MS DOS -- policy is that the running application gets ALL resources. Of course that is "fast". But, if there are more resources than the program uses, these resources are wasted (while that program runs).
Windows (up to 3) -- Event driven graphics library and some utilities on top of MS-DOS. Generally, the underlying policy is that of MS DOS, but co-operative sharing is supported. Resource control is not imposed. Since co-operation is not enforced, performance can be at the "MS DOS" level. Windows 3.1 also supports v86 (crossover with Windows 9x).
Windows 95, 98 -- some forced resource control (underlying v86 "virtual" mode). Above that level, co-operation needed. Programs may not execute quite as fast as MS DOS, but (1) 32 bit is available, and (2) resource sharing and control is possible.
Windows 2000/XP -- Complete resource control, could even extend to multi-user. Again, MS DOS could be faster, but only if interrupt driven i/o, and resource management was written into the DOS application (generally not the case). A "real" filesystem with performance support. Finally, matches Unix in terms of "raw" features.
Windows VISTA -- Evolution of 2000/XP. Doesn't offer additional resource control for the user (UAC is arguable). Does offer controls for "content vendors".
The first version of Windows that could possibly be compared to Unix (Linux) is 2000/XP. Resource control is similar. On this basis, Linux offers a POSIX standard API, multiple filesystems, and good driver support. 2000/XP offers a non-standard API, a single "reasonable" filesystem, and poor driver support. The user interface in 2000/XP is very polished; Linux supports command line interface for most features, which end-users view as a negative. Linux performance is carefully nurtured, with a great deal of research and testing (continual refinement of memory management, processor management, i/o management). Windows offers MAJOR code drops (VISTA) claimed to increase performance, or add features. There is no (or very little) stepwise refinement, or open research. This has made Linux directly competitive with Solaris while Windows offers a more polished end-user experience.
There is NO ADDITIONAL FUNCTIONALITY in an OS. Resource control. If the GUI is included, it is the effective and speedy support of the GUI API (Graphics Programming Model). The standard here, of course, is OpenGL. Is Linux OpenGL support worse than Windows? Possibly, because Windows is the first-tier support for ATI and nVidia. However, MESA is still the standard software renderer for OpenGL. This means that Windows has an advantage in utilizing GPU capability; I can't comment on the management of those features. The (in my opinion) largest driver for VISTA release was sharing GPU capability to allow a "3D desktop" with 3D applications. Linux incorporates compositing, and actually premiered these features before VISTA. However, I find the ATI drivers to not be usable (on an IBM T43 laptop). The nVidia driver is usable, but its resource control cannot be examined. Of course, Microsoft follows a different drummer (little/no standards support), so XP and VISTA use "DirectX" natively.
To be more specific, on the T43, if compositing is used to allow a "3D desktop", xv (hardware video acceleration/scaling) results in a "black window", and OpenGL applications "flicker" rendering them unusable. The desktop compositing has to be disabled before running these applications. Using MESA as the OpenGL engine is possible, but the applications then do not exploit the hardware. This is a big LOSE for ATI, the T43 and Linux.
Which means VISTA wins, based on graphics resource control. The other features do not need more processing (memory, cpu resource control), and VISTA is claimed to have superior caching (as compared to XP). So VISTA should be faster than XP on the T43. I have not personally tried it (POSIX is far to important to me), but here is a comment from the web (google "T43 VISTA"):
"All of the MS specific apps are faster. IE7 is almost
You can't lay the wires or optical -- that's a government grant (allows the phone and cable companies to dig up non-subscribers property to lay the lines). And, believe me, you WON'T get that right.
Which leaves satellite or wireless... and that won't get you "unthrottled super speed".
This IS a monopoly. And should, as such, be government regulated.
Well, I guess it's kind of transparent... but Windows does NOT support older hardware better. It *may* support old device drivers better, but, since these are (almost) universally closed source, changing the kernel will cause worse support. Especially for the IHVs that are now out of business.
BSD and Linux both support open drivers; it is more likely that drivers from BSD will be ported into the Windows ecosystem to provide that support. Unfortunately, VISTA needs driver signing, and I suspect (85% confidence) that WIN7 will also need this. Making it prohibitive for the hobbyist to provide the support.
Of course this doesn't matter, because the Windows adoption model is driven by NEW hardware acquisition; this is the "sheer" fact generally forgotten.
If WIN7 wants to penetrate into that software ecosystem (the one currently ruled by Linux) it will have to provide much better driver support than any previous version of Windows. If WIN7 want to penetrate the enterprise ecosystem, it will have to offer features that compete with Solaris (much more inspection and control, think dtrace and prset). If WIN7 wants to penetrate the existing Windows market, it needs more drivers. The actual ecosystem I think WIN7 wants to occupy is the new system market (which Microsoft owns). Of course WIN7 doesn't have any opinions -- its the design direction for the product.
