And lets keep in mind that ALL the games that he's worked on up to and including Quake 2 have been GPL'd by now.
Very good point. (Not technically correct, as neither the id branded Commander Keen games nor any of the non-id branded games he has worked on are GPL'd...but all the important ones, yes.)
I have heard that Carmack actually develops on a Linux box, but that's just speculation.
No, he uses NT. Perhaps 2k or XP these days, but you'll notice he says "NT is definitely going to be the primary development platform for our next project, but I will be evaluating alternatives for a possible transition after that." That "next project" is Doom 3. (The interview was from Oct '99, late in the Q3:A development process.)
If they decide to release a linux version of Doom3, and given Carmack's good attitude towards open source and OpenGL, I really really would love if they go and piss off Mr Gates by releasing the test for Linux first.
Could happen. For Q3, id released the first test for Mac only, to keep the beta testing group small enough for the first round. (Later tests were of course released for Windows, once the first round of bugs was ironed out.)
I bet that a zillion gamers would install Linux just to be able to test Doom3. They have been waiting for years!!
Wouldn't that be a hoot. Anyways, that just might be the reason for id *not* to release the D3 test on Linux first; it's a lot easier and cheaper for gamers to install Linux to run the D3 test than it is to install Mac OS X, so the test group might not stay small enough for their purposes.
OTOH, the main purpose of the Q3 test was to iron out netcode issues, stuff they couldn't easily test in the lab. As D3 is focused heavily on single player and probably will not contain much new in the way of netcode, they might even do without a test altogether...
Re:OpenGL 2.0 and OpenSource
on
Doom3 and OpenGL2.0
·
· Score: 5, Informative
Also, does anyone know if there will be a supported version of Doom3 for Linux, or will we be relying on ported versions?
id has released nearly simultaneous Linux binaries for all of their games (client and server) since Quake 1, and released Linux patches for Doom 1 and 2 as well. Loki was involved inasmuch as they published a retail box of Linux Q3; however, this was never really important because you could always get Linux Q3 by buying the Windows version and downloading a small patch from id. (Indeed, the retail Linux version sold poorly, probably because it was released a couple weeks after the Windows version and thus many people went the buy-Windows-and-download-patch route.)
I believe id has officially announced that Doom 3 will available for Linux (and Mac), but if not it's still a virtual certainty. id has always been a tremendous supporter of open standards; Carmack chose OpenGL over DirectX for Quake (and thereby single-handedly created the consumer OpenGL market), and in addition to working on Mac, Linux and Windows versions of all 3 Quake games simultaneously, released Doom ports for Next (id developed on Next workstations back then), Solaris, IRIX (I think, or maybe that was unofficial), and I believe even Linux on Alpha in addition to the already mentioned x86 Linux ports.
Again, id has always done the port themselves; most likely, you will have to buy the Windows version and download a patch which will almost certainly be available within days of release.
Their major concern seems to be frame rates in SS and Q3A, two games built on old technology.
Exactly the point: if Parhelia dips as low as 35 fps running SS @ 1024x768 with nothing turned on, then you know for a fact that it will be near-unplayable on newer more demanding games.
But, since the reviews available test a whole lot more than just SS and Q3 engine games, we don't need to rely on that obvious deduction. Parhelia also gets its clock cleaned on newer games like Commanche4, and UT 2003, as well as the mini-games in 3DMark, which are supposed to simulate future game engines. The only difference is, with these games you actually need a GF4 to get decent performance.
What I'm concerned about is high-resolution performance with AA enabled. I have no intention of ever again running a game below 1024x768 with AA enabled. Why would you, when the Parhelia can do it without breaking a sweat?
Because the only games that Parhelia can run at 1024x768 with 16x FAA "without breaking a sweat" are the ones "built on old technology" that you denigrated a sentence ago. Check out the scores for UT 2003 with AA and anisotropic filtering. Parehlia is most definitely breaking a sweat @ 1024x768. (So, of course, is the GF4.) This is for a game that is going to be out in a month; and not only that, but a deathmatch-oriented game which is most definitely tweaked for high framerates. If you never want to run a game below 1024 with AA, then you better either get used to the games that are already out today, or prepare to upgrade video cards very very often.
I'd like to see some benchmarks of the Parhelia running DooM3 at 1024x768 w/ 16xAA.
Too bad: using current drivers, it won't run it. That's because Parhelia's 16x AA is fragment anti-aliasing, which only AA's the edges of polygons, and thus can't deal with a stencil buffer which is used in Doom3 (and many other games) to render cast shadows. Even assuming a driver fix will allow Parhelia to at least run games with a stencil buffer in FAA mode, the shadows themselves will still have jaggies unless you use the much much slower FSAA mode. Given how large a part shadows play in the Doom3 experience, it's doubtful FAA will be better than no AA at all.
If your goal is to never run below 1024x768 with AA, your best bet is probably to buy a GF4 MX 440 today, ride that as long as it will last, and then upgrade in 12-15 months (before you need a DX8 compatible card, which GF4 MX is not) to whatever the best $150 card then is (i.e. equivalent of today's GF4 Ti4200 or Radeon 8500). You'll get your Doom3 at 1024 with AA, and save $175 over buying the Parhelia today to boot.
I wonder what The Carmack has to say about this card.
Me too; in particular, he (or someone) should be able to inform us where the expected throughput benefit for highly multi-textured games has disappeared to. Unfortunately, given the shrug gamers are going to give this card after this morning, he may not even bother.
Sorry to reply to my own post, but some additional news on the DOJ DRAM-industry investigation. First, since yesterday it's been announced that many other memory makers have recieved DOJ subpoenas, including Elpida, Toshiba, Nanya, Hynix, and even little Winbond. It's quite clear now that this has nothing to do with Micron in particular but rather the DRAM industry as a whole.
Second, it's been noted that, sad to say, the DOJ generally only undertakes antitrust investigations in response to a complaint by a US interest claiming to be hurt by the alleged anti-competitive behavior. Since the only major party (that I can think of) hurt by the DRAM price war was Hynix, a South Korean company, it's difficult to imagine on whose behalf the DOJ would be investigating.
Thus it has been suggested that instead the DOJ is investigating the jump in DRAM prices following the end of the price war (i.e. when the Hynix takeover was announced); Michael Dell in particular has publicly complained about high DRAM prices, so some have suggested Dell is behind the complaint.
IMO, if the investigation is focusing on the SDRAM spot prices which shot up after the price war ended, it's misguided. That jump in prices was, IMO, the natural consequence of both the end of an artificial price war and uncertainty on the demand side as Intel introduced its first DDR motherboards and the market had to be sure to have adequate inventory of both SDRAM and DDR to cover for the possibility of either a very successful DDR ramp or a slow one.
Now, if the complaint focuses on the rather outrageous gap between spot and contract prices which existed until it finally close in the last couple weeks, then it might get somewhere. On the one hand it's pretty clear that the pricing anomaly was that the spot price was artificially low, not that contract was artificially high. On the other, there was apparently good evidence that the major memory makers were indeed colluding to keep contract prices that high, so the DOJ might indeed find something to complain about...
The Micron story: First, the DOJ did not bring an action against Micron today. Rather, they sent a subpoena to Micron, seeking information for an investigation into possible anti-competitive behavior in the DRAM market.
This has nothing to do with Micron (with global DRAM marketshare of ~18%) having a "monopoly" in the DRAM market; indeed, in addition to Micron, Infineon and Samsung (and probably others) recieved subpoenas today as well. While the DOJ has not made any public statements about this (that I've seen), it is extraordinarily obvious to anyone who has followed the DRAM industry over the past year or so that they are investigating allegations of dumping on the DRAM spot market. In particular, the spot price for DRAM (128 Mbit SDRAM chips in particular) declined by something like 75% last fall, in the absence of major new supply or significant shortfalls in demand. The resulting price was well below even the marginal cost of producing these chips.
When the price suddenly shot back up a few months ago with the announcement that Micron had reached a tentative deal to buy the already-bankrupt Hynix, it became quite clear that the spot market price was being kept artificially low in order to try to force the ailing Hynix out of business. (The Micron-Hynix deal was later rejected by Hynix's board, causing--wouldn't you know it--another DRAM price war.) The whole thing was made even more fishy by the fact that DRAM contract prices (that is, the DRAM sold in monthly contracts to OEMs like Dell, HP, etc.) stayed way way higher than the spot prices; Hynix was mostly limited to the spot market in its attempts to desperately sell off excess inventory to try to raise cash to pay off their creditors (after all, who wants to sign a contract with a company on the verge of bankruptcy??).
The obvious conclusion is that the other memory makers (Samsung and Micron are #1 and 2 in marketshare) dumped excess inventory on the spot market or even purposely sold at worse prices than they could otherwise get, in an effort to drive Hynix out of business sooner and thus reduce supply and cause prices to be higher than before. By itself, this is probably not illegal, because none of them has a monopoly. What would be illegal is if any of them colluded to keep the spot price artificially low. If they did that, they would be guilty of forming an illegal trust (that's why it's called anti-trust law, not "anti-monopoly") and engaging in illegal dumping or even price-fixing. Personally I would be surprised if this did not occur, but we'll see if the DOJ can find enough concrete evidence to bring a claim.
[For those wondering how the DRAM market turned out--spot prices are way down again across the board, although 128Mbit SDRAM is not nearly as low as it got last fall. (256Mbit and DDR chips are close to record lows, though.) The difference now is that contract prices have finally come down to near-spot levels (in a normal market, contract prices are lower than spot levels), and, considering the recent warnings by Intel and AMD, it looks as if low demand may actually be to blame, not price-fixing.]
