I live in Edmonton, Canada. As such, I have to regard phrases like "even below zero (celsius)" with some amusement. Keeping your core warm certainly does help the extremities, but is not a full solution. Walking to school, I remember many times when my hands/ears/feet were freezing while I was sweating inside a huge coat.
I think this likely varies in different people, temperature, wind, activity, etc.. There's lots of situations where I would have been happy to move heat from arm to hand, and when my core was perfectly toasty.
Your best bet for hooking it up to a monitor would be a "component-to-RGB" type box. These are expensive. With a good TV, though, the progressive component signal would be good enough to use the TV as a monitor.
The real problem is the lack of a good boot solution - which won't be solved unless disks can be made to run on the thing.
Look over his articles. Read them. The man is a hardcore gamer. And he knows what he's talking about.
The CS AWP has been nerfed beyond recognition (and is way expensive, as you mention). From earlier articles, it sounds like he played CS earlier on. In any case, I think his article mostly targets CoD (which I haven't played).
While I'm not an excellent player, I agree with him on most of his points. My FPS of choice is Enemy Territory where, as he mentions, the sniper rifle is well balanced by design (and not just by nerfing cost/damage/accuracy/etc..)
Press release is kind of funny...
on
News from Mars
·
· Score: 5, Funny
...featuring famous landmarks on the surface of Mars 'as seen through European eyes'...
I thank those noble European eyes that were sacrificed in order to make this European mission Euro-possible.
It's ironic. By so blatantly highlighting the Euro-ticity of this mission, they sound very American.
Right now I'm building myself a game controller out of my exercise bike. It should yield a reasonable workout, will actually make for a playable game (turn with handlebars, pedal to go faster), and will certainly cost less that $700.
...and the best solution I've found is to let it keep running until you're ready to have it "help out" by "crashing closed every Explorer window it can find".
It's got some other wierd things going on and has some trouble debugging some stuff, but it's still useful enough that I keep it around.
I suppose its usefulness would correlate pretty well with the size of the largest script you run. For most people creating public-oriented web pages, the largest script they write is going to be pretty small - and a debugger is going to be something they pull out rarely. I end up writing some fairly long scripts at times, and I find it's often worth whatever hassles it presents.
In any case, I too might have to try out some of the alternatives listed in the comments.
There are lots of DHTML debuggers out there. Of course alert() is always handy, but having an actual debugger available really helps out.
Visual Studio came with Visual InterDev, which does a passable job with debugging. Even with its flaws, it has saved me a lot of time over the years.
Nobody needs a true debugger. Heck, nobody needs CVS. Nobody really needs profiling. Nobody needs a relational database.
I think, though, that the best programmers will take the time to "sharpen the saw" and become proficient in using all these tools. If you're programming without a debugger, you're not working as efficiently as you could. And yet half the replies here suggest the poster is asking for something unreasonable or unavailable. Bizarre.
Next, you should never rely on JavaScript to do your form validation
You should never rely on it, but you should always have it (if supported by the browser). Web pages that use only round trips for validation are horrible and often incomprehensible. It's just plain bad interface, and the alternative is simple, easy, and produces a far better result.
User clicks submit. You alert "You need to enter the donation amount" and set the focus to the appropriate field.
If you think it would be better to let the submit happen, and return the page with an error at the top saying "Amount not entered - you must enter something in all the red fields!", you're silly. By the way, I've seen exactly that on a web page - apparently their hatred for employing available browser technology meant they couldn't even set the focus to the missing field.
Since you didn't mention it, everyone should at least let you know it exists: MS Visual Interdev has a script debugger.
It isn't a magically good debugger (like some MS debuggers really are) but it works OK. It also benefits from integration with IE (ie, when Explorer encounters a script error, it can sometimes jump right to the debugger).
It also does the IntelliSense thing (to an extent) allowing you to reference styles and objects without remembering their (often silly) names.
...really puts in perspective the rate of change in computers. It's been a long time since I thought about what I was going to use extended memory for, or strategies for getting a block right on a 64K line (for use in DMA) without wasting space. I suppose in a few years, it will sound just as hokey to be thinking about how you were going to connect to a database.
I didn't know anyone on the BBS's that had Watcom (or knew much about it beyond its memory setup), but most of us wanted it (everyone noticed it in the Doom load screens). Perhaps having it available will usher in a new wave of retro programming from my generation.
I would have paid good money for a free Watcom back in about 1992. Well, unless it was free that is, then I would have kept my money. Watcom was big news back then, and seemed to have all the features my Borland C++ 3 lacked.