Maybe Microsoft is afraid of new system market saturation. Maybe the growing market is in cheap subcompact laptops (OLPC, eeepc). VISTA is not targeted there; XP is barely functional. Is that where WIN7 is going? I *really* don't see a head to head fight against Linux on the old desktop arena, or Solaris/Linux in the datacenter (sure, maybe some token activity, but nothing serious).
"Why? Because it doesn't matter what operating system, brand of computer, or programming language you use."
Them's Fightin' Words... I sentence you to COBOL on IBM 360 DOS.
Come on, it does matter - a programming language is also a way of thinking. As I tell most of my students, go read SICP, do some Scheme and get back to me on that.
As to brand of computer... no, BRAND doesn't matter, unless, of course, you like Apple. Operating system? Isn't that a commodity? If it doesn't come with source, it shouldn't be used. My god, even VAX VMS came with source. Microsoft likes putting in "back compatibility" fixes into the OS core, in order to "protect" the OS investment. I guess you LIKE using an OS whose only standard API is actively poo-pooed by the prime vendor?
"The API-W will not provide any value for our customers or the industry" Hugo Lunardelli, Open Systems and Standards Manager, Microsoft Europe
I guess having API-W would benefit WINE more than Windows developers, right? And Microsoft will be diligent in supporting API "errors" into the future (or forking the API)?
I guess that POSIX really IS meaningless, and being able to support multiple systems from multiple vendors is lame, right?
I don't care. If I get paid, I'll be your bitch. Windows has made me a lot of money over the years.
Encrypting the base image is actually a serious security problem. With typical operating environments, you are giving a potential attacker hundreds of megabytes of known plaintext that has been encrypted. There are known chosen-plaintext attacks on AES, and giving attackers vast quantities of known plaintext is not a good idea.
A direct quote from the article. I guess, in a sense, this is true. It is true that it is impossible to ensure that a lock can be obtained. On a hardware level, no less.
On the other hand, this statement is a load of bullshit.
"Weren't the trade sanctions against Cuba put there and don't they remain there in part because of Cuban human rights abuses?"
Human rights abuses? What the fuck?!? Are you claiming that Batista did better? Are you claiming that Castro tortured and killed over 20,000 during his reign?
"Legal gambling outfits in the US follow strict gambling laws that regulate, among other things, machine calibration, payout ratios, etc."
Only on a state level. But, inter-state gambling is allowed. Are you impugning Antigua's regulatory controls?
These are very strong indictments. Of course the US spin is "human rights (vs. supporting a regime that KILLED citizens), and Regulation (vs. allowing inter-state gambling)". THAT is what the world is PISSED OFF about.
"I don't think an FTP server is quite the same as a distributed network of client resources."
No, its the same as a distributed network of FTP servers, with a single client coordinating transfers. Like I said, you have to reverse "client" and "server" (because the patent gets it the wrong way -- its a computation SERVER carrying on a service for the CLIENT).
And there you have; its morphologically the same. Allow me to quote from "man ftp"
proxy ftp-command
Execute an ftp command on a secondary control connection.
This command allows simultaneous connection to two remote ftp
servers for transferring files between the two servers. The
first proxy command should be an open, to establish the sec-
ondary control connection. Enter the command "proxy ?" to
see other ftp commands executable on the secondary connec-
tion. The following commands behave differently when pref-
aced by proxy: open will not define new macros during the
auto-login process, close will not erase existing macro defi-
nitions, get and mget transfer files from the host on the
primary control connection to the host on the secondary con-
trol connection, and put, mput, and append transfer files
from the host on the secondary control connection to the host
on the primary control connection. Third party file trans-
fers depend upon support of the ftp protocol PASV command by
the server on the secondary control connection.
FTP - a server (from which the client FTP is run), with communications channels between each of the servers (and the servers to each other). A separate dedicated communications channel which tells each server what to do, and gathers results.
FTP uses port 20 for data, and port 21 for control (the separate dedicated channel).
Of course the definition of "client" and "server" is simply reversed from the patent, and the resource is the existence, or desired existence of a data file.
So, yeah, FTP qualifies.
Now, FTP does not actually use a "dedicated" communication channel -- normally the channel is multiplexed using the standard port numbering system. But it behaves that way.
The actual CONTENT -- that is, video and audio encoding; is (or can be) identical between the two formats. The only difference is actually pressing the disks. The music industry supported multiple formats for years (cassette, CD) for the SAME content.
Now, having more capacity or features with one format may induce the producers to actually support those features. For example, near the end, Paramount was "net-enabling" HD DVD releases. Now that it is all Bluray, there is no need to do that (or desire) anymore.