The Rambus story: The short version of events posted here is more or less right: Rambus was a member of JEDEC while the SDRAM standard was being worked on; Rambus did not disclose any patents they had which read on the SDRAM design being discussed, nor did they disclose any intent to seek royalties on the standard; later, after SDRAM was standardized and became the mainstream DRAM type, Rambus sought royalties from anyone producing SDRAM or an SDRAM memory controller, which is to say almost the entire worldwide electronics industry. (Also DDR, which according to Rambus not only infringes their SDRAM patents but also a couple others on double-clocked operation in a DRAM.)
Some memory producers (including Samsung and Elpida) agreed; others (specifically Micron, Infineon and Hynix) refused and were taken to court. (The threatened legal action regarding SDRAM controllers never really happened because Rambus was focusing on memory producers first.) In the first such case to come to trial, in which Rambus sued Infineon in Virginia, not only did the judge rule that Rambus' patents don't actually cover SDRAM/DDR, but the jury found Rambus guilty of fraud for sitting on JEDEC and failing to disclose their intent to seek royalties on SDRAM--which is essentially what the FTC claim charges. (Unlike the DOJ investigation I talked about above, the FTC is past the investigating stage and today actually filed a claim against Rambus.)
However, it's not quite as simple as other/. posts make it out to be. For one thing, Rambus claims that they didn't run afoul of the JEDEC policy requiring attendees to disclose relevant patents, because they did not yet have the patents which they later accused everyone of infringing. What makes the whole thing really complicated is that these "SDRAM patents" were not new patents either, but rather extensions of a previous Rambus patent on RDRAM, which Rambus did have at the time. Basically, patent law definitely allows you to file an extension of an old patent which makes clear that a new product infringes your original invention even if it may not infringe the original patent. How this intersects with the requirement to disclose patents to standards bodies is rather cloudy. As we know, Rambus got convicted of fraud for this--probably helped by documents at trial which pretty much proved Rambus had this plan in mind all along. However the issue is currently on appeal. IMO the Appeals Court's decision on this issue will determine what happens with this FTC claim filed today.
I am incredulous to anyone comparing copyright infringement on Napster to such an event. ... I merely questioned the sanity of someone that wanted to compare copyright infringement to events of that magnitude.
Most of the events listed are only large in magnitude in hindsight. At the time things like the Boston Tea Party (~50 radicals commit vandalism in the midst of a boycott involving a few thousand people) or Rosa Parks' refusal to give up her seat (1 person acting as a test case to spark a boycott involving, again, a few thousand people) were small in magnitude. Or (promised I wouldn't get into it but) Jesus walking around Judea and preaching a new radical brand of Judaism. Only as part of the larger historical movements they are associated with is it clear that the magnitude of their influence was enormous.
By any measure, the magnitude of Napster was much, much larger. 80 million users signed registered with Napster. Billions and billions of songs were traded. Unlike all the events listed, with Napster a substantial portion of those affected by the law actually joined in the "movement" against it.
Now, whether history will judge Napster to have a magnitude nearly as large as what we now credit to things like the Boston Tea Party; well, it's too early to say.
Oh, and fancy phrases will not candy-coat the ENORMOUS magnitude of difference there is in claiming to OWN a human, to feed or not, violate sexually or not, let live or not, to STEALING music. Music theft is what I would term petty. It is stupid.
The comparison is not between the ability to own a person and stealing music but between the ability to own a person and the ability to own an idea. I, like most people raised in the Western intellectual tradition in the past 50 years, of course find owning a person the much more abhorrent idea. But the simple fact is that that is not the response that most people in most civilizations have had. Nor is it clear that the ability to own an idea is not also horrible, albeit something we are nonetheless rather used to.
I also never claimed to AGREE with the copyright laws. All I claimed was that they ARE laws, and stated that the correct way around the problem was to get the law changed!!! Breaking the law is NOT going to get it done. Wake up.
Wake up yourself. Part of the point of the original post was that, in many instances in history, breaking the law was a vital step in the movement to get the law changed (or to have some other similarly far-reaching effect).
you are comparing music theft with Jesus' acts, the boston tea party, and the freedon of slaves?
Thanks for making my point. You really did just fall off the turnip truck.
Are you comparing the Boston Tea Party with Napster?? The Boston Tea Party was an act of outright vandalism in which anti-government reactionaries protesting a somewhat obscure point regarding differential tarriffs and government monopolies broke into ships which were not their property, and destroyed valuable physical property which not only did not belong to them but which other people wished to purchase and now could not. Indeed, the Boston Tea Party was the climactic event in an illegal embargo carried out against colonists as a whole by a small group of radicals who decided that the law regarding tea taxation was wrong and thus that all imports of British tea were to be blocked through threat, vandalism and violence.
Napster, on the other hand, did not involve stealing, vandalism, or destruction of physical property. It almost certainly contributed heftily to the bottom line of the record companies and musicians who brought complaints against it. Use of Napster to share (but not to download) copyrighted music was found to constitute illegal copyright infringement; however, this hinged on a technicality of the 1992 Audio Home Recording Act which did not include mp3s as being under the purview of the rule that noncommercial copying and distribution does not constitute infringement (before the AHRA, all noncommercial copying and distribution were considered fair use, so the AHRA was sharply broadening copyright law as it had been understood for centuries). Unlike the Boston Tea Party which was carried out by a small group of radicals, probably against the wishes of and certainly with economic harm to the large majority, Napster was used by tens of millions of Americans; perhaps an absolute majority of all Internet-connected Americans, and certainly an absolute majority of those with broadband.
The actions taken by the perpetrators of the Boston Tea Party were and are and always will be unambiguously illegal. No society can survive long on the principle that it is okay to break in and destroy others' commercially significant private property as a means of protesting government policy. Napster, on the other hand, was found illegal only after significant court action, and would not have been illegal under copyright law as it existed for centuries prior to 1992.
So why can we view the Boston Tea Party as an admirable event of history? Because it was carried out in the name of greater ideals, namely the proposition that all citizens deserve a voice in the laws that govern them. Note that this proposition is not obviously more correct or more noble than the proposition underlying Napster use, namely that people have the right to share artistic/intellectual content with each other provided they have purchased the means of reproduction and distribution; or, perhaps, that people have the right to use things they have purchased (i.e. CDs, computers and Internet access) as they see fit so long as their actions do not directly hurt anyone else. In any case, the point is that the Boston Tea Party proposition has since been judged by history to be correct (of course, history is written by the winners), whereas the Napster proposition(s) are perhaps almost as controversial today as the proposition that people have a right to self-government was 230 years ago.
As for slavery, we might note that it is not prima facie obvious that allowing people to be owned as private property is any more or less unusual or immoral or against the tenets of natural law than allowing ideas to be owned as private property. Indeed the vast majority human societies throughout history have considered people a valid form of property and ideas not.
As for Jesus...I'm not touching that with a ten foot pole.
But it arguably has the greatest overall loss to society, if you consider intellectual achievements and discoveries to be superior to physical and materialistic ones. Just as normal theft acts as a disincentive toward regular commerce, piracy works as a disincentive toward developing IP that requires a substantial amount of time/effort/money. I could see why potentially robbing the world of future discoveries might be considered the worst form of theft.
But the notion of intellectual "property" itself arguably has the greatest overall loss to society, if you consider intellectual achievements and discoveries to be superior to physical and materialistic ones. Just as normal theft acts as a disincentive toward regular commerce, the notion of intellectual "property" works as a disincentive toward developing new ideas that require the freedom to use and build upon old ideas as the scientific method and all serious forms of intellectual discourse have always done (that is to say, all ideas in a sufficiently advanced society). I could see why potentially robbing the world of future discoveries might be considered the worst form of theft.
Doom took a year? If I remember correctly, Doom was about a year late when it came out...am I wrong?
I believe so. The original design document (pdf) for Doom is dated 11/28/92. The release date was 12/10/93.
Obviously work was started before the design document was finished, but as you can tell by perusing it, the game was still being sketched out at the end of November, 1992. Several reports say that iD began Doom only after the Wolf 3D expansion Spear of Destiny was released, which was September, 1992. So, if we take that as the start date, it looks like about 15 months.
As for an iD game being a year late, I'm not sure, but you may be thinking of Quake 1?
there was a fundamental change somewhere in the last 15 years where things shifted from games using existing hardware fully to where games became the reason themselves to create new, faster hardware devices.
Not that this is bad, nit by any means, but it does give one interesting meat to consider; no one will argue that games are what's driving things like new video card technologies -- when did the chicken outdo the egg?
When games started taking a long time to make. Used to be even a revolutionary game could be made from start to finish in months. Wolfenstein 3-D took about 6 months to make. The original Doom took about a year. iD has been working on Doom3 for about two years and they've still probably got at least a year to go.
When Carmack decided the technical parameters of the Doom3 engine back in 2000, he would have been an idiot if he designed it to only take advantage of the features of existing hardware. Instead he designed it to use features and require performance which he knew would be entering the mainstream by the time the game was released.
Of course, Carmack is unique in that he can actually influence Nvidia, ATi, etc. a bit into supporting the features he wants. On the other hand, you have to realize that back in 2000 Carmack, Nvidia, ATi, etc. already had a very good idea what sort of features would be supported in the 3D cards of 2003. Of course there is some guessing and tweaking involved (which Carmack seems to be particularly good at), but a good 3D engine designer has to design for the hardware of the time when the game will be released, not the hardware of the time when he's designing.
