish on the mess that is OGL 2.1 (much like OGL 2.0 was really just 1.6 with a new name), people got pissed off. And rightfully so.
The only ones pleased with this change as far as I've been able to gather are the CAD people wanting to continue to run their old, stale OpenGL bases code until the end of time. For new development, using OpenGL is a pain in the back side, which is why I just began bringing my renderer up in D3D10.
One step at a time. We're going to end up there but it's silly to try to do it all in one go. Especially if you're AMD.
I can mention lots, but they don't have extremely awesome HDR graphics and player counts in the millions.
Their developers still can make a comfortable living though, and that's really the point. When you're a small team you don't need EA-scale success to make a profit.
Well it depends on the resolution, but Crysis renders between 1 and 2 million triangles every frame and my HD3850 can do around 30 FPS at 1280x800.
Peak throughput with passthrough shaders and no state changes is much higher though, but obviously that isn't very interesting in the real world.
It's time to drop this old complaint. In my experience this hasn't been the case since around the time the Radeon 9700 was king (in Windows). In fact, with the problems Nvidia has been having on Vista I'd say the opposite is closer to the truth. Driver stability just isn't a problem for ATI/AMD any more.
It's not even a hypothesis. For something to be a scientific hypothesis it needs to be testable. We must be able to prove it right or wrong. The basic premise of creationism is that god did it, and that the existence of god cannot be proven. You have to believe, meaning it's simply not science.
About where Dosbox is today I reckon.
The Quake 3 source is released under the GPL, so yes. :)
ATI have open specs. At least for a lot of their hardware. They are releasing more and more documentation as it gets cleaned up and cleared by legal. Open Source ATI drivers are coming on in leaps and bounds as a result.
Too late.
fglrx has seen massive improvement lately. It is supposed to be mostly in sync with the Windows Catalyst drivers these days. It's still a bit off perfect of course, but a lot better than it was.
We already have liboil which does more or less the same thing, so it's not like this is a revolutionary development.
That's a good idea actually. A Fireflyish TV-show about Han Solo's early life could be kind of cool.
Neither KDE nor Gnome are just window managers (that's Metacity and Kwin). Desktop environment is a more fitting term for them. They both aim to include most of what you need for basic day-to-day use of your computer. They also make sure everything they include is nicely consistent, which makes for a good user experience.
As for your speed concerns, I don't see how inclusion of a few new apps will slow down anything? It will take a bit more disk space probably, but it won't slow anything down unless you use these new apps. You're also free to uninstall anything you feel is redundant.
Read it and weep, I know I did. (Last 3 paragraphs are the relevant ones.)
Yeah, I'd be fine with that, but that is unfortunately not how it works. IE8 will default to IE7 mode, and will only be standards compliant if you add the tag telling it you want it to be. That's what the uproar is about. This solution basically makes sure IE7 will be around (and targeted by developers not knowing any better) forever.
What's not to like? This tag allows the people that are writing horrible IE-only HTML/CSS in Visual Studio today to continue their business as usual. There is absolutely no incentive for these people to adhere to standards since the IE7 engine will be kept around for "a couple of web lifetimes" apparently (source). This effectively locks browsers other than IE out of these sites since they won't have this IE7 compatibilty.
Hooray, everybody wins. No, wait, the other one.
I hear people use that excuse all the time, but I really don't buy it. Does Firefox, Opera or Safari break all these old IE6/7 pages? I know they don't render all of them pixel perfect, but it's been a real good long time since I've seen them break a page completely (making it unreadable).
IE8 in standards mode should render pages almost exactly like these browsers, so what's the problem with making that the default again?
Microsoft is afraid of a situation like when they introduced IE7, but that only happened because IE7 wasn't completely standards compliant and instead introduced a new set of quirks and bugs people had to develop hacks to get around. A completely standards compliant browser shouldn't have that much of a problem since people have been targeting all these other browsers for yonks already. The few pages that actually are affected could also easily be updated to force IE into IE7 mode through this new meta tag.
Compared to having to add this tag forever and ever until the end of time to stay out of IE7 mode, that's a tiny price to pay.
Thanks a lot mate. Been experimenting with this a bit now and it seems to work like a charm. I'm able to grow the array by both adding drives and exchanging them all for larger ones. The bad news is that I'm now about to throw $1500 away on a new server with way too much storage for my needs. :)
Heh, I think I'll stick with one completely unfamiliar concept at a time. My experience with Linux RAID stops with once setting it up through OpenFiler on VMWare. :)
Thanks. Just trying it would probably be the best approach yeah. :) Guess I'll have a play in VMWare when I find the time.
Out of interest, how would this work if you switched all the 250GB drives for 500GB drives (one at a time to keep the array alive of course), and wanted to take advantage of the extra space on each new drive? That's a more likely scenario for me and my home server as I don't want to keep adding drive after drive, but I would like to add more space occasionally. Does either Linux RAID or ZFS support that without too much hassle, or would I have to connect all drives (old and new) and do a copy to a new array on the new disks?