Most modern screens can't display deep colour. My Dell U2711 can do it, but you really have to buy a high-end or professional display like it to get 10-bit colour support. Since most displays can't show it, there's not all that much demand to support it.
Colour profiles. Chrome ignores them. On my computers, every image I open up in Chrome is oversaturated compared to opening it in an image editor like Photoshop.
JPEG XR produces images similar to JPEG-2000 while having complexity similar to JPEG, supports transparencies, requires support for lossless compression (unlike JPEG) since lossless is just a quantizer setting, and it's already supported by IE9.
That last bit is probably the most important part. IE's marketshare is shrinking, but it's still big enough that any format it doesn't support is unlikely to see widespread support as the only format available for a site. I doubt IE will ever support WebP, and as such, no website will ever really be able to use WebP. Not unless they do browser detection, and most sites won't bother with multiple image compression formats, they're going to pick the best common one they can, which is currently PNG or JPEG.
Remember PNG alpha support... Until IE supported it, nobody really used it. Once IE did, it became mainstream.
Intel's past use of PowerVR chips was at a time when smartphone screens were still pretty low-res, and the expectations of graphical performance on a smartphone was very different from what was expected on a notebook. Cedertrail (their upcoming Atom product) is using a Series 5 chip (the 545) rather than a Series 5XT chip (like the PowerVR SGX543MP2 in the iPad 2 and iPhone, or the SGX543MP4 in the Playstation Vita). The 545 is certainly an improvement over their previous single-core chips, but I doubt it will compare favourably to the multicore graphics solutions in modern smartphones and tablets. It's a very odd decision on Intel's prt.
Anyhow, the failure of Intel's chips in the smartphone and tablet space has little to do with graphics, and more to do with the fact that Atom performs similar to a Cortex A8 or A9, and yet early Atom chips used two or three times as much power to do that. Intel has narrowed the gap quite a bit, although they're still not there yet. Their next-gen parts might achieve this, but much like other architectures have had trouble displacing Intel for the desktop/notebook crown, so too will Intel have trouble displaying ARM in the embedded space. Merely matching the performance-per-watt of ARM's chips isn't enough, because at that point, people will ask "what's the point of using Intel's chips over ARM, they perform the same but don't have as big an install base so there are no advantages."
The only way Intel will get anywhere is by doing something BETTER than ARM, and they haven't managed that quite yet.
True, but there's also the tendency from researchers to try to make their discoveries seem as important as possible. They'll discover something new and report the benefits it could theoretically bring under optimal conditions, when those conditions might never actually be obtained during real-world use.
Sure, I thought that was implied, since the cases of new engine licensees for id tech 4 is rather minimal, let alone 3 (Urban Terror probably paid almost nothing considering the age of the engine). And any new id tech 4 games that come out are probably from licenses that predate the takeover anyhow.
But I've never seen anything saying Zenimax would not allow licensing. id was bought in 2009, and at QuakeCon 2009 Carmack was saying he intended to petition Zenimax to release the id tech 4 source when Rage came out, and from what we've heard from him, he never really met with any opposition. He even was surprised to get an enthusiastic response from their legal department about the benefits to releasing older source. I can't find any articles that mention a negative response toward this policy by Zenimax.
It came out six years ago. Done by Raven. Not a great game, but not a bad one either. Decent, is the word I'd use. Multiplayer was fun, an update of Quake 3. It can be purchased from a variety of digital distribution networks, such as Steam or Impulse ($20 each), or from online stores like Amazon ($8 for mac version, $38 for PC)
Shadowmaps can still look worse than stencil shadows in many situations. Especially if the stencil shadows are not of sufficiently high resolution to produce smooth edges.
On the other hand, stencil shadows can do soft shadowing a heck of a lot easier.
No, they didn't, they simply said that they would stop licensing id engines for games that are not published by Zenimax. If third party developers want to use an id engine, they need to publish through Zenimax.
... they'll fit right into the steady curve of slowly but steadily increasing battery capacity. People assume that all these battery advancements we keep hearing about never pan out. Well, some of them do, but once the researchers silly claims are brought down to be a bit more realistic, and after the years go by before they actually hit the market, they're just incremental improvements on what was available before they came out.
but why would you be licensing only a binary under BSD? What would be the point of that?
Android is a pretty good example, if we're willing to put the ASL in the place of the BSD here. You can get your hands on a bunch of ASL-licensed binaries on your handset, but nobody is obligated to give you the source.