As for why games take so much longer these days, that's another story but the basic point is that they are not only more complex technically but that there is ever increasing detail in the art, scripting, level design, etc. that it takes much larger teams with much better tools much longer to make a game than in the old days.
I wonder if any of the current laptops will be able to run Doom3... I'm considering buying a laptop with a GF4go as the Radeon7500 based ones seems to be slower... I wonder if its really worth it to go from 32 megs to 64 megs of ram?
Unfortunately, as both the GF4 Go and Mobile Radeon 7500 lack hardware pixel shaders, they will not be able to render Doom3 in its full glory. Of course they will be able to run it, but many of the graphical goodies will either be missing or will need to be (very slowly) computed on the CPU.
As for 32 vs. 64 MB, I'd go for the latter if you want to run Doom3. Surfaces in Doom3 can contain up to 5 texture maps, which means tons of RAM usage at anything but low texture detail. If you run out of room on the card, you need to store textures in main memory and access them over the AGP bus, which is too slow for that sort of thing. IIRC both the GF4 GO and Mobile Radeon 7500 are available with 64 MB, although I suppose one sometimes doesn't get the choice when buying a laptop.
Basically, the top-of-the-line 3D cards of today are going to be necessary to run Doom3 decently, so the top-of-the-line mobile 3D cards--which are about a generation behind the desktop--are going to be able to run it, but somewhat mediocrely. Of course, Doom3 probably won't be out for at least a year, maybe a year and a half. By that time you'll be able to buy a laptop which runs the game beautifully. If you have to buy a laptop now then it'll be a bit tougher. Kind of makes you wish laptop 3D cards were upgradable like desktop ones...
A license is a contract between two parties, in this case the creator and the user. Copyright is a doctrine of federal law, designed to promote the creation and dissemination of original content and (shocking but true) their eventual release into the public domain. If anything, copyright is an agreement between the creator and the government, not between the creator and the user.
Put another way: the fact that I can't go up to you and punch you in the face does not imply that we "have a license". (Except in the very esoteric context of social contract theory.)
In any case, copyright is neither here nor there when it comes to the BSA and software "piracy". (ARRRRRRRRR, matey!) Copyright law prohibits someone who is not the copyright owner from making copies and distributing them. It does not prohibit someone from buying or otherwise obtaining a distributed copy. When it comes to downloading software off the Internet, it's a little murky, because you are in some sense causing a copy to be made even though it is the other guy's machine, not yours, that is doing the copying. (He/his machine is definitely infringing.) But this is a load of BS; by the same token, the use of a browser cache is flagrant and unambiguous copyright infringement.
Point is, using software that someone else burns to a CD and gives to you is not copyright infringement. If it's commercial software then it's probably against the license, though...
Is it a whole new form factor so everyone had to redesign the motherboards and to force incompatability with older systems??
The answer is...yes and no. DDR-II does have a new form factor, with 232 pins, as compared to the 184 currently used in a DDR (I) DIMM. Similarly, a DDR (I) DIMM has a bit more pins than the 168 found in an SDRAM DIMM.
However, this isn't such a big deal for the mobo makers; it's just a matter of putting on a different slot and different traces on the board. One of the things that has made the transition from SDRAM to DDR more "evolutionary" than a transition to (say) RDRAM would have been, is that the same chipset can control both SDRAM and DDR, because they use the same (or at least backwards-compatible) commands. It's not often mentioned on the hardware-enthusiast sites which are only interested in benchmarking the fastest stuff around, but just about every DDR chipset is also available with SDRAM. You need a different motherboard--because the slots are incompatible--but the cost to the mobo makers for offering both versions is pretty small.
In a similar vein, the commands for DDR-II are a superset of the DDR-I command set, such that DDR-II chipsets should very easily be able to detect and use DDR-I as well, just like DDR-I chipsets currently use SDRAM as well. Furthermore there is talk (dunno if it will happen) of releasing DDR-I DIMMs in the 232-pin DDR-II form factor; that way, one could buy a motherboard and use either DDR-II or DDR-I in it, with no problems. Of course old sticks of DDR-I will not fit, and the new ones will not fit in current DDR motherboards.
So, while such a scheme doesn't get rid of all the headaches of an incompatible upgrade path, it does address some, albeit more on the mobo producer end of things than on the IT inventory end of things. It is indeed a pain that, while SDRAM (and the SDRAM form factor) enjoyed around 4 years as the mainstream memory type, DDR-I will only be on top for 2 before DDR-II takes over. And DDR-II will be lucky to have 3 years before something pin-incompatible comes along to replace it.
On the other hand, the SDRAM -> DDR transition probably would have happened a year earlier if Intel hadn't tried to transition to RDRAM instead. And, meanwhile, being stuck with SDRAM for 4 years meant DRAM bandwidth only doubled (and actual performance did less than that) over a period when CPU clock speeds increased by a factor of 5 or so. I think the DRAM industry wants to make speed increases more frequent than they were a few years ago, even if this means more inconvenience for corporate IT departments.
This is important because industrial and corperate-mission-critical is older equipment. and an upgrade path for ram is still important.
I'm not sure what you mean here. As far as the upgrade path for greater RAM capacity goes, standard SDRAM and DDR-I will still be made and sold for quite some time, even if they will eventually be more expensive than their newer and faster brethren. EDO RAM is still being made and sold today.
If you mean an upgrade path for higher-performance, of course you can't just buy faster DRAM and expect it to speed up (or even work in) a system that was built to use a slower type. The system clock sets the DRAM speed, and unless the system has been validated to run at the new DRAM speed, doing so amounts to overclocking. An performance upgrade path for RAM is always going to require purchasing at least a new CPU module and accompanying memory bus, if not a new machine.
Why is it obvious? No-where have iD stated they'll be 'fixing' it. Just because it's something we all want doesn't mean it will obviously be fixed...
1) It's by far the most noticeable graphical flaw in a game that is staking its reputation more than anything on having the best graphics ever seen.
2) It's extremely common for a game at this stage of development to have very few models already built in (after all, the modeling tools and to some degree the engine have to be finished before serious modeling work can begin). And extremely common to make a demo using the few that you already have.
3) It doesn't fit with the premise (i.e. different people are possessed by Hellish, uh, stuff and turned into zombies) for the zombies to look alike.
4) It's something we all want, and id has more than enough time to fix it.
Re:Was at E3 (i'm just an intern at EA)
on
E3 Doom III Preview
·
· Score: 5, Informative
There are about 4 different monsters that I saw, and all other monsters were clones of these monsters. I mean, the big fat belled one, it's very cool and all, but does ID need to have 100's of the same monster running around? What's the logic here?
Doom3 is nowhere near done; the only release date they're specifying at E3 is "2003", and most likely it won't be January either. In other words, they have around a year's work left on this game, maybe more. Presumably they will spend that time adding many many more character models, or even (hopefully) coming up with a way to generate different bump/displacement maps for the same character and thus avoid the "two of the exact same zombie" syndrome that affects the current screenshots.
In any case, it's pretty obvious the issue will be solved in the final game.
And you're forgetting that the radio stations are now required to pay back the RIAA and record labels for every song played, per listener . The streaming radio stations are required to pay double the rate. Trust me, the RIAA is well into the black on their outlay for payola to the radio stations.
Very interesting point, which I'd failed to consider. Actually it is perhaps for that very reason that the record labels don't actually pay the radio stations directly for song placement, but instead pay brokers millions and millions of dollars which they then use to pay the radio stations off.
The reason I know that is that it comes from a truly fascinating Wall Street Journal article (unfortunatly not available online to the general public; I'm gonna see if I can't scare up a copy through Lexis-Nexis) about the millions of dollars poured down the drain by one of the big 5 in an effort to make some chick by the name of Carly Hennessy the next Britney Spears. I think in the end they spent something like $4 or $5 million on her and ended up with roughly 1,000 CD sales and a Wall Street Journal article to show for it. (It may go without saying, but her music SUCKED and I mean was AWFUL; here's a link to her album on CDNow where you, like me, can get a taste of how terrible it is. And I actually LIKE Britney Spears, or at least can appreciate what's well crafted about her good songs.)
Ah; OK, so apparently L-N (or at least the academic version) doesn't get the full text of the Journal, but the piece was syndicated out to the Houston Chronicle in abbreviated form. Although...they don't offer free access to archives either, so I'll just exercise that fair use thing and give some excerpts.
First off, I was wrong; they only wasted $2 mil on her, but then again it was only 378 copies of Ultimate High that got sold. Here's the opening of the story (as excerpted by the Chronicle):
MARINA DEL REY, Calif. - Eighteen-year-old recording artist Carly Hennessy is packing up her small apartment. Her promotional posters will go into storage, and the beige rental couch will be returned. A weight-control message that the slender teen scrawled in marker on the refrigerator - "NO, U R FAT" - will be wiped clean.
For two years, Vivendi Universal SA's MCA Records paid the rent while Hennessy prepared for pop stardom. And that's not all: The label so far has spent about $ 2.2 million to make and market her new album, an upbeat pop recording called Ultimate High. "Some people just struggle," she says. "I was very, very lucky."
Not lucky enough. Ultimate High was released in stores nationwide three months ago. So far, it has sold only 378 copies - amounting to about $ 4,900 at its suggested retail price. In many other industries, this would be considered an extraordinary bomb. But in today's troubled music business, it's routine. Of the thousands of albums released in the United States each year by the five major record companies, fewer than 5 percent become profitable, music executives say.