How much for a player. You claim that the player pricing should be high, because they are expensive. But the only difference between a Bluray player and an HD DVD player is a (cheap) focus mechanism. Bluray players SHOULD be priced at $199 now (the pre-dumping price of the Venture HD DVD player). At Walmart. I will grant a $50 premium for "winning".
on filesystems periodically. This erases all data (although some information may still be recovered from the inodes, and unused parts of existing file blocks). On both hard drives and flash media.
It may take a long time (on either media).
"Unerasing" old drives can be very fruitful -- normally the machines are dumped and resold as bottom-end used computers. The vendor that does this service tries to save money by doing the "refurb" as quickly as possible - no full format, etc.
The remaining parts are: inodes that indicate ownership, size even though the file has been deleted (the filesize could be used as a marker, to give a good idea that a particular file was, in the past, on this computer. For example, the invasion plans, or other document. This can also give a good idea WHO or WHEN a computer was used. When a directory is updated, the filename may not be removed, giving a clue as to contents. This is why full disk encryption is preferred over file-by-file encryption.
"Wear protection" has nothing do with it. If the flash drive has a capacity of N bytes, and N bytes are written, every byte has been used -- even if you could not predict the order the bytes were used in. Wear protection is simply a function randomizing the write order, relative to the usual usage pattern. Since most Operating Environments try to write sequentially (in order to increase performance on magnetic media), and have "hot" areas (think allocation tables and directories) the randomizer is fairly easy to generate and can be quite effective. But the wear protection doesn't magically create capacity (which is what you implied).
Now, it may be that simply zeroing a drive would allow the recovery of the previous state of the drive, but the contents could be randomized. The big issue is the time taken to do this. Easier just to smash the drive, (or, more likely, incinerate it).
I agree with you. Further, I will state that the Toshiba A3/A30 are about the "right price" now (maybe a bit less expensive than they could be). Along with a "7 free movie" offer, an under $100 price point is good. Current BD player prices are completely insane.
And, HD DVD content is being dumped (ei. at reasonable prices). So, it's HD DVD for me, baby.
Also, I can burn HD content on 3xDVD (DVD5 or DVD9) and have my Toshiba play it in glorious HD. I haven't (yet) heard of success in reading DVD9 HD content on a (Sony) Blu Ray player -- can anyone actually report success doing this (the players apparently don't switch into 3x mode, but stay in 1x)? As it is, I can burn 20 to 40 minutes of full 1080i/1080p HD onto DVD media, and play using 3xDVD mode (a movies can be recorded onto 2 or 3 dual-layer DVDs).
I would like to have someone verify that this works on a BD player as well.
I run a wifi hotspot (free). A lot of other people do. My idea is to "mesh the Americas". I don't use anywhere close to my available bandwidth, and I don't mind sharing. In return, I like others to share.
Does Ericsson make money from this? Of course not. 10 Euros a day is a LOT of money for bandwidth. But, for me, its not just the money. I feel good by supplying this service.
Am I going to stop providing wifi? Not unless it is declared illegal. I am not rich enough to provide cell service in the city... although I had considered purchasing an old analog cell tower and becoming the phone company for a cottage area -- the cottagers around a small lake. The plan was to become a co-op phone company offering internet and analog cell service to 50(ish) people. The "stopper" was trying to provide adequate tech support (just not possible, at any price point).
The framerate we are concerned with (this is a video card), is the framerate used to drive the output device. Therefore HD encoder is correct (we are not interested in decoding a stream here, we are encoding the frame buffer to the TV). TV sets follow specific standards, and my original comment holds.
Computer Monitors, on the other hand, are driven in a whole bunch of (wacky) modes -- some VESA defined, some TV spec, others... well, with the advent of "autosyncing", the sky is the limit. Some TV sets (with VGA, DVI, or HDMI inputs) can accept some of these wacky modes as well...
To the issue of tearing/dropping -- dropping can only happen if source material cannot be fed to the video card at the needed rate. Not a problem with modern video cards (may be an issue with HD source material, and some busses). Tearing? Haven't seen it in YEARS. Indicates that there is a memory controller issue.
XvMC? Yes, I did get the letters discombobulated. Sorry. Its driver/hardware assist for motion vector compensation. Not terribly useful these days, and, I will agree that the nVidia driver displayed some anomolies with it, using a 5200 FX, when I last tried it (2 years ago).
As to overscan. Broken? Maybe. TVs do it. If I am trying to display to a TV set, overscan should be accomodated. Most of the MythTV themes DO NOT.
What is a "good video framerate"? Video is 30 fps. No faster, no slower.
The 5200 FX is able to display SD (standard definition) video with no problems. Of course, cards of this class do not have HD encoders.
The 6000 and up series is able to do HD (high definition) video with no problems. I am using a 7300 (AGP 4x bus interface) to do 1080i display (the machine I am typing on, which happens to be my PVR). I am not sure if the card will drive 1080p, but that isn't a "mode" that my TV will do.
