Some of the problems with diesel have been solved/reduced over time, but they are still there. Most of the problems are in comparison to gasoline (petrol). If you drive a diesel you will notice these problems.
Low performance - Diesel engines can produce a large amount of torque, this is because as the combustion is taking place, more fuel comes into the camber. This is like 'adding fuel to the fire'. The problem is that all of the torque and power is at much lower RPM than in a gasoline engine. This gives a very poor acceleration in particular. Some manufactures address this with turbo chargers.
ping ping ping ping - This is the sound of the diesel engine. It is particularly bad when you are starting-up and idling. This is caused by the diesel exploding in the cylinder. This has been alleviated in modern engines to some extent with electronic timing of the injection of the fuel, injection under higher pressure, more injectors spaced out strategically in the cylinder head, and 'combustion shaping'. What this all means is that now with computer simulations engineers have learned how to control the combustion better in the cylinder. For example a small amount of fuel is first injected then more. The small amount of fuel injected earlier first combusts lower down in the cylinder and the combustion 'grows-up' toward the fuel later injected. This reduces the explosion that is heard as a ping to some extent.
emissions - In fact you cannot buy a new diesel in states like CA and NY because of them not meeting emissions requirements in those states. The big deal is that diesel fuel in the USA is not low sulfur like in Europe. (sulfur oxides... sulfuric acid in atmosphere... acid rain....) That is to change for cars by the end of 2006 and by the end of 2010 for buses/trains, so maybe in a few years diesels will be able to be sold again in all states. The other thing is particulate emissions. It used to be that when driving a diesel a plum of black smoke followed the car. This has been reduced greatly but still even on modern diesels when you first start moving after the engine was off and under heavy acceleration you can still SEE the black smoke. These particulate emissions contribute to smog.
winter - The cold does not play well with diesel. You will miss how in the winter you can just expect a gasoline car to start. Diesel has a tendency to freeze in your gas lines. This is worse in the USA because at most filling stations the diesel fuel is not a hot commodity so early in the winter they still have summer blend in the tanks. Solutions like block heaters and glow plugs were used in the past but well formulated fuel and electronic ignition has helped a lot here. Regardlessly, if it ever does get really cold your fuel WILL freeze, period. Also, if you are like me in the winter you go out and start the car while getting everything ready to go out for work in the morning. Forget about idling a diesel to get the heater core all toasty warm for you when you get back. Diesels run COLD and idle even colder. Some manufactures are putting in electrically heated seats in both front and rear of the car to offset this!
Okay, there you go from someone that has had experience in the past with diesel. Those I think are the reasons that gasoline is much more 'convenient' than diesel. You will really miss gasoline if you ever go back to diesel for these reasons, I did when I took a trip to Europe last year and was driving diesels again... (Oh and after you filler-up, your hands will REALLY smell and be oily, I forgot about that wonderful diesel-ism....)
I agree, from what I have read, JPEG2000 is very different from JPEG and not only in the transform. But if indeed they have dropped the Huffman encoder for an arithmetic encoder instead, the standard is patent encumbered. From "Arithmetic coding":
However, IBM and other companies own patents in the United States and other countries on algorithms essential for implementing an arithmetic encoder.
The thing is that possibly the companies like IBM, Canon, and Hitachi (if I remember correctly) have given free licenses to those implementing the JPEG2000 spec. Then it would be patent encumbered, but could be a non-issue for F/OSS.
Since you seem somewhat of an expert in the JPEG2000 spec could you elaborate on the above? Also, are there options for using Huffman or range encoders in lieu of the arithmetic encoder in the spec? And when you implemented JPEG2000, where did you go for the details? Is there something analogous to libjpeg6 from the IJG that is a reference implementation of JPEG2000 available?
It tells about how you can use an nvramrc to change graphic-options on your mac to enable monitor spanning (as apposed to monitor mirroring). It works on Radeon 7500 eMacs but it is unclear yet whether this trick will work on these new eMacs. Let's hope so....
In the last place I worked, we had exactly the set-up I described. In such a system most software that used autoconf works fantastically. You can make a subdir for x86 and one for sparc, cd into that dir, do../configure -exec-prefix=foo -prefix=bar, then make; make install. Then login on the other system and build over nfs.