Is this still a useful product for people? Is the Windows support going to be good enough that it will supplant any of the other development options a Windows user has?
Most importantly, does it support Expanded and Extended memory?
My point I wanted to make was not so much that YOU didn't have the equipment (individual people have all sorts of things, I have arcade sticks hooked up to my PC to play MAME) - the point I was getting at was that that configuration (as well as mine) is rare.
And, despite you being a contrary example, I don't think I'm wrong about this rareness. I'd venture to guess that less than 1% of computer gamers have 4 joypads, and I'd be surprised if even 20% have 2.
Companies like EA do PC ports of a lot of popular sports titles. They're usually sold cheap, and still don't sell nearly what the console versions do.
Why?
Picture you and 3 friends crowded around your PC to play Smash Brothers. Do you have 4 seats at your computer? Do you have 4 gamepads with sufficient buttons? Do you even have 4 USB ports?
Metroid's a little better, and might sell a few copies on the PC, but why would Nintendo dilute the appeal of its flagship unit (the GameCube) in order to sell a few more copies of anything? If it was going to do ports of Smash Brothers, wouldn't it start with a PS2 copy? It would sell lots more...
Until Nintendo doesn't make hardware, it isn't going to be doing a lot of ports.
Until pigs fly, you're not going to see heavy "one machine, many players" on computers. And, yes, I remember Star Control II multiplayer. It's not that multiplayer is unworkable on one computer, just that it isn't nearly as workable as the same thing on a console (or multiplayer over a network).
...is probably somewhat close to what these might end up actually being used for.
I can see these as "convention novelties" wherein the people in attendance are each given one/some, and a prize is given to the person who collects the most losing keys. Convention running people are often fond of games/thingees that invite interactions (especially in narrower industries).
I could see this working, but the keys themselves would have to be very cheap - and would have to be available in logo'ed form.
I like the strategy IBM describes. My strategy is obviously suitable in different places. My intended point was that I didn't figure either would be novel ideas (headlines) to most programmers.
Your benchmarks, on the other hand, are a good headline. Going into a project, you usually have a fair idea of your options for memory management and how long they'll take to implement. However, you don't always have a good grasp of the performance implications - your breakdown is handy.
You have to understand that most software people write isn't like what you're thinking. If you are writing software for a large audience and long term use, you obviously have to be more careful with your strategies. For many apps, though, you don't require this sort of robustness - and you probably aren't going to spend enough effort to do everything well. As such, if you're micro-managing memory then you are likely also creating memory leaks and bugs.
Also, I'm not suggesting you allocate a 30 meg buffer at startup, I'm suggesting you allocate a 30k buffer at startup. Many programs are micro-managing this size of dataset, and it's a waste of time.
or you gotta tell us what OSS projects you've contributed to
I've written a lot of free software. For example, try Jumpman Zero for the Palm (link in sig). If you looked at the code for that, you'd see that I allocate all the space for level data when the program opens and leave it allocated the whole way through. It's a few K spent, but the program never has memory crashes (assuming the Palm has enough memory to start the thing).
I'd suggest this approach only in small footprint sort of apps, or apps where performance footprint or convention means that not much else will likely be running (eg. fullscreen games). In many apps, the memory requirements are consistent enough that total demand is going to be the same either way.
Mostly I just hate people to be doing lots of work in C to save 30k of system memory - and ending up with a buggy program full of memory leaks. Many apps have data sets this small (30k) and yet are spending lots of time/effort managing memory.
Many small programs are no longer memory, or even performance, constrained. As such, a reasonable strategy for a lot of desktop software is to allocate a huge buffer at startup, and do repetitive flushes and complete reloads of data (always using the same pre-allocated buffers).
This is simple to do, and avoids a lot of errors. It's also not much of a headline.
So if anything, Amazon proves click and mortar *doesn't* work.
Amazon lost 149 million last year. Conversely, ToysRUs made 229 million. ToysRUs.com may not have made money, but it is apparently part of a business strategy that is functional. Indigo books made money last year (well, a bit anyways). Barnes & Noble made money last year ($100 million - not too shabby).
Anyways, I just find it entertaining that Amazon is being trotted out as a success in comparison with companies that are actually making money.
..the mean old Dean still won't let the prof enter the class in "Battle of the Bands". It's against school policy - and, besides, what hope does a ragtag bunch of school kids have of winning?