The card/driver does have XV support, which means that "video overlays" and yuv conversion is handled in the card/driver (and is most likely accelerated). MCxV is also supported (but I tend not to use that).
No problems (stability or performance) -- except for software that doesn't understand the concept of overscan, and wants to place things too close to the display edge. Window managers that don't understand 30 fps interlaced, and try to draw large horizontal single pixel lines (resulting in flicker). But these aren't nVidia's issues. As a system integrator, I deal with these problems (my value add when building MythTV boxes).
I try to support AMD (ATI) but nVidia has had an enormous advantage in the Linux sphere for years (things tend to "just work").
Not sure about VISTA. But, take XP PRO SP2. DVD playback software? No, not on the distribution CD. I installed something I purchased called "Power DVD".
Linux (Fedora) at least had something I could download for free ("non-free" programs, a version of mplayer that plays DVDs). Power DVD actually cost me real money.
Now, I suspect that HP/Dell/??? supplies that DVD software with the computer (along with a bunch of other software). I, however, started with a retail copy of XP PRO SP2.
Enforcing 1:1 is a Ponzi scheme -- yes, early downloaders can get it, but it requires INCREASING numbers of later downloaders to maintain. Which is not going to happen. Which is why you are keeping torrents open for weeks to months.
A bittorrent client will pick a piece "at random" to download, and that may well be uploadable right away. So, participation is in the swarm is almost immediate and effective, even if you do NOT wait around for the 1:1.
Seriously, though, NDIS is contained, but a tuner card interacts with the multimedia stack. Not anywhere as contained. It could be done, but the "wrapper" would carry the bulk of WINE with it.
And, it would have to wrap something -- the current multimedia architecture in Linux isn't stable enough to target.
Yes, I would have expected Microsoft to have a compatibility layer -- given that such a layer is the ONLY way to use certain devices moving forward. After all, you don't have source to port forward if needed. Nothing would have prevented the inclusion of the old interfaces.
As an example of doing this the right way, look at the Linux transition from a.out to elf, or oss to alsa.
Put the libraries into a directory, and set LD_PATH to point to that directory. Ensure that your kernel actually supports a.out format. Normally, you would create/linux10/bin/linux10/lib. Create a linux10run script in/usr/local/bin that executes its $0 argument as/linux10/bin/$0 with the reset LD_PATH. Of course, configuration (/etc) may give you a bit of trouble, but usually not for a game (/etc/hosts,/etc/passwd are the same).
The X server is compatible, so it's not a problem. If the application used direct i/o, it already has suid; may be a security risk, but not much of a problem (virtual console support may be, just don't switch away from it once running). If the game uses direct i/o, it (90% confidence) won't run under VMware.
Can't think of much else that would be a problem. Except, of course, the building of the old linux environment. You may have to download and install an antique linux. If you have SATA disks, this will be problematic. Install the antique linux on VMware Server, with IDE (a simple environment) -- and then copy the required/lib and/usr/lib.so files (you can copy both to/linux10/lib, and, given that disk space is fairly cheap, you can copy them all).
If/etc gives you a problem, consider chroot'ing the thing.
But... using this functionality for webkit, obstensibly to not affect performance of some apps...
Setting the control globally for the app (the approved approach) results in fast UGLY drawing, and the controlled setting is a private API, allowing for: nice drawing, with a speed bost when needed.
What is not available is the combination of features.
You mean the Blu Ray VC1 encoder that Microsoft make money from? Thats really hands off.
Yes, Blu Ray has additional space, *but* it has region codes. And, with BD+ key revocations and player bricking are now easy.
On another note:
HD DVD compliance means DVD5/DVD9 HD DVD (3XDVD) support as well. (a standard way to have high def content in on regular DVDs). Which makes it easy to "roll your own", without an HD DVD burner. Now, supposedly, the BD-S300 player (the BD player from WalMart) will play BD9 (the Blu Ray player equivalent), but I don't think anyone has it working yet (the drive continues 1x operation rather than going into 3x mode). The cheap Toshibas play 3XDVD just fine (from any cheap DVD-R media).
*IF* anyone has the BD-S300 and has generated and sucessfully played a BD9 on it, let me know -- I am very curious. I am thinking of stockpiling a few A3 (or A30) players for this reason.
Thank you for your post. Yes, I hold the opinion that Toshiba was bought off. I will hold on to my A3 player, and I hope that Studio Canal will continue pressing movies for it.
Now, I am looking forward to the content firesales as Blockbuster, WalMart et. al. start dumping HD DVD media stock.
Slashdot Mirror
MS DOS -- policy is that the running application gets ALL resources. Of course that is "fast". But, if there are more resources than the program uses, these resources are wasted (while that program runs).
Windows (up to 3) -- Event driven graphics library and some utilities on top of MS-DOS. Generally, the underlying policy is that of MS DOS, but co-operative sharing is supported. Resource control is not imposed. Since co-operation is not enforced, performance can be at the "MS DOS" level. Windows 3.1 also supports v86 (crossover with Windows 9x).