I mostly included that because the "U R FAT" part is funny. Here's the music industry's sob story (notice how it's more sophisticated than the "Napster and CD burning are eating our sales" PR crap that is trotted out for consumer-oriented press):
Music executives also say it is more difficult to launch new acts. Among the reasons: Deregulation of the radio industry in 1996 has led station owners to consolidate into a few big companies, which are under pressure to maximize profits and pull songs off the air that aren't instant nationwide hits.
Superstores such as Wal-Mart, which stock fewer titles than traditional music stores, are the fastest-growing segment of music retailing, making it costlier and more competitive for record companies to secure prime shelf space.
As a result, industry executives estimate that major-label releases must on average sell about 500,000 copies just to break even. Last year, of the 6,455 new albums distributed in the United States by major labels, only 112 have sold at least that many, according to SoundScan, which monitors music sales.
Golly, if only there were some way for those poor big 5 to break the stranglehold on music distribution caused by the Wall-Mart/radio syndicate oligopoly! If only there were a way to reach customers directly, or to encourage the development of new radio stations, or to allow individuals access to more promotional music I'm sure the record labels would jump all over it.
Overall music sales were down 5 percent last year - the steepest decline in a decade.
After the largest yearly increase in a decade the previous year. What happened last year? Napster was shut down. What happened the previous year? Napster ramped from almost 0 users to 80 million registered users. Hmm.
Hennessy, a native of Dublin, Ireland, released her debut musical effort, Carly's Christmas Album, in Ireland at age 10, after performing in Europe as Little Cosette in Les Miserables...
"The most beautiful voice you'd ever heard - and she would have ended up singing in the bath," says her father, Luke Hennessy, a real estate investor.
More local color.
The executives offered her a six-album contract, under which Hennessy would get a $ 100,000 advance for her first album, plus $ 5,000 a month in living expenses while the album was being made. The label would own the recorded music and would front the cost of recording and promotion.
For Hennessy to make any more money, the label would first have to recoup its advance, its recording costs and half the cost of any music videos, as well as her living expenses - meaning the album would have to sell between 500,000 and 700,000 copies, MCA says.
At that point, Hennessy could collect royalties amounting to 15 percent of sales. But she would still owe a cut to a phalanx of producers and managers, as well as other record-company fees - leaving her with at best about 80 cents to $ 1 per album, MCA says.
Such contracts have drawbacks for both sides. Artists can be unceremoniously dropped if they don't live up to expectations. But if they blossom into superstars, they can use their new leverage to demand that their contracts be rewritten to pay them much more.
Hennessy says she let her managers, including her father, worry about the financial details. "Pretty much I was like, 'Is this a good contract, or a bad contract? OK, it's a good contract,' " she recalls. She was not even sure how many albums she owed MCA.
Interesting details.
Ok, let's track that $2.2 million, shall we:
Hennessy and her producer, Dorff, spent about three months recording eight songs, including several he had written. The total tab, including studio time, musicians' salaries, producers' fees and Hennessy's living expenses, was about $ 350,000 - typical for a first pop record, MCA says...
In April 2001, MCA decided to try to get Hennessy notice by releasing her first single, a bouncy tune called I'm Gonna Blow Your Mind. Its opening lines:
"I really really, I really really, I really really, I really really, I really really want to kiss you/
But much more than that/
Boy, I'm gonna blow your mind."
It was a risky choice. MCA realized the song's subject matter - oral sex - made it unlikely to get much exposure on youth-oriented outlets deemed important in launching young artists, such as the Radio Disney network of stations.
Tee hee hee.
But executives felt it was Hennessy's catchiest song. MCA spent $ 250,000 on a video that showed Hennessy dancing in a disco and jumping around with pals in their sleepwear...
The label also earmarked about $ 200,000 to hire independent promoters - middlemen who use their influence with radio program directors to secure airplay.
In addition, MCA spent about $ 100,000 on "imaging" for Hennessy, including photos, clothes and makeup artists. It sent Hennessy on a $ 150,000, four-week promotional tour, where she sang at malls over recorded tapes, backed by two dancers, and at station-sponsored concerts.
Um, so so far we're at $1.05 million. Unfortunately, the Chronicle version skips how the other $1.15 mil was wasted. Essentially what happened was they decided to scrap the first cut of the album and bring in all new writers and producers to write and produce her a new one, and that added another $300,000...and the other $800,000, IIRC, were mostly more payoffs for radio brokers (they tried floating a couple different singles) and the like, although a promotional trip to Europe might have been thrown in there too.
Oh wait: the article was also picked up by the Ottawa Citizen, which has a bit more detail. Re-recording the album actually cost another $600,000, meaning almost a million was spend on recording. And $500,000 more was spent trying to buy radio time for the second single, "Beautiful You".
Grand totals:
$1 mil. for recording, throwing away, and re-recording + living expenses $700,000 to promoters who try to buy radio time (unsuccessfully in this case; the album really sucks) $500,000 on various promotional tours.
And what's even more amazing than this tremendous tale of industry stupidity and obvious waste is that MCA actually *shared* all this information with the Journal *voluntarily* because they thought it proved their point about how they need to overcharge for CDs and rip off established musicians in order to pay for their flops. The shocking truth is that while the Carly Hennessy story has a few extra funny details, fully 19 out of 20 big 5 releases are complete flops just like hers.
And a huge huge proportion of that lost money, according to record execs, is the lost money they throw at radio stations to play their new single. Except that, as you point out, the record labels actually *make* money from radio, because of licensing on the songs radio *does* play. And, as there's only 5 record labels, it's not like there's any competition for this royalty money. Indeed you might almost think there's something slightly dishonest about spinning this money as losses when it is automatically going to be made up for by royalty gains on other songs.
Ugh. What a sad, sad story, and a sad, sad industry. And the real tragedy is, the way things are going it looks as if the labels will have just enough time before they put themselves out of business to buy laws that will prevent any sustainable models for distributing and creating music in the Internet age from being possible, even after the labels are gone. The end result may just be less music available, *and* the creation of a copyright police state.
This guy is virtually unknown to the general public. Does anyone out there think this system would extend well to established authors (or musicians)? As a counter-point, I'd bring up Stephen King's experiment, where he allowed free download of his book and asked for a tiny donation in return. Very few of the people who downloaded the book paid for it and the project was scrapped.
False. For those who've forgotten the facts (i.e. you) here's what happened: in the fall of 2000, Stephen King offered up every chapter of his book-in-progress "The Plant" for free download, with the caveat that if at least 75% of the people downloading each chapter didn't pay $1, he wouldn't release the next one. (The PDFs were encrypted to prevent uncounted freeloaders.)
He released 7 chapters. That means over 75% of people paid him the buck for 6 times in a row. "Very few" indeed. And this was hundreds of thousands of people, despite the fact that almost no one had a dedicated e-book reader at the time and, well, e-books suck. He made hundreds of thousands of dollars off this half-a-book. (Admittedly, this is less than he makes for his paper novels.) And despite the fact that he RAISED THE PRICE in the middle!! (To $2 AFAICT.) Many, many people paid $13 for half of a serialized mystery novel that there was a very real chance they would never get to finish (as indeed happened), even though they didn't have to pay a cent to get it.
[Caveat: I believe King fudged the numbers to allow slightly less than 75% pay rate at one point. OTOH, there were very very widespread reports of corrupted downloads, so many of the "freeloaders" were actually people who downloaded once, paid, and then didn't want to pay again just to download a working copy.]
My basic point is that this guy is getting free advertising by releasing the book for free, which is resulting in some more sales than he would have gotten if nobody had ever heard of him...But the situation is much different when you're talking about an established very well-known author..And the same goes for music. MP3s given away for free by small bands may increase their market..But does anyone hear Britney Spears for the first time on MP3 and think wow, that's great..lets go buy the album? Of course not..And the the RIAA/other publish associations know this, and will quickly discount this guy's story.
First of all, I think you're wrong again; except for a very small number of artists, nobody really has saturation exposure, particularly amongst people who don't listen to pop radio or MTV. Plus you're forgetting that the record labels *pay* very very large amounts of money to *get* these artists songs to be played on pop radio/MTV.
Although no, I don't expect the RIAA will be swayed by this or any other evidence, even the evidence that Napster fueled the largest increase in CD sales in history and that shutting it down directly caused the massive drop they are currently whining about. That's because the big record labels realize that even if they would initially make more money under such a system, they would lose their power. The reason artists sign with the big 5 is that the big 5 have a oligopoly on all the distribution channels; if you add a new distribution channel that cannot be easily controlled, then artists will sign with smaller labels that give them better contracts, or not sign with a label at all.
The sad thing is, even if such a system did decrease the sales of the top few artists (or even the total revenues of the music/publishing industry), that is still not a reason not to move toward it. The only reason we have copyright laws is to encourage as many artists as possible to go into the career of producing valuable art for the rest of us. This motivation is explicitly stated in the Constitution. The copyright laws are *not* there to maximize the earnings of the top artists or of the record labels, only to ensure that the largest amount of music is available to the public.
In other words, if a new system has the effect of increasing the number of people who can make a decent living creating art while decreasing the incomes of the top stars already in the industry, then by the criteria set forth in the Constitution our copyright laws ought to promote that system over the current one. Instead, all the evidence is that the way those laws are currently written prevents rather than promotes their stated purpose.