Then of course it is very simple to have a Makefile in the dir above all of your source that remembers what options to configure to use. You can even have that apply patches if you need for really trick code.
The place I work now, we use AFS here and I really liked the comment about @sys from the AC.
So I guess if you were using binary packages or something having all of the versions prebuilt would be nice, but that would not play too well over the network anyway for anything that uses dynamically linked libraries (almost everything compiled correctly). For most binary file formats you need to specify at build-time where to find the libraries. Most binary packages assume using a tree under something like/,/usr/local,/opt/sw./sw, or something. If you just export one of these schemes as-is over a network filesystem, then the local admins cannot install packages on their workstation any longer. If you decide to put them someplace-else then you need to play with environment variables like LD_LIBRARYPATH and now you really cannot do that too easily for everyone now can you...
(But then again, on the FreeBSD boxes here we DO export/usr/local. With ports it is so insanely easy to keep everything up-to-date, it is like everyhting gets upgraded for everyone automatically. For stuff not in the ports tree, we just use our home directories which are exported over NFS. But this is all so easy for the FreeBSD boxes because they are all x86 of course...)
Trust me, the scheme I described is not anywhere near perfect but it does work better than I imagined at first.
The.app directory idea is gross but it seems 'nifty' to some. This is personal taste and seems to go hand-in-hand with how a person feels abot xml files versus dot files.
I've got an OSX install on purely UFS, and sure enough, it allows you to pack x86 and PPC binaries (or multiple PPC/X86 binaries, for optimization/bitness) into the same *.app so you can have one application file that executes on multiple architectures. It might not be Apple's hacked-up old kludgy way to get a 'fat binary' but it's effectively the same result but done MUCH cleaner and capable of living on many diverse file-systems.
Okay but that is not really necessary on Darwin anyway because it uses Mach-O instead of something like ELF (most modern UNIX-likes) or XCOFF (basically what the PPC data fork code really was prior to MacOS X) and this allows the same binary FILE to have copies for various architectures in it. Check out:
otool -arch
This came from NeXT too. What.app directories do is they have a Contents subdirectory for the various OS/architecture combinations. And that is where the.app should be useful, you can have an application that runs on both MacOS 9/X. But my opinion is that I do not think that is worth it, others disagree. What I think would have been niftier here would have been some fantastic implementation of extended attributes in the filesystem and to use that like the resource/data forks of yore, but most people would think I was a heretic of some sort for thinking that.
Imagine how cool it would be to have ONE shared 'applications' folder mounted read-only on all your clients, the x86 clients execute the x86 code from camino.app and the PPC machines execute the PPC code from the same place. It would be an administrator's utopia!
Did you know about the ARCH variable and the automounter? Do a man automount on solaris say. This is how you can create a map in NIS for/foosw say, where/foosw/bin is different for sparc and x86 while/foosw/include are the same say. Then you have dirs like/export/foosw/bin-x86,/export/foosw/bin-sparc,/export/foosw/include (or you may like to use a structure like/export/foosw/x86/ and/export/foosw/sparc/ with symlinks pointing up a dir for common stuff) which you export over NFS. On solaris check-out isaexec,
isalist, and friends to see how to have different optimized verions of the same binary. (The trick there is with subdirs like sparcv9 etc.) Each other OS (and sometimes it is a compiler-toolchain provided trick) handles this in its own way. You can even have optimized dynamic libraries, in elf just link with the appropriate -R options creating special dirs for the different targets. In solaris you may be able to be even more nifty about this all. Do this sometime on a recent solaris box:
pargs -x $$
Take a look at AT_SUN_PLATFORM. Now do:
elfdump -d/usr/lib/libc.so
Take a look at
AUXILIARY
and this should give you an idea of how to do something similar.
Anyway, the thing you wish for has been solved a long time ago, and in a more clean fashion, without resorting to treating applications like directories.
whatever happened to the uDevGames2003 source?
on
Mac Contest Roundup
·
· Score: 3, Interesting
I heard about uDevGames2003 a long while ago. After the contest the source code and binaries of all of the games was supposed to have been made availble. I just took a look and I cannot find it anywhere on the idevgames site. The closest I could get was this press realease. The Downloads page is essentially empty save for the 21days readme file. Am I just looking in the wrong place? Probably I should use the Contacts page and get a hold of someone there, but maybe another slashdot reader knows so that I do not look stupid twice. I do remember that the idevgames web site 'looked' a whole lot different last year, so it seems someone is maintaining and updating it.