Windows 95, 98 -- some forced resource control (underlying v86 "virtual" mode). Above that level, co-operation needed. Programs may not execute quite as fast as MS DOS, but (1) 32 bit is available, and (2) resource sharing and control is possible.
Windows 2000/XP -- Complete resource control, could even extend to multi-user. Again, MS DOS could be faster, but only if interrupt driven i/o, and resource management was written into the DOS application (generally not the case). A "real" filesystem with performance support. Finally, matches Unix in terms of "raw" features.
Windows VISTA -- Evolution of 2000/XP. Doesn't offer additional resource control for the user (UAC is arguable). Does offer controls for "content vendors".
The first version of Windows that could possibly be compared to Unix (Linux) is 2000/XP. Resource control is similar. On this basis, Linux offers a POSIX standard API, multiple filesystems, and good driver support. 2000/XP offers a non-standard API, a single "reasonable" filesystem, and poor driver support. The user interface in 2000/XP is very polished; Linux supports command line interface for most features, which end-users view as a negative. Linux performance is carefully nurtured, with a great deal of research and testing (continual refinement of memory management, processor management, i/o management). Windows offers MAJOR code drops (VISTA) claimed to increase performance, or add features. There is no (or very little) stepwise refinement, or open research. This has made Linux directly competitive with Solaris while Windows offers a more polished end-user experience.
There is NO ADDITIONAL FUNCTIONALITY in an OS. Resource control. If the GUI is included, it is the effective and speedy support of the GUI API (Graphics Programming Model). The standard here, of course, is OpenGL. Is Linux OpenGL support worse than Windows? Possibly, because Windows is the first-tier support for ATI and nVidia. However, MESA is still the standard software renderer for OpenGL. This means that Windows has an advantage in utilizing GPU capability; I can't comment on the management of those features. The (in my opinion) largest driver for VISTA release was sharing GPU capability to allow a "3D desktop" with 3D applications. Linux incorporates compositing, and actually premiered these features before VISTA. However, I find the ATI drivers to not be usable (on an IBM T43 laptop). The nVidia driver is usable, but its resource control cannot be examined. Of course, Microsoft follows a different drummer (little/no standards support), so XP and VISTA use "DirectX" natively.
To be more specific, on the T43, if compositing is used to allow a "3D desktop", xv (hardware video acceleration/scaling) results in a "black window", and OpenGL applications "flicker" rendering them unusable. The desktop compositing has to be disabled before running these applications. Using MESA as the OpenGL engine is possible, but the applications then do not exploit the hardware. This is a big LOSE for ATI, the T43 and Linux.
Which means VISTA wins, based on graphics resource control. The other features do not need more processing (memory, cpu resource control), and VISTA is claimed to have superior caching (as compared to XP). So VISTA should be faster than XP on the T43. I have not personally tried it (POSIX is far to important to me), but here is a comment from the web (google "T43 VISTA"):
"All of the MS specific apps are faster. IE7 is almost
You can't lay the wires or optical -- that's a government grant (allows the phone and cable companies to dig up non-subscribers property to lay the lines). And, believe me, you WON'T get that right.
Which leaves satellite or wireless... and that won't get you "unthrottled super speed".
This IS a monopoly. And should, as such, be government regulated.
Well, I guess it's kind of transparent... but Windows does NOT support older hardware better. It *may* support old device drivers better, but, since these are (almost) universally closed source, changing the kernel will cause worse support. Especially for the IHVs that are now out of business.
BSD and Linux both support open drivers; it is more likely that drivers from BSD will be ported into the Windows ecosystem to provide that support. Unfortunately, VISTA needs driver signing, and I suspect (85% confidence) that WIN7 will also need this. Making it prohibitive for the hobbyist to provide the support.
Of course this doesn't matter, because the Windows adoption model is driven by NEW hardware acquisition; this is the "sheer" fact generally forgotten.
If WIN7 wants to penetrate into that software ecosystem (the one currently ruled by Linux) it will have to provide much better driver support than any previous version of Windows. If WIN7 want to penetrate the enterprise ecosystem, it will have to offer features that compete with Solaris (much more inspection and control, think dtrace and prset). If WIN7 wants to penetrate the existing Windows market, it needs more drivers. The actual ecosystem I think WIN7 wants to occupy is the new system market (which Microsoft owns). Of course WIN7 doesn't have any opinions -- its the design direction for the product.
Maybe Microsoft is afraid of new system market saturation. Maybe the growing market is in cheap subcompact laptops (OLPC, eeepc). VISTA is not targeted there; XP is barely functional. Is that where WIN7 is going? I *really* don't see a head to head fight against Linux on the old desktop arena, or Solaris/Linux in the datacenter (sure, maybe some token activity, but nothing serious).
