yeah, well, the ESRB initially gave Unreal Tournament 3 a low rating, then drastically revised it later. It seems they toss a dart at a board to determine rating and then revise it later if they're off - definitely not a competent group, IMO.
and FYI, yes, that is a serious screenshot from Vista, not hacked in any way - I avoided upgrading to the 1.1 patch for weeks just so I could show all my friends.
he's a medical doctor and performed at least one partial birth abortion before it was banned, as I recall, so has a strong opinion against it. Abortion is a touchy issue in the states - when is it life, and thereby, when is it murder?
Well, I don't embrace 100% of ANY platform I've ever seen.
I'm pro-immigration, as opposed to Ron Paul. Yes, they may have gotten here and be working here illegally, but as it is, they aren't paying taxes, either, so your choice is spend billions on building a wall and kicking them out just to have them dig a new tunnel to get back in, or making them taxpayers and accept them in.
I'm also pro-abortion to an extent - the day after pill=good, partial term abortion=bad, so at odds with Ron, the Republicans, the Catholics, and everyone else that believe life begins at conception - to be quite honest, life begins before conception and most of those cells die anyway, so IMO, they should draw a realistic line (heartbeat, brain development, etc). If your or your doctor's religious beliefs oppose them, great - either don't get one or find a different doctor.
I'm pro-net neutrality to an extent (unlike Ron) - I'm not opposed to businesses like Comcast building a separate piggy-back network and running their own content as long as they don't selectively throttle internet traffic in so doing - call it net neutrality with a pro business flair.
I'm small government and anti-universal health care, so at odds with the Democrats.
I'm pro-drug/individual rights, so on the same side as the Greens and Libertarians, but at odds with everyone else. I think prescription healthcare needs a serious reform and the war against drugs needs to end. I'm at odds with Dems, Republicans, and drug dealers (for devaluing their trade), so fat chance of that ever happening.
I don't give a rat's ass about gun policy - legalize them all or make em all illegal - I don't care.
The war in Iraq? Justified during the Reagan administration (for known attacks on Halabja and other villages using VX, Tabun, and Sarin nerve agents in 1988 and soviet-style purges), and should have had an international task force from the United Nations, not a US invasion. Oh, but they were ALLIES with us back then, fighting the EVIL Iran, so we couldn't let that happen. The US had no business invading without UN backing and Iraq had been complying with US orders, but I think Bush got a bit pissy because Iraq kept shooting at our jets that were patrolling their skies.
The gold standard was, in fact, being used well into the 20th century - until 1976 in some countries. At some point, however, the resource begins to limit the economy and you need to keep large stockpiles of that commodity just in case someone wants to cash in (e.g. Fort Knox). The reason the gold standard ended in 1971 was because some countries like France wanted us to repay our debts in gold and Nixon didn't want to give it to them. Put in perspective today, a country (the US) that is massively in debt wants to tie its currency to gold, meaning they have to stockpile gold to pay their debt - how the HELL are they going to get it? fiat currency? The other option is mine it, but that means buying mines and hiring workers and basically creating another big-government bureaucratic monopoly (hi postal system - we still love you!). Gold standard years also typically had a higher level of unemployment than afterward, meaning there'd be less fiat currency (in taxes) to buy gold, as well.
you've got some of that right Ray Tracing does in fact scale better than rasterization. I don't remember the Big-Oh numbers exactly, but I believe you're correct in that it is O(logn) with no reflections - more on that later...
With the exception of LOD, none of those are really hacks - they just remove parts of the scene that aren't needed, something rasterizers excel at, but you can't really do with ray tracers because ray tracers need to be aware of the entire scene (since object know about each other). Ray tracers usually throw away objects rather than reduce their detail if the scene is too large, and I'm not sure that's better.
what isn't quite right: Most modern renderers can do single pass shadows with some limitations (e.g. casters can't get too close to the light source).
both raytracing and rastering draw only what's on the screen but need to be aware of objects offscreen for purposes of shadowing and lighting. Rasterization does this by clipping polygons against the view volume, raytracing by casting rays for each pixel shown and ignoring anything outside the display area. Clipping is a hardware operation and doesn't affect runtime performance these days.
HARD shadows in ray tracing come naturally. Ray tracers actually do diffuse lighting poorly, so soft shadows require costly operations such as photon mapping or radiosity.
I'm not so sure about graphics cards ever supporting raytracing - what is probably more likely is a on-dye convergence with the CPUs, and from what I've read, companies like Intel are already working on this. As I said, raytracing needs to be aware of the scene because objects see each other to properly draw reflections and shadows. That means it is best if the entire (viewable) scene is in memory (geometry and textures). GPUs currently work on limited datasets - limited parts of the scene and textures are in memory and they fake shadows and don't do true reflections because objects are not aware of other objects. I believe ray tracing could massively benefit from Unified Shaders if implemented on-dye with a CPU, because they probably can be used like a bunch of high speed parallel processors. Parallel memory access to data may be the biggest time hamper.
other stuff: Ray tracing and pseudo ray tracing are already used in games, just not at the scene level. Look at Steep Parallax Mapping or Relief Mapping - both those techniques use pseudo ray tracers on what is essentially a 3D texture (it's a mapped surface - and some versions I've seen even use true ray tracing for modern hardware).
When you add in a technique that gives good diffuse to a ray traced scene, its big-Oh value is as bad or worse than rasterization (photon mapping is O(nlogn), I think Radiosity is O(n^2)).