The things you mention are basically correct, but they really only apply to modern systems. All this A20 gate keyboard controller stuff dates back to when the first AT systems appeared. Also I believe the terminology for the various modes of Windows is 'correct' ie what Microsoft called it back then, but maybe I remember wrong so many years later...
So what was the deal. Addressing changed a great deal when the 286 came out. In summarized form, essentially you had 24 bit addressing instead of 20 bit addressing. But take an address XT of 0x10ffef (chosen because this was the same as FFFF:FFFF) for example. But you say, "What!? 0x10ffef is more than 0x100000 and you said the XT had 20 bit addressing?" Exactly, that address of 0x10ffef would wrap to 0xffef. Well the 286 in real mode did not behave in such a manner. Why would this matter, well really in 8086 you addressed 16 bits into one of four segments, ugh. So programmers used this wrap-around trick to get to two regions of memory far apart from one another without having to change ES or DS say.
Such a trick did not work on 286 real mode and it was actually used a lot, say to get to the ROM Bios Data Area to find out how much memory was installed on the system. When the AT was being developed they were using a new keyboard controller, the 8042, which had a spare pin on it. So the designers of the AT decided to put that between the processor and the memory. On the 8042 there was a gate the could be on or off, and depending on its state it would either let the memory see the A20 (address line 20 bit) as is from the processor or set it to zero. This was an ugly hack needless to say and when the first 386 systems were coming out there were various bugs in how some motherboards were made that prevented the memory from seeing the correct addreses through the keyboard controller that went unnoticed initially because no one was using protected mode yet.
I am not sure if I remember this acurately but if you do a search for make money fast you now get two sponsored links at the top under the 'Web banner.' I think that in the past google would put the ads to the side in a yellow box. Please tell me that I am simply rememberint this incorrectly, and that my mind was just trained by the old google to avoid those links. If that is the case, then the same will happen again and I will not notice them again soon.
I have not checked those links but one thing got me just now. I looked for Michael Eisner and I do not think that I got the Michael Eisner from Disney that we hear about here all the time. The thing is that I remember a story about FundRace on either NPR or PRI (possibly MarketPlace) and I am almost certain that the story claimed that Disney's Michael Eisner contributed $2000 to the Kerry campaign. It now seems that the reporter just did a search for 90210 just to be cute and see what turned-up and did not check the facts too carefully about what was reported.
Also, I did a neighbor search and "Mona Lisa" turned-up. I know the people who live at the address and they are an elderly retired couple. The husband started a company and his daughter now runs it. I did some more searches on the family and at least the daughter turned up, but with a bogus address. Both had maximum contributions. This seems fishy. Shouldn't real names and addresses be used? If this is wrong, how do you report things like this?
I really cannot believe it. The last time I must have used gopher was probably sometime in the first half of the '90s. I did not even think that anyone was still using gopher, let alone the protocol was supported in the browser!!! I clicked the link in Safari and up popped IE and there were the files. It is somewhat eire that MS IE supports gopher though....
(Just so that you all do not think I am some sort of freak, maybe the fact that I seem so excited in this post has something to do that I have been here at work since 6:30 and I am in the middle of 32 oz of coffee, or maybe I am just a freak after all.)
Are there still any good places to check out with gopher?
Yeah we used to use the phone, but only for one extra player. The guy would play from work during his owl shift:) Anyway, the DM would just roll for him. There was no concern about honesty since he trusted the DM. (Also everyone else was there and could see the DM roll so that could have helped with the DM being honest.) For the stuff we played, we never saw maps. The DM would always explain what we were seeing.
You might try these options to the sample encoder:
--audio-rate-target
--video-rate-target
--audio-quality
--video-quality
This is not two-pass but it may be close enough. I always seem to need to some fudge with XVID using mencoder so this might end-up being as close as you need in practice for now.