Works on Win98, IE6 with the latest security updates installed (even went to windowsupdate to double-check). I had a pretty good game of minesweeper at least. Damn, what a great game.
No, AOL has a vested interest in people using the internet. They have nothing to gain from helping the RIAA make music sharing difficult or irritating.
Except for the fact that AOL owns the fourth largest record label in the world, this would be a really insightful point.
Not to mention that that sentence could be read to suggest that AOL has a policy of making the Internet less annoying.
Note that a G4 1Ghz, pIII 1Ghz and duron 1Ghz all perform about the same.
Precisely. I was going to post the same link, but you beat me to it. While one can certainly diddle a bit about the particulars, this is pretty unimpeachable proof that the G4 is around 2 years behind the PC in terms of general purpose performance.
Proof that clock speed == power
Not at all. Check out the SPEC scores for a 1 GHz Alpha EV6--they're ahead of what a 2.2 GHz P4 can pull. Or look at SPEC scores for 500 MHz Alpha, PA-RISC, MIPS R14k, etc, all of which are roughly competitive with these 1 GHz scores.
Instead what it proves is that the nice PPC ISA (i.e. "RISCness", 32 GPRs, etc.) and relatively short pipeline (7 stages) of the G4+ aren't enough to overcome underpowered execution resources, an obsolete (PC-133) memory bus, and "Out of Order" processing resources so meager as to barely qualify for the term.
Of course, the G4's mediocre (but not quite horrendous) general-purpose performance is very much what you would expect once you realize that the G4 is not meant to be a PC CPU, but rather a chip for embedded signal processing systems. These designs represent the strong majority of Moto's 74xx (i.e. G4) sales, and in fact the G4 is a darn good chip for this market. Because of...
AltiVec. AltiVec is apparently miles ahead of the SIMD resources of x86 chips, even though the latter are getting pretty good. Since most signal processing systems use SIMD-friendly algorithms, the PPC7400 family does great there. But whenever you design your chip to maximize one aspect, you're going to hurt the others--in this case, the aggressive AltiVec design hurt the general-purpose capabilities of the G4 by limiting top clock speed and by taking up much vaster amounts of chip real estate than e.g. the SIMD resources on a P4.
The only acception would be programs like photoshop that are optimised for mac (but how many of these programs really exsist?)
Ah, the $64,000 question. One answer is: surprisingly a lot. In addition to Photoshop, iTunes' MP3 encoding is apparently blazingly fast due to AltiVec, and many of the mathematically intense transforms that make OS X so pretty have recieved AltiVec rewrites as well. (Which is why OS X is nearly unusable on a G3 unless you turn the pretties off.) Presumably voice recognition and video encoding/decoding (although the latter also requires memory bandwidth, something the current G4 platforms sorely lack) can benefit significantly as well.
Then there's 3D games, which can only make limited use of SIMD optimizations (mainly in graphics drivers), but which, on the other hand, are often more dependent on the video card for performance than the CPU. (Although not to the degree that many people think, especially with the newest games.)
At first glance, the above list seems to cover most of the compute-intensive applications that the average consumer might perform on her PC. In other words...sure the G4 is beyond pitiful at the sorts of bandwidth-intensive SISD double-precision float workloads that form the bulk of SPECfp; but so long as nobody's trying to run physics simulations on their G4's (especially without rewriting them to be SIMD if possible), who cares? Sure the 1 GHz G4 isn't going to be any faster than a 1 GHz PIII running SISD programs like Office...but both are more than fast enough for almost all Office tasks. Right?
Well, maybe. Personally, I don't really buy it. For one thing, while there are a number of apps which have been recoded for AltiVec, the vast majority won't be or can't be. In particular, I quite seriously doubt any major open source software projects exist which really take advantage of AltiVec and, unless gcc suddenly grows a (good) auto-vectorization engine really soon, none ever will. Both because most open source programs aren't really amenable to SIMD execution, and because the LinuxPPC community is small and developer resources are tight. At this point, anybody who buys a Macintosh to run Linux on it deserves to have "MORON" tatooed on their forehead.
Moreover, and maybe more importantly, while the number of apps which have some AltiVec routines in them is getting pretty large, it's important to remember that most of the features of these programs can't be rewritten in an SIMD paradigm, and much of what can be rewritten isn't such a great fit that tremendous speed gains are inevitable. Thus, Steve Jobs can always pick 3 or 4 Photoshop filters at which a G4 can run circles around a top-of-the-line P4, but it turns out that if you test the two in a benchmark suite containing *every* Photoshop filter, a 2.2 GHz P4 will win hands down. And then there's that pesky PC-133 and the 1999-style hard drive options shipping with standard Macs (ATA/66 anyone??); add in the time for the application to launch or to open a giant file (100+MB files are of course very common in Photoshop), and the PC blows the Mac away.
Of course the Mac has many strengths: an OS which is gorgeous on top and is UNIX underneath; a suite of strong, free (beer) media manipulation apps like iTunes, iDVD and iMovie; excellent upper-tier media apps like Photoshop and Final Cut Pro; FireWire ports standard; integration with the coolest MP3 player around. Once you take heat/power issues into consideration, the G3/G4 suddenly begin to hold their own with the x86 counterparts; the G4 PowerBook in particular is quite an attractive product. Yes, you can find an x86 laptop as thin and light and fast and with as much battery life for less (but without the big screen), or one with the same screen and a much faster CPU and video chip for less (but without the light weight and long battery life), but it's difficult to find another laptop that beats it all-around. (Although getting close every day; the other disadvantage of Apple products is that their specs stay put for 6 months or more at a time.)
But in terms of raw performance, the G4 is simply outclassed. Perhaps the mythical G5 will arrive soon to even things up (read: lessen the gap). But just because a "G5" based embedded chip is sampling from Moto doesn't mean the Mac is getting a G5 any time soon. Plus there's the rather strong possibility that Moto's semi division will be killed/spun off. (Of course the recent revival in the semi industry makes this less likely.)
Until then, Apple would do well to get DDR mobos here (yesterday, if possible), continue their good work converting anything they can to AltiVec, and...sad to say, but...continue spouting the lie that dual-CPU configurations and AltiVec can actually provide enough performance boost to elevate a new Mac to the performance level of a 6 month-old PC costing half the price (then).
(There are problems that Turing machines can't solve - the classic one is the halting problem. Can you write a program that can look at any other programs (not just most programs, ANY program in the infinite universe, including itself) and say if it'll lock up or not? With a Turing-equivalent machine, it's fundamentally impossible to do that - not just improbable or hard-to-implement, but impossible. The solution is to create a machine that's more capable than a Turing-equivalent machine, but that's rather tricky with macroscopic physics.)
it seems as if you're implying that a quantum computer would be more than Turing complete, and/or might be able to solve the Halting problem. This is wrong. It *may* be the case (AFAIK this hasn't been proven yet) that a quantum computer of arbitrary size could be used to solve all problems in NP in polynomial time/space in the length of the inputs (at least, if you don't believe the many-worlds interpretation of quantum theory) even if NP != P. In other words, a quantum computer of arbitrary size may be able to serve as the oracle that you need for the naive way to solve any problem in NP in polynomial time, even if a Turing machine could never solve NP-Complete problems in poly time in the length of the inputs.
This is *not* the same as being able to solve an incomputable problem--i.e. those which a Turing machine may take infinite time to solve (the original example being the Halting problem)--even though it would represent the first greater-than-polynomial improvement on the Turing machine yet. (Assuming that a quantum computer is, in fact, equivalent to a nondeterministic Turing machine.)
If your company has "Product A", and another company attempts to create "Product B" to replace your product you're not going to be happy.
Yes, but trademark law is not about making the owner of the trademark happy. It's about making society as a whole happy.
Competition is the basis of any free market system and thus our laws are (theoretically) always set up to encourage more fair and transparent competition. Indeed, the Constitution specifically states that copyrights, trademarks, patents, etc. are provided only for their benefits to society, not (unlike, e.g. the rights to free speech or physical property) because they are inherently good in and of themselves.
To this end, the basic rule of thumb with trademarks is that Product B is violating the trademark on Product A if and only if potential customers are likely to be confused into thinking that Product B *is* Product A, or the company that makes Product A makes or has given their support to Product B. That's it. The company that makes Product A might wish for more, but their wishes would be at odds with the wishes of society as a whole.
And if you think about it, this is a very sensible rule, given that the intent of trademarks is to increase fair and transparent competition, not to offer barriers to competition with the trademark-holding company. Of course, if there were no trademarks at all, then "competition" would be increased, but it would not be transparent--the customer would not know whether they were getting the genuine article or a rip-off. Thus things like fake Gucci handbags are violations of trademark law (especially if they say "Gucci" on them or copy the Gucci logo or a logo which has clearly been designed to confuse customers into thinking it's the Gucci logo). But copying the overall design, color scheme, attitude, etc. of Gucci bags, or of any fashion designer, is legal and encouraged.
I think your confusion stems from the widespread use of the term "intellectual property" to describe copyright, trademarks, patents, etc. The term is semantically dishonest, because it makes it seem as if copyrights, trademarks, patents, etc. confer on their owners the same privileges that ownership of property does, when in fact nothing could be farther from the truth. The privileges conferred by copyrights, trademarks, patents, etc. are quite limited, and explicitly delineated (as opposed to the privileges conferred by owning physical property, which are expansive and only the *limitations* of which are explicitly delineated).