Just my 2 cents.
"Why? Because it doesn't matter what operating system, brand of computer, or programming language you use."
Them's Fightin' Words... I sentence you to COBOL on IBM 360 DOS.
Come on, it does matter - a programming language is also a way of thinking. As I tell most of my students, go read SICP, do some Scheme and get back to me on that.
As to brand of computer... no, BRAND doesn't matter, unless, of course, you like Apple. Operating system? Isn't that a commodity? If it doesn't come with source, it shouldn't be used. My god, even VAX VMS came with source. Microsoft likes putting in "back compatibility" fixes into the OS core, in order to "protect" the OS investment. I guess you LIKE using an OS whose only standard API is actively poo-pooed by the prime vendor?
"The API-W will not provide any value for our customers or the industry"
Hugo Lunardelli, Open Systems and Standards Manager,
Microsoft Europe
I guess having API-W would benefit WINE more than Windows developers, right? And Microsoft will be diligent in supporting API "errors" into the future (or forking the API)?
I guess that POSIX really IS meaningless, and being able to support multiple systems from multiple vendors is lame, right?
I don't care. If I get paid, I'll be your bitch. Windows has made me a lot of money over the years.
Encrypting the base image is actually a serious security problem. With typical operating environments, you are giving a potential attacker hundreds of megabytes of known plaintext that has been encrypted. There are known chosen-plaintext attacks on AES, and giving attackers vast quantities of known plaintext is not a good idea.
"Threads are inherently non-deterministic"
A direct quote from the article. I guess, in a sense, this is true. It is true that it is impossible to ensure that a lock can be obtained. On a hardware level, no less.
On the other hand, this statement is a load of bullshit.
"Weren't the trade sanctions against Cuba put there and don't they remain there in part because of Cuban human rights abuses?"
Human rights abuses? What the fuck?!? Are you claiming that Batista did better? Are you claiming that Castro tortured and killed over 20,000 during his reign?
"Legal gambling outfits in the US follow strict gambling laws that regulate, among other things, machine calibration, payout ratios, etc."
Only on a state level. But, inter-state gambling is allowed. Are you impugning Antigua's regulatory controls?
These are very strong indictments. Of course the US spin is "human rights (vs. supporting a regime that KILLED citizens), and Regulation (vs. allowing inter-state gambling)". THAT is what the world is PISSED OFF about.
"I don't think an FTP server is quite the same as a distributed network of client resources."
No, its the same as a distributed network of FTP servers, with a single client coordinating transfers. Like I said, you have to reverse "client" and "server" (because the patent gets it the wrong way -- its a computation SERVER carrying on a service for the CLIENT).
And there you have; its morphologically the same. Allow me to quote from "man ftp"
proxy ftp-command
Execute an ftp command on a secondary control connection.
This command allows simultaneous connection to two remote ftp
servers for transferring files between the two servers. The
first proxy command should be an open, to establish the sec-
ondary control connection. Enter the command "proxy ?" to
see other ftp commands executable on the secondary connec-
tion. The following commands behave differently when pref-
aced by proxy: open will not define new macros during the
auto-login process, close will not erase existing macro defi-
nitions, get and mget transfer files from the host on the
primary control connection to the host on the secondary con-
trol connection, and put, mput, and append transfer files
from the host on the secondary control connection to the host
on the primary control connection. Third party file trans-
fers depend upon support of the ftp protocol PASV command by
the server on the secondary control connection.
FTP - a server (from which the client FTP is run), with communications channels between each of the servers (and the servers to each other). A separate dedicated communications channel which tells each server what to do, and gathers results.
FTP uses port 20 for data, and port 21 for control (the separate dedicated channel).
Of course the definition of "client" and "server" is simply reversed from the patent, and the resource is
the existence, or desired existence of a data file.
So, yeah, FTP qualifies.
Now, FTP does not actually use a "dedicated" communication channel -- normally the channel is multiplexed using the standard port numbering system. But it behaves that way.
Your analysis is fascinating...
The actual CONTENT -- that is, video and audio encoding; is (or can be) identical between the two formats. The only difference is actually pressing the disks. The music industry supported multiple formats for years (cassette, CD) for the SAME content.
Now, having more capacity or features with one format may induce the producers to actually support those features. For example, near the end, Paramount was "net-enabling" HD DVD releases. Now that it is all Bluray, there is no need to do that (or desire) anymore.
How much for a player. You claim that the player pricing should be high, because they are expensive. But the only difference between a Bluray player and an HD DVD player is a (cheap) focus mechanism. Bluray players SHOULD be priced at $199 now (the pre-dumping price of the Venture HD DVD player). At Walmart. I will grant a $50 premium for "winning".