Here is a more complete excerpt. This is how he explained how he was able to approximate the root so quickly:
The number was 1729.03. I happened to know that a cubic foot contains 1728 cubic inches, so the answer is a tiny bit more than 12. The excess, 1.03 is only one part in nearly 2000, and I had learned in calculus that for small fractions, the cube root's excess is one-third of the number's excess. So all I had to do is find the fraction 1/1728, and multiply by 4 (divide by 3 and multiply by 12). So I was able to pull out a whole lot of digits that way.
They are zip files, just extract them and follow the instructions inside. But are you really going to trust some links on slashdot? The way I found these is I went to sunsolve and searched for 108993 and 113476 in PatchFinder.
Agreed, I cannot tell you how horrible it was to change all of my scripts to use the pwd command instead of the pwd shell built-in all because bash disguised as sh likes to treat symlinks 'special'.
I am talking about awk, grep, diff (still no unified diffs!) and the like. The default shells -- sh and csh -- do not even allow for command line editing. make is outdated. vi borks if you extend your xterm too wide.
Yes I have once run into a bug in awk with escaping characters on solaris, the thing is that you get three awk commands, so I just used a different one.
What is the bug in grep that you allude to? Sure GNU grep has more options, but GNU loves to name them --foobarbaz so that you 'need' command line history and editing.
Solaris supports unified diffs since version 8 or 9 but people like me accustomed to solaris can't wrap our heads arround it and like context diffs better anyway. But patch files on the net are usually in unified format... Personally I dislike the bug in in GNU diff that does not give minimal diffs unless you ask it to (option --minimal believe it or not) so it really does not behave like the shortest-common-subsequence diff algorithm that is taught as the example for dynamic programming at university.
No command line editing in the default shells? I personally like ksh on solaris, maybe the up arrow does not work by default or something, but it uses similar commands that the shell on the embedded systems I maintain do.
I agree the vi included in solaris is krufty and old, but it will get you out a bind if you need it and prefer it to using ed. Good luck using vim or emacs when you are in trouble. Vim comes on the freeware CD as well, so it is a breeze to install. The same is true for all of the other examples you gave.
This is an article about security, why was this troll fed? Moderators?
'so-1997-and-1994' should give a hint that these are not rampant vulnerabilities in solaris, though there are what -- probably six or so vulnerabilities a year related to PAM on various systems. I think the submitter was not bashing solaris, so much as pointing out a vulnerability that so far has only showed-up in CIAC (a little read and usually slow security news bulletin run by the DOE) and patches with their associated docs from Sun.
The thing is that authentication and PAM is just so complicated that effectively exploits of the passwd command show up with some regularity. A more seasoned admin than I here at work commented that this feels like 1982 all over again. I do agree that if there was a similar story like this submitted about linux, it would probably be rejected and not make front page news, though.
Here is a post from the IWW news mailing list that includes the text of an article from the Seattle Post-Intelligencer about the case. In fact the details dealt with butchers rather than deli workers (a deli may sell precut/prepared meats exclusively thus not requiring butchers).
There was a guy I knew, he did not use macs so I do not know how much this applies, but he had a studio that he set-up when he was in California. From what I saw and what he explained to me, it was set-up where all the analog instruments, mixers, mics, etc fed into boxes that sent digitized signals to a computer that was on the other side of a wall. The boxes sent the digitized audio over usb and optical connections. The reason for this set-up was because the computer was loud, and with it being in another room, the mics would not pick it up in the studio. With essentially a hole in the wall, he fed his mouse, keyboard, and monitor cable to the other room on the other side of the wall. He also used the special boxes (I do not remember what company made them) to digitize and send the audio that way to the computer, because he learned that using a board in the computer itself was a bad idea because of all the electrical (hissing and pops) noise that crept in. He noted that the optical connection was the best because there was no possible way for electrical noise to interfere.
I wish I could remember more details, I am not a musician. I thought that this might help though. I am a computer professional and it is probably a bad idea to put things like speaker dampening material in/around your case like others seem to have suggested. You really do not want to interfere with the cooling system for of computer unless you do not want it last long.
I use DVD Backup to copy a DVD to my iBook when I take a trip but do not want to take my original DVD with me. For a thousand DVDs you will need more than a terabyte of storage, but you should be able to setup a machine to serve that over nfs maybe with a few mounts. Hook-up a mac to your plasma screen and use the DVD Player included with MacOS X to play your movies. DVD Player has a menu item 'File -> Open VIDEO_TS Folder..." that does the trick. Plus you can script DVD Player with applescript, so you can quickly hack something together that lets you choose the movie you wish to play. Then you can navigate the usual DVD menus as you wish. You can get a wireless keyboard and mouse to make navigation from your couch easier.