"Free speech" is often trotted out to defend things that don't particularly matter - like depictions of Jesus done in elephant poo. These cases are rightly decided largely as prophylactic; we need to defend the elephant poo speech or soon we'll lose our rights to important speech.
And what is that important speech, what is the speech that is in most need of defense from government? How about speech about that government.
If this site was a parody of a commercial or personal site, I'd say take it down. You need to follow the rules if you're going to say just some everyday thing. But since it's a parody of the site promoting the party in power, I think the call has to be to treat it extremely delicately.
The right to criticize the government needs the widest possible of berths. It's the treasure that "free speech" in general provides the wall around. It is the voice that needs the most protection.
HTA's are about being able to use HTML to create a desktop application, and treating the result as a desktop application (ie, different security arrangement and display). I can imagine how this would be useful for simple apps, especially to programmers accustomed to HTML/Script. HTA's are treated as executable code, and are not (barring an exploit) able to be popped up via web page. They are not connected to the web particularly other than that they share an underlying language, HTML. Regardless of browser, I don't imagine anyone sees them very often.
I think the whole idea went out of favor at MS a long time ago - I haven't seen an HTA article there for a while. They apparently weren't too memorable, the comments I've read thusfar betray no understanding of what they are/were.
The game should have had a commanding tutorial. Perhaps the full version does. Barring that, the best practice is to play as a non-commander (to see what weapons/towers/strategies do, and to hopefully see what a good commander does) and as commander on an empty server.
..is indeed a problem in Savage. It's a problem that's solveable (there's a vote to impeach command), but players often won't bother to vote - kind of like reality.
A smaller threshold (30% or so, with longer vote time) to impeach would largely solve the problem on public servers. Alternatively, command could be restricted to those who scored well in the previous round (or they could be allowed to choose commander). These players will not be horrible commanders, or will at least be able to choose a non-newbie to command. The game also needs to handle "command vacuum" better. For example, requests for funds should be auto-approved.
I guess what I'm saying is that the problems are not so much in concept but in implementation. Like other game features in the past, "team with commander" is something that needs to be ironed out. On the whole, though, it makes for a rewarding gameplay experience - it's a good way for allowing individuals to play together cohesively without being game experts.
Slashdot Mirror
I live in Edmonton, Canada. As such, I have to regard phrases like "even below zero (celsius)" with some amusement. Keeping your core warm certainly does help the extremities, but is not a full solution. Walking to school, I remember many times when my hands/ears/feet were freezing while I was sweating inside a huge coat.
I think this likely varies in different people, temperature, wind, activity, etc.. There's lots of situations where I would have been happy to move heat from arm to hand, and when my core was perfectly toasty.
...though it's not widely available.
Your best bet for hooking it up to a monitor would be a "component-to-RGB" type box. These are expensive. With a good TV, though, the progressive component signal would be good enough to use the TV as a monitor.
The real problem is the lack of a good boot solution - which won't be solved unless disks can be made to run on the thing.
Look over his articles. Read them. The man is a hardcore gamer. And he knows what he's talking about.
The CS AWP has been nerfed beyond recognition (and is way expensive, as you mention). From earlier articles, it sounds like he played CS earlier on. In any case, I think his article mostly targets CoD (which I haven't played).
While I'm not an excellent player, I agree with him on most of his points. My FPS of choice is Enemy Territory where, as he mentions, the sniper rifle is well balanced by design (and not just by nerfing cost/damage/accuracy/etc..)
...featuring famous landmarks on the surface of Mars 'as seen through European eyes'...
I thank those noble European eyes that were sacrificed in order to make this European mission Euro-possible.
It's ironic. By so blatantly highlighting the Euro-ticity of this mission, they sound very American.
Right now I'm building myself a game controller out of my exercise bike. It should yield a reasonable workout, will actually make for a playable game (turn with handlebars, pedal to go faster), and will certainly cost less that $700.
...and the best solution I've found is to let it keep running until you're ready to have it "help out" by "crashing closed every Explorer window it can find".
It's got some other wierd things going on and has some trouble debugging some stuff, but it's still useful enough that I keep it around.
I suppose its usefulness would correlate pretty well with the size of the largest script you run. For most people creating public-oriented web pages, the largest script they write is going to be pretty small - and a debugger is going to be something they pull out rarely. I end up writing some fairly long scripts at times, and I find it's often worth whatever hassles it presents.
In any case, I too might have to try out some of the alternatives listed in the comments.