But, old versions of the files may be in the unused space. I usually recommend running
/dev/urandom
# possibly use
dd=if=/dev/zero of=freespace bs=512; rm freespace
on filesystems periodically. This erases all data (although some information may still be recovered from the inodes, and unused parts of existing file blocks). On both hard drives and flash media.
It may take a long time (on either media).
"Unerasing" old drives can be very fruitful -- normally the machines are dumped and resold as bottom-end used computers. The vendor that does this service tries to save money by doing the "refurb" as quickly as possible - no full format, etc.
The remaining parts are: inodes that indicate ownership, size even though the file has been deleted (the filesize could be used as a marker, to give a good idea that a particular file was, in the past, on this computer. For example, the invasion plans, or other document. This can also give a good idea WHO or WHEN a computer was used. When a directory is updated, the filename may not be removed, giving a clue as to contents. This is why full disk encryption is preferred over file-by-file encryption.
"Wear protection" has nothing do with it. If the flash drive has a capacity of N bytes, and N bytes are written, every byte has been used -- even if you could not predict the order the bytes were used in. Wear protection is simply a function randomizing the write order, relative to the usual usage pattern. Since most Operating Environments try to write sequentially (in order to increase performance on magnetic media), and have "hot" areas (think allocation tables and directories) the randomizer is fairly easy to generate and can be quite effective. But the wear protection doesn't magically create capacity (which is what you implied).
Now, it may be that simply zeroing a drive would allow the recovery of the previous state of the drive, but the contents could be randomized. The big issue is the time taken to do this. Easier just to smash the drive, (or, more likely, incinerate it).
Preach on!
I agree with you. Further, I will state that the Toshiba A3/A30 are about the "right price" now (maybe a bit less expensive than they could be). Along with a "7 free movie" offer, an under $100 price point is good. Current BD player prices are completely insane.
And, HD DVD content is being dumped (ei. at reasonable prices). So, it's HD DVD for me, baby.
Also, I can burn HD content on 3xDVD (DVD5 or DVD9) and have my Toshiba play it in glorious HD. I haven't (yet) heard of success in reading DVD9 HD content on a (Sony) Blu Ray player -- can anyone actually report success doing this (the players apparently don't switch into 3x mode, but stay in 1x)? As it is, I can burn 20 to 40 minutes of full 1080i/1080p HD onto DVD media, and play using 3xDVD mode (a movies can be recorded onto 2 or 3 dual-layer DVDs).
I would like to have someone verify that this works on a BD player as well.
Ericsson, eh?
I run a wifi hotspot (free). A lot of other people do. My idea is to "mesh the Americas". I don't use anywhere close to my available bandwidth, and I don't mind sharing. In return, I like others to share.
Does Ericsson make money from this? Of course not. 10 Euros a day is a LOT of money for bandwidth. But, for me, its not just the money. I feel good by supplying this service.
Am I going to stop providing wifi? Not unless it is declared illegal. I am not rich enough to provide cell service in the city... although I had considered purchasing an old analog cell tower and becoming the phone company for a cottage area -- the cottagers around a small lake. The plan was to become a co-op phone company offering internet and analog cell service to 50(ish) people. The "stopper" was trying to provide adequate tech support (just not possible, at any price point).
And I'm going to hold my breath until I get them! ..blue..bluer..bluest..
29.97 to be specific.
The framerate we are concerned with (this is a video card), is the framerate used to drive the output device. Therefore HD encoder is correct (we are not interested in decoding a stream here, we are encoding the frame buffer to the TV). TV sets follow specific standards, and my original comment holds.
Computer Monitors, on the other hand, are driven in a whole bunch of (wacky) modes -- some VESA defined, some TV spec, others... well, with the advent of "autosyncing", the sky is the limit. Some TV sets (with VGA, DVI, or HDMI inputs) can accept some of these wacky modes as well...
To the issue of tearing/dropping -- dropping can only happen if source material cannot be fed to the video card at the needed rate. Not a problem with modern video cards (may be an issue with HD source material, and some busses). Tearing? Haven't seen it in YEARS. Indicates that there is a memory controller issue.
XvMC? Yes, I did get the letters discombobulated. Sorry. Its driver/hardware assist for motion vector compensation. Not terribly useful these days, and, I will agree that the nVidia driver displayed some anomolies with it, using a 5200 FX, when I last tried it (2 years ago).
As to overscan. Broken? Maybe. TVs do it. If I am trying to display to a TV set, overscan should be accomodated. Most of the MythTV themes DO NOT.
What is a "good video framerate"? Video is 30 fps. No faster, no slower.
The 5200 FX is able to display SD (standard definition) video with no problems. Of course, cards of this class do not have HD encoders.
The 6000 and up series is able to do HD (high definition) video with no problems. I am using a 7300 (AGP 4x bus interface) to do 1080i display (the machine I am typing on, which happens to be my PVR). I am not sure if the card will drive 1080p, but that isn't a "mode" that my TV will do.