Of course, neither do the numerous pentium3 (And 4) based beowulf clusters out there.
Interesting, I thought that ever since the i80x87 line of chips you had 80-bit floating point. It even has support for many rounding modes. This is superior precision to the IEEE 64-bit spec.
Please also understand that the 6.2 GFlops you cite is a theoretical peak, while you state that your dual P3 does not even get 1 GFlop this was surely with some lame test you ran. The P3 using SSE can do much better than what you claim.
There was a serious paper presented in 1962 titled The One-Way Manned Space Mission by John M. Cord and Leonard M. Seale. It described sending a Mercury sized capsule to the moon with a single astronaut. The capsule would have a cylindrical container attached to it with supplies. Before and during the mission more capsules and cylinders with supplies would be launched to the landing site. The astronaut was to use the empty cylinders as shelter. Later a two man expedition would be sent to retrieve the original astronaut when sufficiently powerful rockets were developed.
Maybe this current plan for Mars is just a similar situation where in the eagerness of the moment some wild ideas like this get tossed about until technology catches-up.
The idea for the moon mission lead to the novel The Pilgrim Project by Hank Searls which lead to the movie
Countdown directed by Robert Altman (of M*A*S*H fame) starring James Caan and Robert Duvall which was eclipsed by a certain other movie set in space released shortly after this one.
Slashdot Mirror
Some of the problems with diesel have been solved/reduced over time, but they are still there. Most of the problems are in comparison to gasoline (petrol). If you drive a diesel you will notice these problems.
Low performance - Diesel engines can produce a large amount of torque, this is because as the combustion is taking place, more fuel comes into the camber. This is like 'adding fuel to the fire'. The problem is that all of the torque and power is at much lower RPM than in a gasoline engine. This gives a very poor acceleration in particular. Some manufactures address this with turbo chargers.
ping ping ping ping - This is the sound of the diesel engine. It is particularly bad when you are starting-up and idling. This is caused by the diesel exploding in the cylinder. This has been alleviated in modern engines to some extent with electronic timing of the injection of the fuel, injection under higher pressure, more injectors spaced out strategically in the cylinder head, and 'combustion shaping'. What this all means is that now with computer simulations engineers have learned how to control the combustion better in the cylinder. For example a small amount of fuel is first injected then more. The small amount of fuel injected earlier first combusts lower down in the cylinder and the combustion 'grows-up' toward the fuel later injected. This reduces the explosion that is heard as a ping to some extent.
emissions - In fact you cannot buy a new diesel in states like CA and NY because of them not meeting emissions requirements in those states. The big deal is that diesel fuel in the USA is not low sulfur like in Europe. (sulfur oxides... sulfuric acid in atmosphere... acid rain....) That is to change for cars by the end of 2006 and by the end of 2010 for buses/trains, so maybe in a few years diesels will be able to be sold again in all states. The other thing is particulate emissions. It used to be that when driving a diesel a plum of black smoke followed the car. This has been reduced greatly but still even on modern diesels when you first start moving after the engine was off and under heavy acceleration you can still SEE the black smoke. These particulate emissions contribute to smog.
winter - The cold does not play well with diesel. You will miss how in the winter you can just expect a gasoline car to start. Diesel has a tendency to freeze in your gas lines. This is worse in the USA because at most filling stations the diesel fuel is not a hot commodity so early in the winter they still have summer blend in the tanks. Solutions like block heaters and glow plugs were used in the past but well formulated fuel and electronic ignition has helped a lot here. Regardlessly, if it ever does get really cold your fuel WILL freeze, period. Also, if you are like me in the winter you go out and start the car while getting everything ready to go out for work in the morning. Forget about idling a diesel to get the heater core all toasty warm for you when you get back. Diesels run COLD and idle even colder. Some manufactures are putting in electrically heated seats in both front and rear of the car to offset this!