In summary, Ray Tracing excels at: Specular Lighting, reflections, and hard shadowing rendering complex scenes rendering any mathematically defined object (spheres, tori, etc), not just polygons visual accuracy (spheres are round, for instance). is easily parallel-izable (each ray can be calculated independently).
Ray Tracing stinks at: Diffuse lighting and related effects such as soft shadows. Scene reduction (not suitable for GPU) speed on smaller scenes Spacial indexing (realtime tree recalculation is much worse than for rasterization, as I recall, so it works best on static scenes)
If you haven't noticed, I don't favor one or the other, I think they both have merits and drawbacks. In college I estimated by Moore's law that by 2011 individual CPUs would be faster than rasterizers for most scenes provided Z-buffering replaced the painter's algorithm (which I expected because I used hardware accelerated IRIX workstations), but I totally missed the scaling and power of GPU hardware and multi-core processors, so it's amusing that that date actually may still be pretty close to correct.
There may be an interesting battle coming up. I think the next gen shader will be a tessellation shader, so you fee
I'm nitpicking a bit since GameFAQs was purchased by the CNET conglomerate which also owns GameSpot, but it technically is not GameSpot's cheat site - GameSpot has their own Cheats and FAQs.
sudo asks for a password, not a Yes click, and you only do it once when installing, not all the fricking time.
what poster was saying is MS should host a repository for drivers so the latest can always be retrieved. I'd say that is asking a bit much, but some sort of package management or driver management would be nice (like through automatically followed official links to signed drivers).
Vista is quite slow and has a MUCH larger memory footprint than XP. You can tack on a bunch of third party extensions and still use less memory and run faster than Vista.
My biggest beefs with Windows Vista: - when you install a driver, it means you want it turned on, as well, not requiring manually going to another dialog in a separate part of the OS and enabling it (e.g. File and Print Sharing). - Too many popups. - DirectX 10 exclusive platform in an attempt to force migration, but has little benefit. Cg and OpenGL can be used to work around this on XP. - Too large of memory and drive space footprint. - It still pisses me off that you can't switch languages except on the expensive professional and Ultimate. I know plenty of Spanish speaking US citizens that can't find a laptop that defaults to Spanish (and nobody sells cheap laptops with Pro or Ultimate). Not that they can't speak English, but I'm sure they feel most at home in their native tongue, as I certainly do. I know localization isn't cheap, but North American laptops should at least allow switching between French, Spanish and English, you greedy little pricks at M$. - BIOS needs to die, and since 32 bit processors are basically dead, 32 bit Vista should die, too (64 bit Vista supports EFI, which can be run in a BIOS compatibility mode). - still has stuff run in user mode that should be in kernel mode or other protected space. - laptop refers me to patches that are supposed to fix certain crashes, but when I attempt to install them, the patch says it doesn't apply to my OS. - Windows blocks a driver from starting up even though I've told it multiple times to allow that driver. - occasional services crashes on startup (possibly laptop related, but I haven't seen updated drivers from ASUS, either).
Actually, it is - acronyms typically are 1 letter per word but not always and can skip words (see NORAD below). The only requirement is that it is an abbreviation the is pronounceable as a word. Something without a vowel like STL (Standard Template Library) or the alphabet agencies (AAA, FCC, FSA FDIC, SEC and many more) are called Initialisms and usually are one letter per word, with the exception of anything that wouldn't be capitalized in a title (prepositions, articles, etc).
For instance NORAD - NORth american Aerospace Defense command RADAR - RAdio Detection And Ranging
or the recursive VISA - Visa International Service Association
After learning computer basics on an Apple ][ in elementary school, I remember moving to my uncle's PET and spending hours trying to make it work (with no instruction). Part of that was learning how to backspace. I did finally manage to figure out a few things, but it was very limited in what it could do (I think it had 8k of RAM).
I had completely forgotten that old Mattel machine/keyboard. I remember playing with it in a store back when I was a Mattel junkie (Intellivision). I managed to miss about half of those, including the PC Jr.
Still, I wish Solaris's keyboard had made it - the control and caps lock key are swapped, which differs from the typewriter standard location for caps lock, and it has buttons nobody uses (OK, I have, but I had to program them).
Computers of that era could move memory in larger blocks than the processor would allow, as well - in fact, I (and others - it was documented in The Black Art of 3D Game Programming, or some other book in that era) exploited that fact to copy blocks of aligned 'short' data using cast to double to quadruple the memory move speed for software blitters. Even more fun was using asynchronous floating point units in parallel with integer units on some processors (macs and many UNIX machines first, later Intel, though Intel had its serial MMX unit by that point which gave equivalent or better speed) to process.
Incidentally, they really didn't teach recursion in my Scheme classes, those focused almost entirely on list processing, especially linked lists. Recursion was, in fact, taught in my Software Engineering I class in C++. I don't think the concept is easier or harder in any particular language, but Scheme and LISP have a bit nicer syntax for recursive functions (mainly, they are inline in a parenthesis chain and not spread into a function and main section). I did a lot of recursion in LISP for my AI classes.