One of the events that prompted the founding of the FSF was when Stallman wanted to fix a problem with a printer driver, but was forbidden from getting the source. That's not all that different from wanting to fix a problem with your Android handset, but being forbidden from getting the source.
So if I have a BSD-licensed binary, but not the source or any way to get the source, it's not opensource, or free software? If the BSD is only sometimes a free software or opensource license, it's not really free or open at all.
Well, I always just assumed while watching it that the movie was *set* in Vancouver, although it wasn't. Anyhow, being made in Canada certainly influenced the production, but it's ultimately up to the director/producers/etc to make decisions. You've got movies made in Canada that have no indication that they were. Source Code was made here in Montreal, but I had no idea until the credits rolled. That's the way it normally is; we watch a movie, and only when we start seeing a lot of French Canadian names in the credits does it become obvious the movie was made locally.
Some of those US TV series, though, were basically Canadian shows in everything but budget. Stargate SG-1, for example. My understanding is that the owners paid the budget, but almost all of the writers/producers/executive producers/etc were Canadian. So we like to take credit for shows like that;)
Please don't blame Canada for that movie. We didn't write it, we didn't direct it, we didn't produce it. We just provided facilities and most of the cast and crew, not the creative team that actually made the film.
- You have the freedom to run the program as you wish, for any purpose. - You have the freedom to modify the program to suit your needs. (To make this freedom effective in practice, you must have access to the source code, since making changes in a program without having the source code is exceedingly difficult.) - You have the freedom to redistribute copies, either gratis or for a fee. - You have the freedom to distribute modified versions of the program, so that the community can benefit from your improvements.
The Apache and BSD licenses violate the second criteria for free software licenses, they do not grant you the freedom to modify the program to suit your needs, because they do not grant you access to the source. You can throw around the binaries to an Apache or BSD licensed program yourself, but nobody is obligated to give you the source. In that way, they're no different than most other freeware. Heck, some of Microsoft's shared source licenses are more open than them.
- You have the freedom to run the program as you wish, for any purpose. - You have the freedom to modify the program to suit your needs. (To make this freedom effective in practice, you must have access to the source code, since making changes in a program without having the source code is exceedingly difficult.) - You have the freedom to redistribute copies, either gratis or for a fee. - You have the freedom to distribute modified versions of the program, so that the community can benefit from your improvements.
According to RMS himself, founder of the FSF, the Apache and BSD licenses are not free software licenses because they specifically violate the second value; the freedom to modify, which requires source code.
There's a tremendous amount of FUD being spread about key competitors. As you've noted, the "Android's not really free/libre" is one of the FUD talking points.
Uh, well, it's not, really. The Apache license (as well as BSD licenses) make no requirement that the source be provided, so they're not really "free software" licenses in the "freedom" sense... They're certainly not opensource licenses since they make no requirement that the source be open. They're basically just free-as-in-beer distribution licenses; you can share the binaries all you want, but you can't really change them, because nobody is obligated to give you the source to go with those binaries.
I built a ZFS-based home file server a few years ago using OpenSolaris for ZFS support. However, OpenSolaris is dead, and I was never all that comfortable with it anyhow, so I'd like to move to FreeBSD.
Yeah and of course you can't blame Ubuntu for not sticking with Gnome 2. What were their options for their GTK DE then if they didn't care for the direction Gnome was heading? XFCE? I don't like particularly like Unity and maybe they should have just taken advantage of the customizability of Gnome shell and made and maintained a very unique and cool Ubuntu flavor of Gnome 3 but I like the idea of having multiple unique desktop environments out there competing. Maybe in a few years Unity becomes something pretty cool for the standard user...
But Unity isn't a desktop environment. It's a GNOME shell replacement, running on top of GNOME.
Linux Mint's approach for their upcoming release is going to be to migrate to GNOME 3, but supplement/replace GNOME Shell 3 (which looks a lot like Unity) with something that looks a lot more like GNOME Shell 2. In other words, they're going to run the latest and greatest, but tweak and add to it to provide a more traditional UI.
As somebody who is shortly going to be migrating from OpenSolaris (and I never really figured package management out there either) to FreeBSD, could you elaborate a bit on FreeBSD's package management? I assumed that it had something comparable to apt.
Slashdot Mirror
Most cameras from high end point & shoot on up (basically, anything that takes RAW) is already recording 10+ bpp.
Most modern screens can't display deep colour. My Dell U2711 can do it, but you really have to buy a high-end or professional display like it to get 10-bit colour support. Since most displays can't show it, there's not all that much demand to support it.