Okay, there you go from someone that has had experience in the past with diesel. Those I think are the reasons that gasoline is much more 'convenient' than diesel. You will really miss gasoline if you ever go back to diesel for these reasons, I did when I took a trip to Europe last year and was driving diesels again... (Oh and after you filler-up, your hands will REALLY smell and be oily, I forgot about that wonderful diesel-ism....)
The thing is that possibly the companies like IBM, Canon, and Hitachi (if I remember correctly) have given free licenses to those implementing the JPEG2000 spec. Then it would be patent encumbered, but could be a non-issue for F/OSS.
Since you seem somewhat of an expert in the JPEG2000 spec could you elaborate on the above? Also, are there options for using Huffman or range encoders in lieu of the arithmetic encoder in the spec? And when you implemented JPEG2000, where did you go for the details? Is there something analogous to libjpeg6 from the IJG that is a reference implementation of JPEG2000 available?
Thanks...
I do not know about Princeton but the JET Experiment reached Breakeven in 1997. Spot-on about the neutron production issues though.
http://www.rutemoeller.com/mp/ibook/ibook_e.html
It tells about how you can use an nvramrc to change graphic-options on your mac to enable monitor spanning (as apposed to monitor mirroring). It works on Radeon 7500 eMacs but it is unclear yet whether this trick will work on these new eMacs. Let's hope so....
In the last place I worked, we had exactly the set-up I described. In such a system most software that used autoconf works fantastically. You can make a subdir for x86 and one for sparc, cd into that dir, do ../configure -exec-prefix=foo -prefix=bar, then make; make install. Then login on the other system and build over nfs.
/, /usr/local, /opt/sw. /sw, or something. If you just export one of these schemes as-is over a network filesystem, then the local admins cannot install packages on their workstation any longer. If you decide to put them someplace-else then you need to play with environment variables like LD_LIBRARYPATH and now you really cannot do that too easily for everyone now can you...
/usr/local. With ports it is so insanely easy to keep everything up-to-date, it is like everyhting gets upgraded for everyone automatically. For stuff not in the ports tree, we just use our home directories which are exported over NFS. But this is all so easy for the FreeBSD boxes because they are all x86 of course...)
Then of course it is very simple to have a Makefile in the dir above all of your source that remembers what options to configure to use. You can even have that apply patches if you need for really trick code.
The place I work now, we use AFS here and I really liked the comment about @sys from the AC.
So I guess if you were using binary packages or something having all of the versions prebuilt would be nice, but that would not play too well over the network anyway for anything that uses dynamically linked libraries (almost everything compiled correctly). For most binary file formats you need to specify at build-time where to find the libraries. Most binary packages assume using a tree under something like
(But then again, on the FreeBSD boxes here we DO export
Trust me, the scheme I described is not anywhere near perfect but it does work better than I imagined at first.
Okay but that is not really necessary on Darwin anyway because it uses Mach-O instead of something like ELF (most modern UNIX-likes) or XCOFF (basically what the PPC data fork code really was prior to MacOS X) and this allows the same binary FILE to have copies for various architectures in it. Check out:
This came from NeXT too. WhatDid you know about the ARCH variable and the automounter? Do a man automount on solaris say. This is how you can create a map in NIS for /foosw say, where /foosw/bin is different for sparc and x86 while /foosw/include are the same say. Then you have dirs like /export/foosw/bin-x86, /export/foosw/bin-sparc, /export/foosw/include (or you may like to use a structure like /export/foosw/x86/ and /export/foosw/sparc/ with symlinks pointing up a dir for common stuff) which you export over NFS. On solaris check-out isaexec,
isalist, and friends to see how to have different optimized verions of the same binary. (The trick there is with subdirs like sparcv9 etc.) Each other OS (and sometimes it is a compiler-toolchain provided trick) handles this in its own way. You can even have optimized dynamic libraries, in elf just link with the appropriate -R options creating special dirs for the different targets. In solaris you may be able to be even more nifty about this all. Do this sometime on a recent solaris box:
Take a look at AT_SUN_PLATFORM. Now do:Take a look at and this should give you an idea of how to do something similar.Anyway, the thing you wish for has been solved a long time ago, and in a more clean fashion, without resorting to treating applications like directories.