With the same respect, I've never had a C or C++ course that covered function pointers or threads, didn't have networking or databases in my specialty of usability and computer graphics, had minimal instruction on user and file/data manipulation (e.g. regular expressions like awk and a number of features from the standard template library), etc, so often there are lots of things you can learn on your own. Most computer languages are fairly easy to learn but hard to master.
no, the submitter should not have capitalized the 'and' and (optionally) capitalize the A in fast to emphasize the acronym - it should be FAst Search and Transfer
So what you're saying is the government pays for roads to be built so they can get paid in taxes. Great - I liked the other guy's way of thinking better, as a free ride on a space shuttle would be pretty cool. Realistically, I probably only paid for a couple kilos of rocket fuel, but as long as I don't get bumped a half centimeter off the ground, it seems like a good deal to me (yes, I used metric - NASA uses metric).
Forth and Postscript are useful for what, learning stack based languages? I learned Forth for one reason - to hack the PPC mac OpenBoot (open firmware), also used on some Solaris and IBM systems. Postscript I mucked with a bit also for hacking (I enjoyed trying to get root in college when I was a bored lab TA). I was much more successful with OpenBoot, as postscript is more geared to page layout and it's harder to do programming type operations.
Lisp and Scheme I'll lump in "parenthesis hell." I've never seen the allure of list processing languages - they drive me nuts. In the real world, you'll probably never see them even for what it's known best for, AI (python or lua are much more useful these days).
Haskell I've never used and seems to be stuck in the university wasteland. Ruby and Rails seem more practical, but no more than Python. To be honest, I dislike python because it uses indent significance, the one thing I despised most about Make. I've been meaning to look at Ruby, as well, because I like the Smalltalk object model, but I'm not sure how much I like linebreak significant line endings.
As abusive as it sounds, languages like COBOL (banking/finance) or Ada (government/military) or even FORTRAN (mech-E) will get you a job faster than Haskell. Or learn RPG and dedicate your life to IBM and Unisys mainframes (*shudder*).
weird - I could swear I read GPL for OGRE3D just recently, but it does appear LGPL - good catch. Digging around the license page, I see OGRE also has a commercial license available. I may have been thinking of NeoEngine, which has a similar dual license (it is GPL/commercial). NeoEngine (#1) was terribly broken, and even the main developer admitted it, but I know nothing of his new project NeoEngine2 so I'll reserve judgment.
CS does have a number of non-graphical features, but is not considered a complete game engine (missing such features as physics and vegetation rendering, though they are available as plug-ins). CS also is organized a bit like SDL, where you have a core library and a bunch of optional libraries (in SDL, stuff like SDL_net and SDL_image). OGRE 3d does not have these optional libs, only 3rd party plugins for external libs, thus the distinction you're making (but even wikipedia doesn't consider it a complete game engine).
When I evaluated both CS and OGRE (3, maybe 4 years ago) I was really disappointed with the ease of use for both compared to engines I've written myself, and I found plenty of places in the code I wrote that I felt should have been abstracted. OGRE definitely had the edge in framerates in my tests (it was at least 8-10FPS faster with 200000ish polygons), but I confess it was a series of models I created in high end CAD (Unigraphics), exported to STEP (or possibly IGES, my memory on that is a bit foggy), opened in Blender and saved to other formats, not a real game scene (basically, OGRE was faster at model rendering and true scene rendering wasn't tested). I should also say the CS people were rewriting a lot of their engine at that time, so it's entirely possible some of the complexity and framerate issues were resolved since then.
heh - probably the name it should have had - the Panther was never released and the Jaguar wasn't exactly a 64 bit system, it was a 32 bit CPU (the motorola 68000) and 32 bit coprocessor (essentially a GPU) with a 64 bit ALU (essentially an integer math coprocessor). Using the same argument, many of the 32 bit machines with 64 bit FPUs (like my mom's mac circa 1996) were also 64 bit machines.
you are correct - Nylon was invented in 1935, but was not really used in clothing until the silk shortage during World War II.
The 1930s prohibition era had a lot of anti-Mexican messages in the propaganda films, and I remember seeing one by a fairly famous director, but I've forgotten the name of it. The most well known is probably the 1936 "Tell Your Children" (Reefer Madness), which shows clean cut kids taking one puff and becoming sexually active and listening to that crazy jazz music.
not to mention there are LOTS of other companies that sell cripple-ware hardware. Verizon was notorious for it with their cell phones, though I believe they recently announced they will end this practice. Some of these phones had to be essentially cracked to unlock them (and yes, Apple does it, too).
actually, none of them are full game engines. Most interface with other freely available libraries such as ODE for physics. What you may be thinking of is the license - Ogre3D is GPL, so can't be used for commercial software, while Crystal Space, Irrlicht, and Lightfeather can.
Good points, and yeah "The Design of Everyday Things" has been a staple for UI and Mech-E ergonomics classes for many years and is quite a funny read.
A few things I remember from my UI classes
- Try to avoid non-movable, modal dialogs whenever possible - use them for emphasis on critical actions such as "Are you sure you want to Quit without saving?" - Group things that go together, together. I see this all the time by untrained dialog creators (as a matter of fact, I'm reading a spec with lots of these). Say you have a user entry box, a list box and 4 buttons Add User, Remove User, Cancel and Finish. The User Entry Box, list box, Add and Remove buttons should be grouped because they go together, and the finalizing actions Cancel and Finish should be separate. Generally, this is done with a line box around the grouped items and/or extra white space between the two groups. - Use page flow for whatever language you use. Usually this means flow your dialog from top left to bottom right. A language like Hebrew would go from top right to bottom left. There are some arguments on whether buttons such as OK and Cancel should be bottom left or bottom right, but that is nitpicking - as long as it flows from top to bottom, great. There are some exceptions to this for some cultures, so it's more a guideline than a rule. - Use appropriate controls for the dialog. For instance, use checkboxes for multiple choice and radio buttons for "exclusive or" choices. Make sure they're grouped appropriately, so if you have two separate sets of radio buttons, you can distinguish between the sets (I used to see this a lot, but not much anymore). - Avoid complex language and unusual phrasing whenever possible. A PRIME example of this is MS Word in the 1990s, where they had a dialog "Are you sure you want to Revert your document?" This option also happened to be close to Save and I know many people that lost documents because of it (I worked in a liberal arts computer lab). - Leave some space between buttons. I see this mostly with Java dialogs, where 8 square buttons are crammed together without any space between them and only an outline to distinguish where one ends and the next begins.