It could be the first image compression that uses Mechanical Turk as a core component ;)
Colour profiles. Chrome ignores them. On my computers, every image I open up in Chrome is oversaturated compared to opening it in an image editor like Photoshop.
JPEG XR produces images similar to JPEG-2000 while having complexity similar to JPEG, supports transparencies, requires support for lossless compression (unlike JPEG) since lossless is just a quantizer setting, and it's already supported by IE9.
That last bit is probably the most important part. IE's marketshare is shrinking, but it's still big enough that any format it doesn't support is unlikely to see widespread support as the only format available for a site. I doubt IE will ever support WebP, and as such, no website will ever really be able to use WebP. Not unless they do browser detection, and most sites won't bother with multiple image compression formats, they're going to pick the best common one they can, which is currently PNG or JPEG.
Remember PNG alpha support... Until IE supported it, nobody really used it. Once IE did, it became mainstream.
Intel's past use of PowerVR chips was at a time when smartphone screens were still pretty low-res, and the expectations of graphical performance on a smartphone was very different from what was expected on a notebook. Cedertrail (their upcoming Atom product) is using a Series 5 chip (the 545) rather than a Series 5XT chip (like the PowerVR SGX543MP2 in the iPad 2 and iPhone, or the SGX543MP4 in the Playstation Vita). The 545 is certainly an improvement over their previous single-core chips, but I doubt it will compare favourably to the multicore graphics solutions in modern smartphones and tablets. It's a very odd decision on Intel's prt.
Anyhow, the failure of Intel's chips in the smartphone and tablet space has little to do with graphics, and more to do with the fact that Atom performs similar to a Cortex A8 or A9, and yet early Atom chips used two or three times as much power to do that. Intel has narrowed the gap quite a bit, although they're still not there yet. Their next-gen parts might achieve this, but much like other architectures have had trouble displacing Intel for the desktop/notebook crown, so too will Intel have trouble displaying ARM in the embedded space. Merely matching the performance-per-watt of ARM's chips isn't enough, because at that point, people will ask "what's the point of using Intel's chips over ARM, they perform the same but don't have as big an install base so there are no advantages."
The only way Intel will get anywhere is by doing something BETTER than ARM, and they haven't managed that quite yet.
True, but there's also the tendency from researchers to try to make their discoveries seem as important as possible. They'll discover something new and report the benefits it could theoretically bring under optimal conditions, when those conditions might never actually be obtained during real-world use.
Sure, I thought that was implied, since the cases of new engine licensees for id tech 4 is rather minimal, let alone 3 (Urban Terror probably paid almost nothing considering the age of the engine). And any new id tech 4 games that come out are probably from licenses that predate the takeover anyhow.
But I've never seen anything saying Zenimax would not allow licensing. id was bought in 2009, and at QuakeCon 2009 Carmack was saying he intended to petition Zenimax to release the id tech 4 source when Rage came out, and from what we've heard from him, he never really met with any opposition. He even was surprised to get an enthusiastic response from their legal department about the benefits to releasing older source. I can't find any articles that mention a negative response toward this policy by Zenimax.
Really?
http://en.wikipedia.org/wiki/Quake_4
It came out six years ago. Done by Raven. Not a great game, but not a bad one either. Decent, is the word I'd use. Multiplayer was fun, an update of Quake 3. It can be purchased from a variety of digital distribution networks, such as Steam or Impulse ($20 each), or from online stores like Amazon ($8 for mac version, $38 for PC)
Shadowmaps can still look worse than stencil shadows in many situations. Especially if the stencil shadows are not of sufficiently high resolution to produce smooth edges.
On the other hand, stencil shadows can do soft shadowing a heck of a lot easier.
No, they didn't, they simply said that they would stop licensing id engines for games that are not published by Zenimax. If third party developers want to use an id engine, they need to publish through Zenimax.
... they'll fit right into the steady curve of slowly but steadily increasing battery capacity. People assume that all these battery advancements we keep hearing about never pan out. Well, some of them do, but once the researchers silly claims are brought down to be a bit more realistic, and after the years go by before they actually hit the market, they're just incremental improvements on what was available before they came out.
There's nothing wrong with that.
but why would you be licensing only a binary under BSD? What would be the point of that?
Android is a pretty good example, if we're willing to put the ASL in the place of the BSD here. You can get your hands on a bunch of ASL-licensed binaries on your handset, but nobody is obligated to give you the source.