I heard about uDevGames2003 a long while ago. After the contest the source code and binaries of all of the games was supposed to have been made availble. I just took a look and I cannot find it anywhere on the idevgames site. The closest I could get was this press realease. The Downloads page is essentially empty save for the 21days readme file. Am I just looking in the wrong place? Probably I should use the Contacts page and get a hold of someone there, but maybe another slashdot reader knows so that I do not look stupid twice. I do remember that the idevgames web site 'looked' a whole lot different last year, so it seems someone is maintaining and updating it.
The things you mention are basically correct, but they really only apply to modern systems. All this A20 gate keyboard controller stuff dates back to when the first AT systems appeared. Also I believe the terminology for the various modes of Windows is 'correct' ie what Microsoft called it back then, but maybe I remember wrong so many years later...
So what was the deal. Addressing changed a great deal when the 286 came out. In summarized form, essentially you had 24 bit addressing instead of 20 bit addressing. But take an address XT of 0x10ffef (chosen because this was the same as FFFF:FFFF) for example. But you say, "What!? 0x10ffef is more than 0x100000 and you said the XT had 20 bit addressing?" Exactly, that address of 0x10ffef would wrap to 0xffef. Well the 286 in real mode did not behave in such a manner. Why would this matter, well really in 8086 you addressed 16 bits into one of four segments, ugh. So programmers used this wrap-around trick to get to two regions of memory far apart from one another without having to change ES or DS say.
Such a trick did not work on 286 real mode and it was actually used a lot, say to get to the ROM Bios Data Area to find out how much memory was installed on the system. When the AT was being developed they were using a new keyboard controller, the 8042, which had a spare pin on it. So the designers of the AT decided to put that between the processor and the memory. On the 8042 there was a gate the could be on or off, and depending on its state it would either let the memory see the A20 (address line 20 bit) as is from the processor or set it to zero. This was an ugly hack needless to say and when the first 386 systems were coming out there were various bugs in how some motherboards were made that prevented the memory from seeing the correct addreses through the keyboard controller that went unnoticed initially because no one was using protected mode yet.
I am not sure if I remember this acurately but if you do a search for make money fast you now get two sponsored links at the top under the 'Web banner.' I think that in the past google would put the ads to the side in a yellow box. Please tell me that I am simply rememberint this incorrectly, and that my mind was just trained by the old google to avoid those links. If that is the case, then the same will happen again and I will not notice them again soon.
Also, I did a neighbor search and "Mona Lisa" turned-up. I know the people who live at the address and they are an elderly retired couple. The husband started a company and his daughter now runs it. I did some more searches on the family and at least the daughter turned up, but with a bogus address. Both had maximum contributions. This seems fishy. Shouldn't real names and addresses be used? If this is wrong, how do you report things like this?
I really cannot believe it. The last time I must have used gopher was probably sometime in the first half of the '90s. I did not even think that anyone was still using gopher, let alone the protocol was supported in the browser!!! I clicked the link in Safari and up popped IE and there were the files. It is somewhat eire that MS IE supports gopher though....
(Just so that you all do not think I am some sort of freak, maybe the fact that I seem so excited in this post has something to do that I have been here at work since 6:30 and I am in the middle of 32 oz of coffee, or maybe I am just a freak after all.)
Are there still any good places to check out with gopher?
Yeah we used to use the phone, but only for one extra player. The guy would play from work during his owl shift :) Anyway, the DM would just roll for him. There was no concern about honesty since he trusted the DM. (Also everyone else was there and could see the DM roll so that could have helped with the DM being honest.) For the stuff we played, we never saw maps. The DM would always explain what we were seeing.
You might try these options to the sample encoder: --audio-rate-target --video-rate-target --audio-quality --video-quality This is not two-pass but it may be close enough. I always seem to need to some fudge with XVID using mencoder so this might end-up being as close as you need in practice for now.
113476
They are zip files, just extract them and follow the instructions inside. But are you really going to trust some links on slashdot? The way I found these is I went to sunsolve and searched for 108993 and 113476 in PatchFinder.
Agreed, I cannot tell you how horrible it was to change all of my scripts to use the pwd command instead of the pwd shell built-in all because bash disguised as sh likes to treat symlinks 'special'.
I am talking about awk, grep, diff (still no unified diffs!) and the like. The default shells -- sh and csh -- do not even allow for command line editing. make is outdated. vi borks if you extend your xterm too wide.