I agree HIG are GUIDELINES not rules. They help you design an experience in a consistent way, but they can't predict every use of every feature, so there are times you will need to stray. Apple says a menu bar should always be on top, but that might not work for a game or tray widget.
Irrlicht is a toy - you can dump a scene down easily, but if you want to do much more than that, the structuring is very restrictive and many things are not possible. The lightfeather project, which started as an Irrlicht fork and then rewrote most of the engine is much more flexible, but not as easy to use and not as feature-rich (stuff like DirectX support were dumped and it is OpenGL only, however, it also supports many modern GPU features that are not supported in Irrlicht).
If I had to pick one, though, I'd probably say Ogre3d is the best game oriented engine out there at the moment. CS has been doing a lot of work to modernize, but the performance still has issues (which they noted) and when I worked with it before I found it one of the most cumbersome engines to work with (out of the 5 or so I tested, 3 of which are dead now).
NWN2 has nicer graphics, yes, but performance is terrible for me on both of my machines and I believe it is CPU bound and probably single threaded (meaning dual core doesn't help). Neither of my machines technically meet the required CPU GHz speed, but in reality both are considered faster than a Pentium 4 those settings are based on (a 2.17GHz Athlon and a Core 2 Duo 2GHz). All other system settings exceed the recommended rating. On my fastest machine (the Core 2 Duo with 2GB RAM and an 8600M GPU - a fairly high end laptop) I need to have everything crammed down to low to even play the game and not fall asleep as it lags. This same machine playing UT3, Crysis, Bioshock or the Witcher (using an updated NWN1 Aurora engine) runs everything at medium to high settings. If I run Vista on it (It triple boots to XP and Ubuntu now since it shipped with Vista and I had an extra copy of XP lying around), the baseline rating is 4.9 due to CPU, with disk next at 5.3 and graphics I believe was 5.9, which should be more than enough for NWN2.
I never bothered trying the multiplayer game - single player is unbearable unless I cram everything down to lowest, and then it looks crappy but is at least playable. Unfortunately, I wasn't too interested in the main plot story and I've always disliked all the forced alignment choices. My evil character may be saving the town, but he may have ulterior motivations the computer can't account for (I want to loot it for myself, then maybe burn it to the ground). A GM can account for those things, but a script can't handle it, so I'd rather just get the evil action every time or the good action on a good char.
kill someone in the defense of you family or nation is perhaps regrettable but not immoral
incidentally, that's pretty much exactly what jihad means - radicals twist the wording to mean an offensive is OK to accomplish this.
Part of the translation problem comes when a 1-to-1 word translation doesn't exist - for instance, in Aramaic (or maybe it was ancient Hebrew), hand and wrist are the same word, so the common biblical translation says Jesus was crucified through the hand, even though the bones in the hand cannot hold up the weight of the human body. Languages also change with age, so the word may not mean the same thing as it meant in the original writings. Thirdly, many of the oldest texts contain some damage, so a best guess approach was taken. The number of the beast may be 666 or a nearby number like 667.
Furthermore, no original Aramaic text of the bible exists. The only known copy burned in the library of Alexandria. Thus existing translations into English are at least third generation, if not fourth. Some, like the King James with the Apocrypha might be hybrids because the Textus Receptus is third generation Latin and the Apocrypha is third generation Greek (as I recall).
I read DADOES before I saw Blade Runner (wanted to go but mom made me see something like ET instead - I was a bit young for that one, though I did catch the director's cut in theaters many years later) and the movie is vastly different in setting than the book - a fallout devastated, half-populated San Fransisco vs a overpopulated mega-city. On the other hand, both are cities in decay, just a different sort. Dick's book is essentially a reference to the Holocaust and World War II and Blade Runner more a parable about technology running out of control. If I had made the same movie based on the same source material, it would look a lot more like Cyborg than Blade Runner, and I don't think Gibson could say his world's look-and-feel was derived from that book, but it definitely is from the movie.
The 'androids' really aren't what we think of as androids, which are anthropomorphic robots, either - they're more like human manufactured organisms (almost clones) with certain features programmed in (like cell lifespan), as are the replicated extinct animals.
Stuff like the Penfield mood organ I think are really a result of the 60s climate of change and drug culture, and probably have something to do with wanting to straddle the line between the conservative 50s and the wildly liberal 60s. Still, yeah, it didn't have a whole lot to do with driving the plot along, so it probably wasn't necessary. The Buster Friendly vs Mercer stuff (the religious thread) I didn't understand at the time - I'd probably have to re-read the book and I didn't like it that much to begin with. I agree with you with Dick having great ideas and poor execution.