The card/driver does have XV support, which means that "video overlays" and yuv conversion is handled in the card/driver (and is most likely accelerated). MCxV is also supported (but I tend not to use that).
No problems (stability or performance) -- except for software that doesn't understand the concept of overscan, and wants to place things too close to the display edge. Window managers that don't understand 30 fps interlaced, and try to draw large horizontal single pixel lines (resulting in flicker). But these aren't nVidia's issues. As a system integrator, I deal with these problems (my value add when building MythTV boxes).
I try to support AMD (ATI) but nVidia has had an enormous advantage in the Linux sphere for years (things tend to "just work").
Not sure about VISTA. But, take XP PRO SP2. DVD playback software? No, not on the distribution CD. I installed something I purchased called "Power DVD".
Linux (Fedora) at least had something I could download for free ("non-free" programs, a version of mplayer that plays DVDs). Power DVD actually cost me real money.
Now, I suspect that HP/Dell/??? supplies that DVD software with the computer (along with a bunch of other software). I, however, started with a retail copy of XP PRO SP2.
I cannot comment on VISTA at this time...
Enforcing 1:1 is a Ponzi scheme -- yes, early downloaders can get it, but it requires INCREASING numbers of later downloaders to maintain. Which is not going to happen. Which is why you are keeping torrents open for weeks to months.
A bittorrent client will pick a piece "at random" to download, and that may well be uploadable right away. So, participation is in the swarm is almost immediate and effective, even if you do NOT wait around for the 1:1.
The wrapper for your tuner is called VMware....
Seriously, though, NDIS is contained, but a tuner card interacts with the multimedia stack. Not anywhere as contained. It could be done, but the "wrapper" would carry the bulk of WINE with it.
And, it would have to wrap something -- the current multimedia architecture in Linux isn't stable enough to target.
Yes, I would have expected Microsoft to have a compatibility layer -- given that such a layer is the ONLY way to use certain devices moving forward. After all, you don't have source to port forward if needed. Nothing would have prevented the inclusion of the old interfaces.
As an example of doing this the right way, look at the Linux transition from a.out to elf, or oss to alsa.
Put the libraries into a directory, and set LD_PATH to point to that directory. Ensure that your kernel actually supports a.out format. Normally, you would create /linux10/bin /linux10/lib. Create a linux10run script in /usr/local/bin that executes its $0 argument as /linux10/bin/$0 with the reset LD_PATH. Of course, configuration (/etc) may give you a bit of trouble, but usually not for a game (/etc/hosts, /etc/passwd are the same).
/lib and /usr/lib .so files (you can copy both to /linux10/lib, and, given that disk space is fairly cheap, you can copy them all).
/etc gives you a problem, consider chroot'ing the thing.
The X server is compatible, so it's not a problem. If the application used direct i/o, it already has suid; may be a security risk, but not much of a problem (virtual console support may be, just don't switch away from it once running). If the game uses direct i/o, it (90% confidence) won't run under VMware.
Can't think of much else that would be a problem. Except, of course, the building of the old linux environment. You may have to download and install an antique linux. If you have SATA disks, this will be problematic. Install the antique linux on VMware Server, with IDE (a simple environment) -- and then copy the required
If
Have fun with the game!
But... using this functionality for webkit, obstensibly to not affect performance of some apps...
Setting the control globally for the app (the approved approach) results in fast UGLY drawing, and the controlled setting is a private API, allowing for: nice drawing, with a speed bost when needed.
What is not available is the combination of features.
What IS a "programming superstar"?
Windows applications? Unix applications? System level? Oracle? SAP?
Social Networking sites are one element of your answer. You need to be "LinkedIn" to the circles of your developers.
You mean the Blu Ray VC1 encoder that Microsoft make money from? Thats really hands off.
Yes, Blu Ray has additional space, *but* it has region codes. And, with BD+ key revocations and player bricking are now easy.
On another note:
HD DVD compliance means DVD5/DVD9 HD DVD (3XDVD) support as well. (a standard way to have high def content in on regular DVDs). Which makes it easy to "roll your own", without an HD DVD burner. Now, supposedly, the BD-S300 player (the BD player from WalMart) will play BD9 (the Blu Ray player equivalent), but I don't think anyone has it working yet (the drive continues 1x operation rather than going into 3x mode). The cheap Toshibas play 3XDVD just fine (from any cheap DVD-R media).
*IF* anyone has the BD-S300 and has generated and sucessfully played a BD9 on it, let me know -- I am very curious. I am thinking of stockpiling a few A3 (or A30) players for this reason.
Thank you for your post. Yes, I hold the opinion that Toshiba was bought off. I will hold on to my A3 player, and I hope that Studio Canal will continue pressing movies for it.
Now, I am looking forward to the content firesales as Blockbuster, WalMart et. al. start dumping HD DVD media stock.