When you start thinking of copyrights, patents, trademarks, etc. as the (intellectual) "property" of the individuals or companies involved, you start thinking that they get to limit the way that their patented/trademarked/copyrighted product can be used to just those uses which they approve of. (Sort of like real property.) But, again, that's not what the law says, nor has it ever been the intent of the law; instead, the intent of the law is to prevent certain specific uses of the product or ideas associated with the product, such that the people who come up with those ideas are given a decent financial incentive to continue coming up with ideas, and such that the market remains more transparent to the public. (Patents do allow the patent-holder pretty wide lattitude to decide the uses to which their invention can be applied, but in exchange for that, patents are much more difficult to get than copyrights or trademarks, and last for a shorter time.)
Of course, if you start asking yourself "well, if that's the case, how on earth could laws like the Sonny Bono Copyright Extension Act"--which extends copyright to the life of the artist + 75 years, even though a much shorter term would never stop any reasonable person from creating and releasing a copyrightable work--"or the DMCA"--which, under the banner of copyright grants to copyright holders the ability to restrict the uses of their products to a degree much greater than the restrictions granted by copyrights, and even makes it illegal to attempt to use the product in ways which are legal but which the copyright holder does not like--"be constitutional?" The answer is that they're probably not, and simply haven't been around long enough to recieve a challenge at the Supreme Court level (IIRC there's a Sonny Bono case appearing before the Court now). What's scary is that the fact that these laws and other exist has confused many people into thinking that they must be constitutional, and thus into thinking that copyright (and, by extention, patents, trademarks, etc.) confers many more priviliges than it actually does. And, worst of all, thinking that it "ought to".
Slashdot Mirror
"The development/test platform is Windows 2000" for Doom 3.
And lets keep in mind that ALL the games that he's worked on up to and including Quake 2 have been GPL'd by now.
Very good point. (Not technically correct, as neither the id branded Commander Keen games nor any of the non-id branded games he has worked on are GPL'd...but all the important ones, yes.)
I have heard that Carmack actually develops on a Linux box, but that's just speculation.
No, he uses NT. Perhaps 2k or XP these days, but you'll notice he says "NT is definitely going to be the primary development platform for our next project, but I will be evaluating alternatives for a possible transition after that." That "next project" is Doom 3. (The interview was from Oct '99, late in the Q3:A development process.)
Nope, the first Q3 test was for Linux; after that the Mac and windows versions.
Bzzzt, wrong. (Sorry, had to say that.)
Q3test 1.0 was released for Mac on April 24, 1999; the first q3test release for Linux, 1.03, came out on April 29. The first q3test client for Win32, 1.05, was released May 10.
If they decide to release a linux version of Doom3, and given Carmack's good attitude towards open source and OpenGL, I really really would love if they go and piss off Mr Gates by releasing the test for Linux first.
Could happen. For Q3, id released the first test for Mac only, to keep the beta testing group small enough for the first round. (Later tests were of course released for Windows, once the first round of bugs was ironed out.)
I bet that a zillion gamers would install Linux just to be able to test Doom3. They have been waiting for years!!
Wouldn't that be a hoot. Anyways, that just might be the reason for id *not* to release the D3 test on Linux first; it's a lot easier and cheaper for gamers to install Linux to run the D3 test than it is to install Mac OS X, so the test group might not stay small enough for their purposes.
OTOH, the main purpose of the Q3 test was to iron out netcode issues, stuff they couldn't easily test in the lab. As D3 is focused heavily on single player and probably will not contain much new in the way of netcode, they might even do without a test altogether...
Also, does anyone know if there will be a supported version of Doom3 for Linux, or will we be relying on ported versions?
id has released nearly simultaneous Linux binaries for all of their games (client and server) since Quake 1, and released Linux patches for Doom 1 and 2 as well. Loki was involved inasmuch as they published a retail box of Linux Q3; however, this was never really important because you could always get Linux Q3 by buying the Windows version and downloading a small patch from id. (Indeed, the retail Linux version sold poorly, probably because it was released a couple weeks after the Windows version and thus many people went the buy-Windows-and-download-patch route.)
I believe id has officially announced that Doom 3 will available for Linux (and Mac), but if not it's still a virtual certainty. id has always been a tremendous supporter of open standards; Carmack chose OpenGL over DirectX for Quake (and thereby single-handedly created the consumer OpenGL market), and in addition to working on Mac, Linux and Windows versions of all 3 Quake games simultaneously, released Doom ports for Next (id developed on Next workstations back then), Solaris, IRIX (I think, or maybe that was unofficial), and I believe even Linux on Alpha in addition to the already mentioned x86 Linux ports.
Again, id has always done the port themselves; most likely, you will have to buy the Windows version and download a patch which will almost certainly be available within days of release.
Their major concern seems to be frame rates in SS and Q3A, two games built on old technology.
Exactly the point: if Parhelia dips as low as 35 fps running SS @ 1024x768 with nothing turned on, then you know for a fact that it will be near-unplayable on newer more demanding games.
But, since the reviews available test a whole lot more than just SS and Q3 engine games, we don't need to rely on that obvious deduction. Parhelia also gets its clock cleaned on newer games like Commanche4, and UT 2003, as well as the mini-games in 3DMark, which are supposed to simulate future game engines. The only difference is, with these games you actually need a GF4 to get decent performance.
What I'm concerned about is high-resolution performance with AA enabled. I have no intention of ever again running a game below 1024x768 with AA enabled. Why would you, when the Parhelia can do it without breaking a sweat?
Because the only games that Parhelia can run at 1024x768 with 16x FAA "without breaking a sweat" are the ones "built on old technology" that you denigrated a sentence ago. Check out the scores for UT 2003 with AA and anisotropic filtering. Parehlia is most definitely breaking a sweat @ 1024x768. (So, of course, is the GF4.) This is for a game that is going to be out in a month; and not only that, but a deathmatch-oriented game which is most definitely tweaked for high framerates. If you never want to run a game below 1024 with AA, then you better either get used to the games that are already out today, or prepare to upgrade video cards very very often.
I'd like to see some benchmarks of the Parhelia running DooM3 at 1024x768 w/ 16xAA.
Too bad: using current drivers, it won't run it. That's because Parhelia's 16x AA is fragment anti-aliasing, which only AA's the edges of polygons, and thus can't deal with a stencil buffer which is used in Doom3 (and many other games) to render cast shadows. Even assuming a driver fix will allow Parhelia to at least run games with a stencil buffer in FAA mode, the shadows themselves will still have jaggies unless you use the much much slower FSAA mode. Given how large a part shadows play in the Doom3 experience, it's doubtful FAA will be better than no AA at all.
If your goal is to never run below 1024x768 with AA, your best bet is probably to buy a GF4 MX 440 today, ride that as long as it will last, and then upgrade in 12-15 months (before you need a DX8 compatible card, which GF4 MX is not) to whatever the best $150 card then is (i.e. equivalent of today's GF4 Ti4200 or Radeon 8500). You'll get your Doom3 at 1024 with AA, and save $175 over buying the Parhelia today to boot.
I wonder what The Carmack has to say about this card.
Me too; in particular, he (or someone) should be able to inform us where the expected throughput benefit for highly multi-textured games has disappeared to. Unfortunately, given the shrug gamers are going to give this card after this morning, he may not even bother.
Sorry to reply to my own post, but some additional news on the DOJ DRAM-industry investigation. First, since yesterday it's been announced that many other memory makers have recieved DOJ subpoenas, including Elpida, Toshiba, Nanya, Hynix, and even little Winbond. It's quite clear now that this has nothing to do with Micron in particular but rather the DRAM industry as a whole.
Second, it's been noted that, sad to say, the DOJ generally only undertakes antitrust investigations in response to a complaint by a US interest claiming to be hurt by the alleged anti-competitive behavior. Since the only major party (that I can think of) hurt by the DRAM price war was Hynix, a South Korean company, it's difficult to imagine on whose behalf the DOJ would be investigating.
Thus it has been suggested that instead the DOJ is investigating the jump in DRAM prices following the end of the price war (i.e. when the Hynix takeover was announced); Michael Dell in particular has publicly complained about high DRAM prices, so some have suggested Dell is behind the complaint.
IMO, if the investigation is focusing on the SDRAM spot prices which shot up after the price war ended, it's misguided. That jump in prices was, IMO, the natural consequence of both the end of an artificial price war and uncertainty on the demand side as Intel introduced its first DDR motherboards and the market had to be sure to have adequate inventory of both SDRAM and DDR to cover for the possibility of either a very successful DDR ramp or a slow one.
Now, if the complaint focuses on the rather outrageous gap between spot and contract prices which existed until it finally close in the last couple weeks, then it might get somewhere. On the one hand it's pretty clear that the pricing anomaly was that the spot price was artificially low, not that contract was artificially high. On the other, there was apparently good evidence that the major memory makers were indeed colluding to keep contract prices that high, so the DOJ might indeed find something to complain about...
The Micron story: First, the DOJ did not bring an action against Micron today. Rather, they sent a subpoena to Micron, seeking information for an investigation into possible anti-competitive behavior in the DRAM market.
/. posts make it out to be. For one thing, Rambus claims that they didn't run afoul of the JEDEC policy requiring attendees to disclose relevant patents, because they did not yet have the patents which they later accused everyone of infringing. What makes the whole thing really complicated is that these "SDRAM patents" were not new patents either, but rather extensions of a previous Rambus patent on RDRAM, which Rambus did have at the time. Basically, patent law definitely allows you to file an extension of an old patent which makes clear that a new product infringes your original invention even if it may not infringe the original patent. How this intersects with the requirement to disclose patents to standards bodies is rather cloudy. As we know, Rambus got convicted of fraud for this--probably helped by documents at trial which pretty much proved Rambus had this plan in mind all along. However the issue is currently on appeal. IMO the Appeals Court's decision on this issue will determine what happens with this FTC claim filed today.