Slashdot Mirror
yeah, well, the ESRB initially gave Unreal Tournament 3 a low rating, then drastically revised it later. It seems they toss a dart at a board to determine rating and then revise it later if they're off - definitely not a competent group, IMO.
and FYI, yes, that is a serious screenshot from Vista, not hacked in any way - I avoided upgrading to the 1.1 patch for weeks just so I could show all my friends.
he's a medical doctor and performed at least one partial birth abortion before it was banned, as I recall, so has a strong opinion against it. Abortion is a touchy issue in the states - when is it life, and thereby, when is it murder?
Well, I don't embrace 100% of ANY platform I've ever seen.
I'm pro-immigration, as opposed to Ron Paul. Yes, they may have gotten here and be working here illegally, but as it is, they aren't paying taxes, either, so your choice is spend billions on building a wall and kicking them out just to have them dig a new tunnel to get back in, or making them taxpayers and accept them in.
I'm also pro-abortion to an extent - the day after pill=good, partial term abortion=bad, so at odds with Ron, the Republicans, the Catholics, and everyone else that believe life begins at conception - to be quite honest, life begins before conception and most of those cells die anyway, so IMO, they should draw a realistic line (heartbeat, brain development, etc). If your or your doctor's religious beliefs oppose them, great - either don't get one or find a different doctor.
I'm pro-net neutrality to an extent (unlike Ron) - I'm not opposed to businesses like Comcast building a separate piggy-back network and running their own content as long as they don't selectively throttle internet traffic in so doing - call it net neutrality with a pro business flair.
I'm small government and anti-universal health care, so at odds with the Democrats.
I'm pro-drug/individual rights, so on the same side as the Greens and Libertarians, but at odds with everyone else. I think prescription healthcare needs a serious reform and the war against drugs needs to end. I'm at odds with Dems, Republicans, and drug dealers (for devaluing their trade), so fat chance of that ever happening.
I don't give a rat's ass about gun policy - legalize them all or make em all illegal - I don't care.
The war in Iraq? Justified during the Reagan administration (for known attacks on Halabja and other villages using VX, Tabun, and Sarin nerve agents in 1988 and soviet-style purges), and should have had an international task force from the United Nations, not a US invasion. Oh, but they were ALLIES with us back then, fighting the EVIL Iran, so we couldn't let that happen. The US had no business invading without UN backing and Iraq had been complying with US orders, but I think Bush got a bit pissy because Iraq kept shooting at our jets that were patrolling their skies.
The gold standard was, in fact, being used well into the 20th century - until 1976 in some countries. At some point, however, the resource begins to limit the economy and you need to keep large stockpiles of that commodity just in case someone wants to cash in (e.g. Fort Knox). The reason the gold standard ended in 1971 was because some countries like France wanted us to repay our debts in gold and Nixon didn't want to give it to them. Put in perspective today, a country (the US) that is massively in debt wants to tie its currency to gold, meaning they have to stockpile gold to pay their debt - how the HELL are they going to get it? fiat currency? The other option is mine it, but that means buying mines and hiring workers and basically creating another big-government bureaucratic monopoly (hi postal system - we still love you!). Gold standard years also typically had a higher level of unemployment than afterward, meaning there'd be less fiat currency (in taxes) to buy gold, as well.
you've got some of that right
Ray Tracing does in fact scale better than rasterization. I don't remember the Big-Oh numbers exactly, but I believe you're correct in that it is O(logn) with no reflections - more on that later...
With the exception of LOD, none of those are really hacks - they just remove parts of the scene that aren't needed, something rasterizers excel at, but you can't really do with ray tracers because ray tracers need to be aware of the entire scene (since object know about each other). Ray tracers usually throw away objects rather than reduce their detail if the scene is too large, and I'm not sure that's better.
what isn't quite right:
Most modern renderers can do single pass shadows with some limitations (e.g. casters can't get too close to the light source).
both raytracing and rastering draw only what's on the screen but need to be aware of objects offscreen for purposes of shadowing and lighting. Rasterization does this by clipping polygons against the view volume, raytracing by casting rays for each pixel shown and ignoring anything outside the display area. Clipping is a hardware operation and doesn't affect runtime performance these days.
HARD shadows in ray tracing come naturally. Ray tracers actually do diffuse lighting poorly, so soft shadows require costly operations such as photon mapping or radiosity.
I'm not so sure about graphics cards ever supporting raytracing - what is probably more likely is a on-dye convergence with the CPUs, and from what I've read, companies like Intel are already working on this. As I said, raytracing needs to be aware of the scene because objects see each other to properly draw reflections and shadows. That means it is best if the entire (viewable) scene is in memory (geometry and textures). GPUs currently work on limited datasets - limited parts of the scene and textures are in memory and they fake shadows and don't do true reflections because objects are not aware of other objects. I believe ray tracing could massively benefit from Unified Shaders if implemented on-dye with a CPU, because they probably can be used like a bunch of high speed parallel processors. Parallel memory access to data may be the biggest time hamper.
other stuff:
Ray tracing and pseudo ray tracing are already used in games, just not at the scene level. Look at Steep Parallax Mapping or Relief Mapping - both those techniques use pseudo ray tracers on what is essentially a 3D texture (it's a mapped surface - and some versions I've seen even use true ray tracing for modern hardware).
When you add in a technique that gives good diffuse to a ray traced scene, its big-Oh value is as bad or worse than rasterization (photon mapping is O(nlogn), I think Radiosity is O(n^2)).