Yes I have once run into a bug in awk with escaping characters on solaris, the thing is that you get three awk commands, so I just used a different one.
What is the bug in grep that you allude to? Sure GNU grep has more options, but GNU loves to name them --foobarbaz so that you 'need' command line history and editing.
Solaris supports unified diffs since version 8 or 9 but people like me accustomed to solaris can't wrap our heads arround it and like context diffs better anyway. But patch files on the net are usually in unified format... Personally I dislike the bug in in GNU diff that does not give minimal diffs unless you ask it to (option --minimal believe it or not) so it really does not behave like the shortest-common-subsequence diff algorithm that is taught as the example for dynamic programming at university.
No command line editing in the default shells? I personally like ksh on solaris, maybe the up arrow does not work by default or something, but it uses similar commands that the shell on the embedded systems I maintain do.
I agree the vi included in solaris is krufty and old, but it will get you out a bind if you need it and prefer it to using ed. Good luck using vim or emacs when you are in trouble. Vim comes on the freeware CD as well, so it is a breeze to install. The same is true for all of the other examples you gave.
This is an article about security, why was this troll fed? Moderators?
The thing is that authentication and PAM is just so complicated that effectively exploits of the passwd command show up with some regularity. A more seasoned admin than I here at work commented that this feels like 1982 all over again. I do agree that if there was a similar story like this submitted about linux, it would probably be rejected and not make front page news, though.
correct link
Here is a post from the IWW news mailing list that includes the text of an article from the Seattle Post-Intelligencer about the case. In fact the details dealt with butchers rather than deli workers (a deli may sell precut/prepared meats exclusively thus not requiring butchers).
There was a guy I knew, he did not use macs so I do not know how much this applies, but he had a studio that he set-up when he was in California. From what I saw and what he explained to me, it was set-up where all the analog instruments, mixers, mics, etc fed into boxes that sent digitized signals to a computer that was on the other side of a wall. The boxes sent the digitized audio over usb and optical connections. The reason for this set-up was because the computer was loud, and with it being in another room, the mics would not pick it up in the studio. With essentially a hole in the wall, he fed his mouse, keyboard, and monitor cable to the other room on the other side of the wall. He also used the special boxes (I do not remember what company made them) to digitize and send the audio that way to the computer, because he learned that using a board in the computer itself was a bad idea because of all the electrical (hissing and pops) noise that crept in. He noted that the optical connection was the best because there was no possible way for electrical noise to interfere.
I wish I could remember more details, I am not a musician. I thought that this might help though. I am a computer professional and it is probably a bad idea to put things like speaker dampening material in/around your case like others seem to have suggested. You really do not want to interfere with the cooling system for of computer unless you do not want it last long.
I use DVD Backup to copy a DVD to my iBook when I take a trip but do not want to take my original DVD with me. For a thousand DVDs you will need more than a terabyte of storage, but you should be able to setup a machine to serve that over nfs maybe with a few mounts. Hook-up a mac to your plasma screen and use the DVD Player included with MacOS X to play your movies. DVD Player has a menu item 'File -> Open VIDEO_TS Folder..." that does the trick. Plus you can script DVD Player with applescript, so you can quickly hack something together that lets you choose the movie you wish to play. Then you can navigate the usual DVD menus as you wish. You can get a wireless keyboard and mouse to make navigation from your couch easier.
Interesting, I thought that ever since the i80x87 line of chips you had 80-bit floating point. It even has support for many rounding modes. This is superior precision to the IEEE 64-bit spec.
Please also understand that the 6.2 GFlops you cite is a theoretical peak, while you state that your dual P3 does not even get 1 GFlop this was surely with some lame test you ran. The P3 using SSE can do much better than what you claim.
Are you a Sony fanboy by any chance :)
Maybe this current plan for Mars is just a similar situation where in the eagerness of the moment some wild ideas like this get tossed about until technology catches-up.
The idea for the moon mission lead to the novel The Pilgrim Project by Hank Searls which lead to the movie Countdown directed by Robert Altman (of M*A*S*H fame) starring James Caan and Robert Duvall which was eclipsed by a certain other movie set in space released shortly after this one.
Here is another good article from The Austrailian with more details. The pressure rose from 13.92 psi to 14.11 psi when the leak was sealed.