This has nothing to do with Micron (with global DRAM marketshare of ~18%) having a "monopoly" in the DRAM market; indeed, in addition to Micron, Infineon and Samsung (and probably others) recieved subpoenas today as well. While the DOJ has not made any public statements about this (that I've seen), it is extraordinarily obvious to anyone who has followed the DRAM industry over the past year or so that they are investigating allegations of dumping on the DRAM spot market. In particular, the spot price for DRAM (128 Mbit SDRAM chips in particular) declined by something like 75% last fall, in the absence of major new supply or significant shortfalls in demand. The resulting price was well below even the marginal cost of producing these chips.
When the price suddenly shot back up a few months ago with the announcement that Micron had reached a tentative deal to buy the already-bankrupt Hynix, it became quite clear that the spot market price was being kept artificially low in order to try to force the ailing Hynix out of business. (The Micron-Hynix deal was later rejected by Hynix's board, causing--wouldn't you know it--another DRAM price war.) The whole thing was made even more fishy by the fact that DRAM contract prices (that is, the DRAM sold in monthly contracts to OEMs like Dell, HP, etc.) stayed way way higher than the spot prices; Hynix was mostly limited to the spot market in its attempts to desperately sell off excess inventory to try to raise cash to pay off their creditors (after all, who wants to sign a contract with a company on the verge of bankruptcy??).
The obvious conclusion is that the other memory makers (Samsung and Micron are #1 and 2 in marketshare) dumped excess inventory on the spot market or even purposely sold at worse prices than they could otherwise get, in an effort to drive Hynix out of business sooner and thus reduce supply and cause prices to be higher than before. By itself, this is probably not illegal, because none of them has a monopoly. What would be illegal is if any of them colluded to keep the spot price artificially low. If they did that, they would be guilty of forming an illegal trust (that's why it's called anti-trust law, not "anti-monopoly") and engaging in illegal dumping or even price-fixing. Personally I would be surprised if this did not occur, but we'll see if the DOJ can find enough concrete evidence to bring a claim.
[For those wondering how the DRAM market turned out--spot prices are way down again across the board, although 128Mbit SDRAM is not nearly as low as it got last fall. (256Mbit and DDR chips are close to record lows, though.) The difference now is that contract prices have finally come down to near-spot levels (in a normal market, contract prices are lower than spot levels), and, considering the recent warnings by Intel and AMD, it looks as if low demand may actually be to blame, not price-fixing.]
The Rambus story: The short version of events posted here is more or less right: Rambus was a member of JEDEC while the SDRAM standard was being worked on; Rambus did not disclose any patents they had which read on the SDRAM design being discussed, nor did they disclose any intent to seek royalties on the standard; later, after SDRAM was standardized and became the mainstream DRAM type, Rambus sought royalties from anyone producing SDRAM or an SDRAM memory controller, which is to say almost the entire worldwide electronics industry. (Also DDR, which according to Rambus not only infringes their SDRAM patents but also a couple others on double-clocked operation in a DRAM.)
Some memory producers (including Samsung and Elpida) agreed; others (specifically Micron, Infineon and Hynix) refused and were taken to court. (The threatened legal action regarding SDRAM controllers never really happened because Rambus was focusing on memory producers first.) In the first such case to come to trial, in which Rambus sued Infineon in Virginia, not only did the judge rule that Rambus' patents don't actually cover SDRAM/DDR, but the jury found Rambus guilty of fraud for sitting on JEDEC and failing to disclose their intent to seek royalties on SDRAM--which is essentially what the FTC claim charges. (Unlike the DOJ investigation I talked about above, the FTC is past the investigating stage and today actually filed a claim against Rambus.)
However, it's not quite as simple as other
I am incredulous to anyone comparing copyright infringement on Napster to such an event.
...
I merely questioned the sanity of someone that wanted to compare copyright infringement to events of that magnitude.
Most of the events listed are only large in magnitude in hindsight. At the time things like the Boston Tea Party (~50 radicals commit vandalism in the midst of a boycott involving a few thousand people) or Rosa Parks' refusal to give up her seat (1 person acting as a test case to spark a boycott involving, again, a few thousand people) were small in magnitude. Or (promised I wouldn't get into it but) Jesus walking around Judea and preaching a new radical brand of Judaism. Only as part of the larger historical movements they are associated with is it clear that the magnitude of their influence was enormous.
By any measure, the magnitude of Napster was much, much larger. 80 million users signed registered with Napster. Billions and billions of songs were traded. Unlike all the events listed, with Napster a substantial portion of those affected by the law actually joined in the "movement" against it.
Now, whether history will judge Napster to have a magnitude nearly as large as what we now credit to things like the Boston Tea Party; well, it's too early to say.
Oh, and fancy phrases will not candy-coat the ENORMOUS magnitude of difference there is in claiming to OWN a human, to feed or not, violate sexually or not, let live or not, to STEALING music. Music theft is what I would term petty. It is stupid.
The comparison is not between the ability to own a person and stealing music but between the ability to own a person and the ability to own an idea. I, like most people raised in the Western intellectual tradition in the past 50 years, of course find owning a person the much more abhorrent idea. But the simple fact is that that is not the response that most people in most civilizations have had. Nor is it clear that the ability to own an idea is not also horrible, albeit something we are nonetheless rather used to.
I also never claimed to AGREE with the copyright laws. All I claimed was that they ARE laws, and stated that the correct way around the problem was to get the law changed!!! Breaking the law is NOT going to get it done. Wake up.
Wake up yourself. Part of the point of the original post was that, in many instances in history, breaking the law was a vital step in the movement to get the law changed (or to have some other similarly far-reaching effect).
you are comparing music theft with Jesus' acts, the boston tea party, and the freedon of slaves?
Thanks for making my point. You really did just fall off the turnip truck.
Are you comparing the Boston Tea Party with Napster?? The Boston Tea Party was an act of outright vandalism in which anti-government reactionaries protesting a somewhat obscure point regarding differential tarriffs and government monopolies broke into ships which were not their property, and destroyed valuable physical property which not only did not belong to them but which other people wished to purchase and now could not. Indeed, the Boston Tea Party was the climactic event in an illegal embargo carried out against colonists as a whole by a small group of radicals who decided that the law regarding tea taxation was wrong and thus that all imports of British tea were to be blocked through threat, vandalism and violence.
Napster, on the other hand, did not involve stealing, vandalism, or destruction of physical property. It almost certainly contributed heftily to the bottom line of the record companies and musicians who brought complaints against it. Use of Napster to share (but not to download) copyrighted music was found to constitute illegal copyright infringement; however, this hinged on a technicality of the 1992 Audio Home Recording Act which did not include mp3s as being under the purview of the rule that noncommercial copying and distribution does not constitute infringement (before the AHRA, all noncommercial copying and distribution were considered fair use, so the AHRA was sharply broadening copyright law as it had been understood for centuries). Unlike the Boston Tea Party which was carried out by a small group of radicals, probably against the wishes of and certainly with economic harm to the large majority, Napster was used by tens of millions of Americans; perhaps an absolute majority of all Internet-connected Americans, and certainly an absolute majority of those with broadband.
The actions taken by the perpetrators of the Boston Tea Party were and are and always will be unambiguously illegal. No society can survive long on the principle that it is okay to break in and destroy others' commercially significant private property as a means of protesting government policy. Napster, on the other hand, was found illegal only after significant court action, and would not have been illegal under copyright law as it existed for centuries prior to 1992.
So why can we view the Boston Tea Party as an admirable event of history? Because it was carried out in the name of greater ideals, namely the proposition that all citizens deserve a voice in the laws that govern them. Note that this proposition is not obviously more correct or more noble than the proposition underlying Napster use, namely that people have the right to share artistic/intellectual content with each other provided they have purchased the means of reproduction and distribution; or, perhaps, that people have the right to use things they have purchased (i.e. CDs, computers and Internet access) as they see fit so long as their actions do not directly hurt anyone else. In any case, the point is that the Boston Tea Party proposition has since been judged by history to be correct (of course, history is written by the winners), whereas the Napster proposition(s) are perhaps almost as controversial today as the proposition that people have a right to self-government was 230 years ago.
As for slavery, we might note that it is not prima facie obvious that allowing people to be owned as private property is any more or less unusual or immoral or against the tenets of natural law than allowing ideas to be owned as private property. Indeed the vast majority human societies throughout history have considered people a valid form of property and ideas not.
As for Jesus...I'm not touching that with a ten foot pole.
But it arguably has the greatest overall loss to society, if you consider intellectual achievements and discoveries to be superior to physical and materialistic ones. Just as normal theft acts as a disincentive toward regular commerce, piracy works as a disincentive toward developing IP that requires a substantial amount of time/effort/money. I could see why potentially robbing the world of future discoveries might be considered the worst form of theft.
But the notion of intellectual "property" itself arguably has the greatest overall loss to society, if you consider intellectual achievements and discoveries to be superior to physical and materialistic ones. Just as normal theft acts as a disincentive toward regular commerce, the notion of intellectual "property" works as a disincentive toward developing new ideas that require the freedom to use and build upon old ideas as the scientific method and all serious forms of intellectual discourse have always done (that is to say, all ideas in a sufficiently advanced society). I could see why potentially robbing the world of future discoveries might be considered the worst form of theft.