In summary, Ray Tracing excels at:
Specular Lighting, reflections, and hard shadowing
rendering complex scenes
rendering any mathematically defined object (spheres, tori, etc), not just polygons
visual accuracy (spheres are round, for instance).
is easily parallel-izable (each ray can be calculated independently).
Ray Tracing stinks at:
Diffuse lighting and related effects such as soft shadows.
Scene reduction (not suitable for GPU)
speed on smaller scenes
Spacial indexing (realtime tree recalculation is much worse than for rasterization, as I recall, so it works best on static scenes)
If you haven't noticed, I don't favor one or the other, I think they both have merits and drawbacks. In college I estimated by Moore's law that by 2011 individual CPUs would be faster than rasterizers for most scenes provided Z-buffering replaced the painter's algorithm (which I expected because I used hardware accelerated IRIX workstations), but I totally missed the scaling and power of GPU hardware and multi-core processors, so it's amusing that that date actually may still be pretty close to correct.
There may be an interesting battle coming up. I think the next gen shader will be a tessellation shader, so you fee
I'm nitpicking a bit since GameFAQs was purchased by the CNET conglomerate which also owns GameSpot, but it technically is not GameSpot's cheat site - GameSpot has their own Cheats and FAQs.
sudo asks for a password, not a Yes click, and you only do it once when installing, not all the fricking time.
what poster was saying is MS should host a repository for drivers so the latest can always be retrieved. I'd say that is asking a bit much, but some sort of package management or driver management would be nice (like through automatically followed official links to signed drivers).
Vista is quite slow and has a MUCH larger memory footprint than XP. You can tack on a bunch of third party extensions and still use less memory and run faster than Vista.
My biggest beefs with Windows Vista:
- when you install a driver, it means you want it turned on, as well, not requiring manually going to another dialog in a separate part of the OS and enabling it (e.g. File and Print Sharing).
- Too many popups.
- DirectX 10 exclusive platform in an attempt to force migration, but has little benefit. Cg and OpenGL can be used to work around this on XP.
- Too large of memory and drive space footprint.
- It still pisses me off that you can't switch languages except on the expensive professional and Ultimate. I know plenty of Spanish speaking US citizens that can't find a laptop that defaults to Spanish (and nobody sells cheap laptops with Pro or Ultimate). Not that they can't speak English, but I'm sure they feel most at home in their native tongue, as I certainly do. I know localization isn't cheap, but North American laptops should at least allow switching between French, Spanish and English, you greedy little pricks at M$.
- BIOS needs to die, and since 32 bit processors are basically dead, 32 bit Vista should die, too (64 bit Vista supports EFI, which can be run in a BIOS compatibility mode).
- still has stuff run in user mode that should be in kernel mode or other protected space.
- laptop refers me to patches that are supposed to fix certain crashes, but when I attempt to install them, the patch says it doesn't apply to my OS.
- Windows blocks a driver from starting up even though I've told it multiple times to allow that driver.
- occasional services crashes on startup (possibly laptop related, but I haven't seen updated drivers from ASUS, either).
Actually, it is - acronyms typically are 1 letter per word but not always and can skip words (see NORAD below). The only requirement is that it is an abbreviation the is pronounceable as a word. Something without a vowel like STL (Standard Template Library) or the alphabet agencies (AAA, FCC, FSA FDIC, SEC and many more) are called Initialisms and usually are one letter per word, with the exception of anything that wouldn't be capitalized in a title (prepositions, articles, etc).
For instance
NORAD - NORth american Aerospace Defense command
RADAR - RAdio Detection And Ranging
or the recursive
VISA - Visa International Service Association
After learning computer basics on an Apple ][ in elementary school, I remember moving to my uncle's PET and spending hours trying to make it work (with no instruction). Part of that was learning how to backspace. I did finally manage to figure out a few things, but it was very limited in what it could do (I think it had 8k of RAM).
I had completely forgotten that old Mattel machine/keyboard. I remember playing with it in a store back when I was a Mattel junkie (Intellivision). I managed to miss about half of those, including the PC Jr.
Still, I wish Solaris's keyboard had made it - the control and caps lock key are swapped, which differs from the typewriter standard location for caps lock, and it has buttons nobody uses (OK, I have, but I had to program them).
Computers of that era could move memory in larger blocks than the processor would allow, as well - in fact, I (and others - it was documented in The Black Art of 3D Game Programming, or some other book in that era) exploited that fact to copy blocks of aligned 'short' data using cast to double to quadruple the memory move speed for software blitters. Even more fun was using asynchronous floating point units in parallel with integer units on some processors (macs and many UNIX machines first, later Intel, though Intel had its serial MMX unit by that point which gave equivalent or better speed) to process.
Incidentally, they really didn't teach recursion in my Scheme classes, those focused almost entirely on list processing, especially linked lists. Recursion was, in fact, taught in my Software Engineering I class in C++. I don't think the concept is easier or harder in any particular language, but Scheme and LISP have a bit nicer syntax for recursive functions (mainly, they are inline in a parenthesis chain and not spread into a function and main section). I did a lot of recursion in LISP for my AI classes.