There are lots of DHTML debuggers out there. Of course alert() is always handy, but having an actual debugger available really helps out.
Visual Studio came with Visual InterDev, which does a passable job with debugging. Even with its flaws, it has saved me a lot of time over the years.
Nobody needs a true debugger. Heck, nobody needs CVS. Nobody really needs profiling. Nobody needs a relational database.
I think, though, that the best programmers will take the time to "sharpen the saw" and become proficient in using all these tools. If you're programming without a debugger, you're not working as efficiently as you could. And yet half the replies here suggest the poster is asking for something unreasonable or unavailable. Bizarre.
Next, you should never rely on JavaScript to do your form validation
You should never rely on it, but you should always have it (if supported by the browser). Web pages that use only round trips for validation are horrible and often incomprehensible. It's just plain bad interface, and the alternative is simple, easy, and produces a far better result.
User clicks submit. You alert "You need to enter the donation amount" and set the focus to the appropriate field.
If you think it would be better to let the submit happen, and return the page with an error at the top saying "Amount not entered - you must enter something in all the red fields!", you're silly. By the way, I've seen exactly that on a web page - apparently their hatred for employing available browser technology meant they couldn't even set the focus to the missing field.
Since you didn't mention it, everyone should at least let you know it exists: MS Visual Interdev has a script debugger.
It isn't a magically good debugger (like some MS debuggers really are) but it works OK. It also benefits from integration with IE (ie, when Explorer encounters a script error, it can sometimes jump right to the debugger).
It also does the IntelliSense thing (to an extent) allowing you to reference styles and objects without remembering their (often silly) names.
...really puts in perspective the rate of change in computers. It's been a long time since I thought about what I was going to use extended memory for, or strategies for getting a block right on a 64K line (for use in DMA) without wasting space. I suppose in a few years, it will sound just as hokey to be thinking about how you were going to connect to a database.
I didn't know anyone on the BBS's that had Watcom (or knew much about it beyond its memory setup), but most of us wanted it (everyone noticed it in the Doom load screens). Perhaps having it available will usher in a new wave of retro programming from my generation.
I would have paid good money for a free Watcom back in about 1992. Well, unless it was free that is, then I would have kept my money. Watcom was big news back then, and seemed to have all the features my Borland C++ 3 lacked.
Is this still a useful product for people? Is the Windows support going to be good enough that it will supplant any of the other development options a Windows user has?
Most importantly, does it support Expanded and Extended memory?
My point I wanted to make was not so much that YOU didn't have the equipment (individual people have all sorts of things, I have arcade sticks hooked up to my PC to play MAME) - the point I was getting at was that that configuration (as well as mine) is rare.
And, despite you being a contrary example, I don't think I'm wrong about this rareness. I'd venture to guess that less than 1% of computer gamers have 4 joypads, and I'd be surprised if even 20% have 2.
Companies like EA do PC ports of a lot of popular sports titles. They're usually sold cheap, and still don't sell nearly what the console versions do.
Why?
Picture you and 3 friends crowded around your PC to play Smash Brothers. Do you have 4 seats at your computer? Do you have 4 gamepads with sufficient buttons? Do you even have 4 USB ports?
Metroid's a little better, and might sell a few copies on the PC, but why would Nintendo dilute the appeal of its flagship unit (the GameCube) in order to sell a few more copies of anything? If it was going to do ports of Smash Brothers, wouldn't it start with a PS2 copy? It would sell lots more...
Until Nintendo doesn't make hardware, it isn't going to be doing a lot of ports.
Until pigs fly, you're not going to see heavy "one machine, many players" on computers. And, yes, I remember Star Control II multiplayer. It's not that multiplayer is unworkable on one computer, just that it isn't nearly as workable as the same thing on a console (or multiplayer over a network).
...is probably somewhat close to what these might end up actually being used for.
I can see these as "convention novelties" wherein the people in attendance are each given one/some, and a prize is given to the person who collects the most losing keys. Convention running people are often fond of games/thingees that invite interactions (especially in narrower industries).
I could see this working, but the keys themselves would have to be very cheap - and would have to be available in logo'ed form.
You call the Verisign guy, and he tells you what to do. I'm not sure why the instructions aren't on the page - just call.
I like the strategy IBM describes. My strategy is obviously suitable in different places. My intended point was that I didn't figure either would be novel ideas (headlines) to most programmers.
Your benchmarks, on the other hand, are a good headline. Going into a project, you usually have a fair idea of your options for memory management and how long they'll take to implement. However, you don't always have a good grasp of the performance implications - your breakdown is handy.