One of the events that prompted the founding of the FSF was when Stallman wanted to fix a problem with a printer driver, but was forbidden from getting the source. That's not all that different from wanting to fix a problem with your Android handset, but being forbidden from getting the source.
So if I have a BSD-licensed binary, but not the source or any way to get the source, it's not opensource, or free software? If the BSD is only sometimes a free software or opensource license, it's not really free or open at all.
Well, I always just assumed while watching it that the movie was *set* in Vancouver, although it wasn't. Anyhow, being made in Canada certainly influenced the production, but it's ultimately up to the director/producers/etc to make decisions. You've got movies made in Canada that have no indication that they were. Source Code was made here in Montreal, but I had no idea until the credits rolled. That's the way it normally is; we watch a movie, and only when we start seeing a lot of French Canadian names in the credits does it become obvious the movie was made locally.
Some of those US TV series, though, were basically Canadian shows in everything but budget. Stargate SG-1, for example. My understanding is that the owners paid the budget, but almost all of the writers/producers/executive producers/etc were Canadian. So we like to take credit for shows like that ;)
Please don't blame Canada for that movie. We didn't write it, we didn't direct it, we didn't produce it. We just provided facilities and most of the cast and crew, not the creative team that actually made the film.
Sorry, my notebook only comes with one keyboard, I can't "turn off" the noise the switches make.
The founder of the Free Software Foundation appears to agree with me:
- You have the freedom to run the program as you wish, for any purpose.
- You have the freedom to modify the program to suit your needs. (To make this freedom effective in practice, you must have access to the source code, since making changes in a program without having the source code is exceedingly difficult.)
- You have the freedom to redistribute copies, either gratis or for a fee.
- You have the freedom to distribute modified versions of the program, so that the community can benefit from your improvements.
The Apache and BSD licenses violate the second criteria for free software licenses, they do not grant you the freedom to modify the program to suit your needs, because they do not grant you access to the source. You can throw around the binaries to an Apache or BSD licensed program yourself, but nobody is obligated to give you the source. In that way, they're no different than most other freeware. Heck, some of Microsoft's shared source licenses are more open than them.
Let's look at RMS's own definitions of free software
- You have the freedom to run the program as you wish, for any purpose.
- You have the freedom to modify the program to suit your needs. (To make this freedom effective in practice, you must have access to the source code, since making changes in a program without having the source code is exceedingly difficult.)
- You have the freedom to redistribute copies, either gratis or for a fee.
- You have the freedom to distribute modified versions of the program, so that the community can benefit from your improvements.
According to RMS himself, founder of the FSF, the Apache and BSD licenses are not free software licenses because they specifically violate the second value; the freedom to modify, which requires source code.
There's a tremendous amount of FUD being spread about key competitors. As you've noted, the "Android's not really free/libre" is one of the FUD talking points.
Uh, well, it's not, really. The Apache license (as well as BSD licenses) make no requirement that the source be provided, so they're not really "free software" licenses in the "freedom" sense... They're certainly not opensource licenses since they make no requirement that the source be open. They're basically just free-as-in-beer distribution licenses; you can share the binaries all you want, but you can't really change them, because nobody is obligated to give you the source to go with those binaries.
The article mentions price, but claims 35 GBP for 100Mbps...
Right now, I'm paying the equivalent of 75 GBP for 50Mbps, so to me, 35 seems super cheap.
I built a ZFS-based home file server a few years ago using OpenSolaris for ZFS support. However, OpenSolaris is dead, and I was never all that comfortable with it anyhow, so I'd like to move to FreeBSD.
Yeah and of course you can't blame Ubuntu for not sticking with Gnome 2. What were their options for their GTK DE then if they didn't care for the direction Gnome was heading? XFCE? I don't like particularly like Unity and maybe they should have just taken advantage of the customizability of Gnome shell and made and maintained a very unique and cool Ubuntu flavor of Gnome 3 but I like the idea of having multiple unique desktop environments out there competing. Maybe in a few years Unity becomes something pretty cool for the standard user...
But Unity isn't a desktop environment. It's a GNOME shell replacement, running on top of GNOME.
Linux Mint's approach for their upcoming release is going to be to migrate to GNOME 3, but supplement/replace GNOME Shell 3 (which looks a lot like Unity) with something that looks a lot more like GNOME Shell 2. In other words, they're going to run the latest and greatest, but tweak and add to it to provide a more traditional UI.
As somebody who is shortly going to be migrating from OpenSolaris (and I never really figured package management out there either) to FreeBSD, could you elaborate a bit on FreeBSD's package management? I assumed that it had something comparable to apt.