With the same respect, I've never had a C or C++ course that covered function pointers or threads, didn't have networking or databases in my specialty of usability and computer graphics, had minimal instruction on user and file/data manipulation (e.g. regular expressions like awk and a number of features from the standard template library), etc, so often there are lots of things you can learn on your own. Most computer languages are fairly easy to learn but hard to master.
no, the submitter should not have capitalized the 'and' and (optionally) capitalize the A in fast to emphasize the acronym - it should be
FAst Search and Transfer
So what you're saying is the government pays for roads to be built so they can get paid in taxes. Great - I liked the other guy's way of thinking better, as a free ride on a space shuttle would be pretty cool. Realistically, I probably only paid for a couple kilos of rocket fuel, but as long as I don't get bumped a half centimeter off the ground, it seems like a good deal to me (yes, I used metric - NASA uses metric).
Forth and Postscript are useful for what, learning stack based languages? I learned Forth for one reason - to hack the PPC mac OpenBoot (open firmware), also used on some Solaris and IBM systems. Postscript I mucked with a bit also for hacking (I enjoyed trying to get root in college when I was a bored lab TA). I was much more successful with OpenBoot, as postscript is more geared to page layout and it's harder to do programming type operations.
Lisp and Scheme I'll lump in "parenthesis hell." I've never seen the allure of list processing languages - they drive me nuts. In the real world, you'll probably never see them even for what it's known best for, AI (python or lua are much more useful these days).
Haskell I've never used and seems to be stuck in the university wasteland. Ruby and Rails seem more practical, but no more than Python. To be honest, I dislike python because it uses indent significance, the one thing I despised most about Make. I've been meaning to look at Ruby, as well, because I like the Smalltalk object model, but I'm not sure how much I like linebreak significant line endings.
As abusive as it sounds, languages like COBOL (banking/finance) or Ada (government/military) or even FORTRAN (mech-E) will get you a job faster than Haskell. Or learn RPG and dedicate your life to IBM and Unisys mainframes (*shudder*).
weird - I could swear I read GPL for OGRE3D just recently, but it does appear LGPL - good catch. Digging around the license page, I see OGRE also has a commercial license available. I may have been thinking of NeoEngine, which has a similar dual license (it is GPL/commercial). NeoEngine (#1) was terribly broken, and even the main developer admitted it, but I know nothing of his new project NeoEngine2 so I'll reserve judgment.
CS does have a number of non-graphical features, but is not considered a complete game engine (missing such features as physics and vegetation rendering, though they are available as plug-ins). CS also is organized a bit like SDL, where you have a core library and a bunch of optional libraries (in SDL, stuff like SDL_net and SDL_image). OGRE 3d does not have these optional libs, only 3rd party plugins for external libs, thus the distinction you're making (but even wikipedia doesn't consider it a complete game engine).
When I evaluated both CS and OGRE (3, maybe 4 years ago) I was really disappointed with the ease of use for both compared to engines I've written myself, and I found plenty of places in the code I wrote that I felt should have been abstracted. OGRE definitely had the edge in framerates in my tests (it was at least 8-10FPS faster with 200000ish polygons), but I confess it was a series of models I created in high end CAD (Unigraphics), exported to STEP (or possibly IGES, my memory on that is a bit foggy), opened in Blender and saved to other formats, not a real game scene (basically, OGRE was faster at model rendering and true scene rendering wasn't tested). I should also say the CS people were rewriting a lot of their engine at that time, so it's entirely possible some of the complexity and framerate issues were resolved since then.
heh - probably the name it should have had - the Panther was never released and the Jaguar wasn't exactly a 64 bit system, it was a 32 bit CPU (the motorola 68000) and 32 bit coprocessor (essentially a GPU) with a 64 bit ALU (essentially an integer math coprocessor). Using the same argument, many of the 32 bit machines with 64 bit FPUs (like my mom's mac circa 1996) were also 64 bit machines.
you are correct - Nylon was invented in 1935, but was not really used in clothing until the silk shortage during World War II.
The 1930s prohibition era had a lot of anti-Mexican messages in the propaganda films, and I remember seeing one by a fairly famous director, but I've forgotten the name of it. The most well known is probably the 1936 "Tell Your Children" (Reefer Madness), which shows clean cut kids taking one puff and becoming sexually active and listening to that crazy jazz music.
so the testbed for this tech was what, Vista users?
not to mention there are LOTS of other companies that sell cripple-ware hardware. Verizon was notorious for it with their cell phones, though I believe they recently announced they will end this practice. Some of these phones had to be essentially cracked to unlock them (and yes, Apple does it, too).
actually, none of them are full game engines. Most interface with other freely available libraries such as ODE for physics. What you may be thinking of is the license - Ogre3D is GPL, so can't be used for commercial software, while Crystal Space, Irrlicht, and Lightfeather can.
Good points, and yeah "The Design of Everyday Things" has been a staple for UI and Mech-E ergonomics classes for many years and is quite a funny read.
A few things I remember from my UI classes
- Try to avoid non-movable, modal dialogs whenever possible - use them for emphasis on critical actions such as "Are you sure you want to Quit without saving?"
- Group things that go together, together. I see this all the time by untrained dialog creators (as a matter of fact, I'm reading a spec with lots of these). Say you have a user entry box, a list box and 4 buttons Add User, Remove User, Cancel and Finish. The User Entry Box, list box, Add and Remove buttons should be grouped because they go together, and the finalizing actions Cancel and Finish should be separate. Generally, this is done with a line box around the grouped items and/or extra white space between the two groups.
- Use page flow for whatever language you use. Usually this means flow your dialog from top left to bottom right. A language like Hebrew would go from top right to bottom left. There are some arguments on whether buttons such as OK and Cancel should be bottom left or bottom right, but that is nitpicking - as long as it flows from top to bottom, great. There are some exceptions to this for some cultures, so it's more a guideline than a rule.