You have to understand that most software people write isn't like what you're thinking. If you are writing software for a large audience and long term use, you obviously have to be more careful with your strategies. For many apps, though, you don't require this sort of robustness - and you probably aren't going to spend enough effort to do everything well. As such, if you're micro-managing memory then you are likely also creating memory leaks and bugs.
Also, I'm not suggesting you allocate a 30 meg buffer at startup, I'm suggesting you allocate a 30k buffer at startup. Many programs are micro-managing this size of dataset, and it's a waste of time.
or you gotta tell us what OSS projects you've contributed to
I've written a lot of free software. For example, try Jumpman Zero for the Palm (link in sig). If you looked at the code for that, you'd see that I allocate all the space for level data when the program opens and leave it allocated the whole way through. It's a few K spent, but the program never has memory crashes (assuming the Palm has enough memory to start the thing).
I'd suggest this approach only in small footprint sort of apps, or apps where performance footprint or convention means that not much else will likely be running (eg. fullscreen games). In many apps, the memory requirements are consistent enough that total demand is going to be the same either way.
Mostly I just hate people to be doing lots of work in C to save 30k of system memory - and ending up with a buggy program full of memory leaks. Many apps have data sets this small (30k) and yet are spending lots of time/effort managing memory.
Many small programs are no longer memory, or even performance, constrained. As such, a reasonable strategy for a lot of desktop software is to allocate a huge buffer at startup, and do repetitive flushes and complete reloads of data (always using the same pre-allocated buffers).
This is simple to do, and avoids a lot of errors. It's also not much of a headline.
So if anything, Amazon proves click and mortar *doesn't* work.
Amazon lost 149 million last year. Conversely, ToysRUs made 229 million. ToysRUs.com may not have made money, but it is apparently part of a business strategy that is functional. Indigo books made money last year (well, a bit anyways). Barnes & Noble made money last year ($100 million - not too shabby).
Anyways, I just find it entertaining that Amazon is being trotted out as a success in comparison with companies that are actually making money.
..the mean old Dean still won't let the prof enter the class in "Battle of the Bands". It's against school policy - and, besides, what hope does a ragtag bunch of school kids have of winning?
"Free speech" is often trotted out to defend things that don't particularly matter - like depictions of Jesus done in elephant poo. These cases are rightly decided largely as prophylactic; we need to defend the elephant poo speech or soon we'll lose our rights to important speech.
And what is that important speech, what is the speech that is in most need of defense from government? How about speech about that government.
If this site was a parody of a commercial or personal site, I'd say take it down. You need to follow the rules if you're going to say just some everyday thing. But since it's a parody of the site promoting the party in power, I think the call has to be to treat it extremely delicately.
The right to criticize the government needs the widest possible of berths. It's the treasure that "free speech" in general provides the wall around. It is the voice that needs the most protection.
HTA's are about being able to use HTML to create a desktop application, and treating the result as a desktop application (ie, different security arrangement and display). I can imagine how this would be useful for simple apps, especially to programmers accustomed to HTML/Script. HTA's are treated as executable code, and are not (barring an exploit) able to be popped up via web page. They are not connected to the web particularly other than that they share an underlying language, HTML. Regardless of browser, I don't imagine anyone sees them very often.
I think the whole idea went out of favor at MS a long time ago - I haven't seen an HTA article there for a while. They apparently weren't too memorable, the comments I've read thusfar betray no understanding of what they are/were.
The game should have had a commanding tutorial. Perhaps the full version does. Barring that, the best practice is to play as a non-commander (to see what weapons/towers/strategies do, and to hopefully see what a good commander does) and as commander on an empty server.
..is indeed a problem in Savage. It's a problem that's solveable (there's a vote to impeach command), but players often won't bother to vote - kind of like reality.
A smaller threshold (30% or so, with longer vote time) to impeach would largely solve the problem on public servers. Alternatively, command could be restricted to those who scored well in the previous round (or they could be allowed to choose commander). These players will not be horrible commanders, or will at least be able to choose a non-newbie to command. The game also needs to handle "command vacuum" better. For example, requests for funds should be auto-approved.
I guess what I'm saying is that the problems are not so much in concept but in implementation. Like other game features in the past, "team with commander" is something that needs to be ironed out. On the whole, though, it makes for a rewarding gameplay experience - it's a good way for allowing individuals to play together cohesively without being game experts.