- Use appropriate controls for the dialog. For instance, use checkboxes for multiple choice and radio buttons for "exclusive or" choices. Make sure they're grouped appropriately, so if you have two separate sets of radio buttons, you can distinguish between the sets (I used to see this a lot, but not much anymore).
- Avoid complex language and unusual phrasing whenever possible. A PRIME example of this is MS Word in the 1990s, where they had a dialog "Are you sure you want to Revert your document?" This option also happened to be close to Save and I know many people that lost documents because of it (I worked in a liberal arts computer lab).
- Leave some space between buttons. I see this mostly with Java dialogs, where 8 square buttons are crammed together without any space between them and only an outline to distinguish where one ends and the next begins.
I agree HIG are GUIDELINES not rules. They help you design an experience in a consistent way, but they can't predict every use of every feature, so there are times you will need to stray. Apple says a menu bar should always be on top, but that might not work for a game or tray widget.
Irrlicht is a toy - you can dump a scene down easily, but if you want to do much more than that, the structuring is very restrictive and many things are not possible. The lightfeather project, which started as an Irrlicht fork and then rewrote most of the engine is much more flexible, but not as easy to use and not as feature-rich (stuff like DirectX support were dumped and it is OpenGL only, however, it also supports many modern GPU features that are not supported in Irrlicht).
If I had to pick one, though, I'd probably say Ogre3d is the best game oriented engine out there at the moment. CS has been doing a lot of work to modernize, but the performance still has issues (which they noted) and when I worked with it before I found it one of the most cumbersome engines to work with (out of the 5 or so I tested, 3 of which are dead now).
NWN2 has nicer graphics, yes, but performance is terrible for me on both of my machines and I believe it is CPU bound and probably single threaded (meaning dual core doesn't help). Neither of my machines technically meet the required CPU GHz speed, but in reality both are considered faster than a Pentium 4 those settings are based on (a 2.17GHz Athlon and a Core 2 Duo 2GHz). All other system settings exceed the recommended rating. On my fastest machine (the Core 2 Duo with 2GB RAM and an 8600M GPU - a fairly high end laptop) I need to have everything crammed down to low to even play the game and not fall asleep as it lags. This same machine playing UT3, Crysis, Bioshock or the Witcher (using an updated NWN1 Aurora engine) runs everything at medium to high settings. If I run Vista on it (It triple boots to XP and Ubuntu now since it shipped with Vista and I had an extra copy of XP lying around), the baseline rating is 4.9 due to CPU, with disk next at 5.3 and graphics I believe was 5.9, which should be more than enough for NWN2.
I never bothered trying the multiplayer game - single player is unbearable unless I cram everything down to lowest, and then it looks crappy but is at least playable. Unfortunately, I wasn't too interested in the main plot story and I've always disliked all the forced alignment choices. My evil character may be saving the town, but he may have ulterior motivations the computer can't account for (I want to loot it for myself, then maybe burn it to the ground). A GM can account for those things, but a script can't handle it, so I'd rather just get the evil action every time or the good action on a good char.
incidentally, that's pretty much exactly what jihad means - radicals twist the wording to mean an offensive is OK to accomplish this.
Part of the translation problem comes when a 1-to-1 word translation doesn't exist - for instance, in Aramaic (or maybe it was ancient Hebrew), hand and wrist are the same word, so the common biblical translation says Jesus was crucified through the hand, even though the bones in the hand cannot hold up the weight of the human body. Languages also change with age, so the word may not mean the same thing as it meant in the original writings. Thirdly, many of the oldest texts contain some damage, so a best guess approach was taken. The number of the beast may be 666 or a nearby number like 667.
Furthermore, no original Aramaic text of the bible exists. The only known copy burned in the library of Alexandria. Thus existing translations into English are at least third generation, if not fourth. Some, like the King James with the Apocrypha might be hybrids because the Textus Receptus is third generation Latin and the Apocrypha is third generation Greek (as I recall).
I read DADOES before I saw Blade Runner (wanted to go but mom made me see something like ET instead - I was a bit young for that one, though I did catch the director's cut in theaters many years later) and the movie is vastly different in setting than the book - a fallout devastated, half-populated San Fransisco vs a overpopulated mega-city. On the other hand, both are cities in decay, just a different sort. Dick's book is essentially a reference to the Holocaust and World War II and Blade Runner more a parable about technology running out of control. If I had made the same movie based on the same source material, it would look a lot more like Cyborg than Blade Runner, and I don't think Gibson could say his world's look-and-feel was derived from that book, but it definitely is from the movie.
The 'androids' really aren't what we think of as androids, which are anthropomorphic robots, either - they're more like human manufactured organisms (almost clones) with certain features programmed in (like cell lifespan), as are the replicated extinct animals.
Stuff like the Penfield mood organ I think are really a result of the 60s climate of change and drug culture, and probably have something to do with wanting to straddle the line between the conservative 50s and the wildly liberal 60s. Still, yeah, it didn't have a whole lot to do with driving the plot along, so it probably wasn't necessary. The Buster Friendly vs Mercer stuff (the religious thread) I didn't understand at the time - I'd probably have to re-read the book and I didn't like it that much to begin with. I agree with you with Dick having great ideas and poor execution.
If you know how, let me know - the desktop machine Remote Desktop always disconnects when I connect the laptop to it.
I may also try Sunbelt, as well - I believe we have a site + home corporate license (will need to check on that).