Doom took a year? If I remember correctly, Doom was about a year late when it came out...am I wrong?
I believe so. The original design document (pdf) for Doom is dated 11/28/92. The release date was 12/10/93.
Obviously work was started before the design document was finished, but as you can tell by perusing it, the game was still being sketched out at the end of November, 1992. Several reports say that iD began Doom only after the Wolf 3D expansion Spear of Destiny was released, which was September, 1992. So, if we take that as the start date, it looks like about 15 months.
As for an iD game being a year late, I'm not sure, but you may be thinking of Quake 1?
there was a fundamental change somewhere in the last 15 years where things shifted from games using existing hardware fully to where games became the reason themselves to create new, faster hardware devices.
Not that this is bad, nit by any means, but it does give one interesting meat to consider; no one will argue that games are what's driving things like new video card technologies -- when did the chicken outdo the egg?
When games started taking a long time to make. Used to be even a revolutionary game could be made from start to finish in months. Wolfenstein 3-D took about 6 months to make. The original Doom took about a year. iD has been working on Doom3 for about two years and they've still probably got at least a year to go.
When Carmack decided the technical parameters of the Doom3 engine back in 2000, he would have been an idiot if he designed it to only take advantage of the features of existing hardware. Instead he designed it to use features and require performance which he knew would be entering the mainstream by the time the game was released.
Of course, Carmack is unique in that he can actually influence Nvidia, ATi, etc. a bit into supporting the features he wants. On the other hand, you have to realize that back in 2000 Carmack, Nvidia, ATi, etc. already had a very good idea what sort of features would be supported in the 3D cards of 2003. Of course there is some guessing and tweaking involved (which Carmack seems to be particularly good at), but a good 3D engine designer has to design for the hardware of the time when the game will be released, not the hardware of the time when he's designing.
As for why games take so much longer these days, that's another story but the basic point is that they are not only more complex technically but that there is ever increasing detail in the art, scripting, level design, etc. that it takes much larger teams with much better tools much longer to make a game than in the old days.
I wonder if any of the current laptops will be able to run Doom3... I'm considering buying a laptop with a GF4go as the Radeon7500 based ones seems to be slower... I wonder if its really worth it to go from 32 megs to 64 megs of ram?
Unfortunately, as both the GF4 Go and Mobile Radeon 7500 lack hardware pixel shaders, they will not be able to render Doom3 in its full glory. Of course they will be able to run it, but many of the graphical goodies will either be missing or will need to be (very slowly) computed on the CPU.
As for 32 vs. 64 MB, I'd go for the latter if you want to run Doom3. Surfaces in Doom3 can contain up to 5 texture maps, which means tons of RAM usage at anything but low texture detail. If you run out of room on the card, you need to store textures in main memory and access them over the AGP bus, which is too slow for that sort of thing. IIRC both the GF4 GO and Mobile Radeon 7500 are available with 64 MB, although I suppose one sometimes doesn't get the choice when buying a laptop.
Basically, the top-of-the-line 3D cards of today are going to be necessary to run Doom3 decently, so the top-of-the-line mobile 3D cards--which are about a generation behind the desktop--are going to be able to run it, but somewhat mediocrely. Of course, Doom3 probably won't be out for at least a year, maybe a year and a half. By that time you'll be able to buy a laptop which runs the game beautifully. If you have to buy a laptop now then it'll be a bit tougher. Kind of makes you wish laptop 3D cards were upgradable like desktop ones...
So how does copyright NOT imply a license?
A license is a contract between two parties, in this case the creator and the user. Copyright is a doctrine of federal law, designed to promote the creation and dissemination of original content and (shocking but true) their eventual release into the public domain. If anything, copyright is an agreement between the creator and the government, not between the creator and the user.
Put another way: the fact that I can't go up to you and punch you in the face does not imply that we "have a license". (Except in the very esoteric context of social contract theory.)
In any case, copyright is neither here nor there when it comes to the BSA and software "piracy". (ARRRRRRRRR, matey!) Copyright law prohibits someone who is not the copyright owner from making copies and distributing them. It does not prohibit someone from buying or otherwise obtaining a distributed copy. When it comes to downloading software off the Internet, it's a little murky, because you are in some sense causing a copy to be made even though it is the other guy's machine, not yours, that is doing the copying. (He/his machine is definitely infringing.) But this is a load of BS; by the same token, the use of a browser cache is flagrant and unambiguous copyright infringement.
Point is, using software that someone else burns to a CD and gives to you is not copyright infringement. If it's commercial software then it's probably against the license, though...
Is it a whole new form factor so everyone had to redesign the motherboards and to force incompatability with older systems??
The answer is...yes and no. DDR-II does have a new form factor, with 232 pins, as compared to the 184 currently used in a DDR (I) DIMM. Similarly, a DDR (I) DIMM has a bit more pins than the 168 found in an SDRAM DIMM.
However, this isn't such a big deal for the mobo makers; it's just a matter of putting on a different slot and different traces on the board. One of the things that has made the transition from SDRAM to DDR more "evolutionary" than a transition to (say) RDRAM would have been, is that the same chipset can control both SDRAM and DDR, because they use the same (or at least backwards-compatible) commands. It's not often mentioned on the hardware-enthusiast sites which are only interested in benchmarking the fastest stuff around, but just about every DDR chipset is also available with SDRAM. You need a different motherboard--because the slots are incompatible--but the cost to the mobo makers for offering both versions is pretty small.
In a similar vein, the commands for DDR-II are a superset of the DDR-I command set, such that DDR-II chipsets should very easily be able to detect and use DDR-I as well, just like DDR-I chipsets currently use SDRAM as well. Furthermore there is talk (dunno if it will happen) of releasing DDR-I DIMMs in the 232-pin DDR-II form factor; that way, one could buy a motherboard and use either DDR-II or DDR-I in it, with no problems. Of course old sticks of DDR-I will not fit, and the new ones will not fit in current DDR motherboards.
So, while such a scheme doesn't get rid of all the headaches of an incompatible upgrade path, it does address some, albeit more on the mobo producer end of things than on the IT inventory end of things. It is indeed a pain that, while SDRAM (and the SDRAM form factor) enjoyed around 4 years as the mainstream memory type, DDR-I will only be on top for 2 before DDR-II takes over. And DDR-II will be lucky to have 3 years before something pin-incompatible comes along to replace it.
On the other hand, the SDRAM -> DDR transition probably would have happened a year earlier if Intel hadn't tried to transition to RDRAM instead. And, meanwhile, being stuck with SDRAM for 4 years meant DRAM bandwidth only doubled (and actual performance did less than that) over a period when CPU clock speeds increased by a factor of 5 or so. I think the DRAM industry wants to make speed increases more frequent than they were a few years ago, even if this means more inconvenience for corporate IT departments.
This is important because industrial and corperate-mission-critical is older equipment. and an upgrade path for ram is still important.
I'm not sure what you mean here. As far as the upgrade path for greater RAM capacity goes, standard SDRAM and DDR-I will still be made and sold for quite some time, even if they will eventually be more expensive than their newer and faster brethren. EDO RAM is still being made and sold today.
If you mean an upgrade path for higher-performance, of course you can't just buy faster DRAM and expect it to speed up (or even work in) a system that was built to use a slower type. The system clock sets the DRAM speed, and unless the system has been validated to run at the new DRAM speed, doing so amounts to overclocking. An performance upgrade path for RAM is always going to require purchasing at least a new CPU module and accompanying memory bus, if not a new machine.
Why is it obvious? No-where have iD stated they'll be 'fixing' it. Just because it's something we all want doesn't mean it will obviously be fixed...
1) It's by far the most noticeable graphical flaw in a game that is staking its reputation more than anything on having the best graphics ever seen.
2) It's extremely common for a game at this stage of development to have very few models already built in (after all, the modeling tools and to some degree the engine have to be finished before serious modeling work can begin). And extremely common to make a demo using the few that you already have.
3) It doesn't fit with the premise (i.e. different people are possessed by Hellish, uh, stuff and turned into zombies) for the zombies to look alike.
4) It's something we all want, and id has more than enough time to fix it.
There are about 4 different monsters that I saw, and all other monsters were clones of these monsters. I mean, the big fat belled one, it's very cool and all, but does ID need to have 100's of the same monster running around? What's the logic here?
Doom3 is nowhere near done; the only release date they're specifying at E3 is "2003", and most likely it won't be January either. In other words, they have around a year's work left on this game, maybe more. Presumably they will spend that time adding many many more character models, or even (hopefully) coming up with a way to generate different bump/displacement maps for the same character and thus avoid the "two of the exact same zombie" syndrome that affects the current screenshots.
In any case, it's pretty obvious the issue will be solved in the final game.
And you're forgetting that the radio stations are now required to pay back the RIAA and record labels for every song played, per listener . The streaming radio stations are required to pay double the rate. Trust me, the RIAA is well into the black on their outlay for payola to the radio stations.
Very interesting point, which I'd failed to consider. Actually it is perhaps for that very reason that the record labels don't actually pay the radio stations directly for song placement, but instead pay brokers millions and millions of dollars which they then use to pay the radio stations off.
The reason I know that is that it comes from a truly fascinating Wall Street Journal article (unfortunatly not available online to the general public; I'm gonna see if I can't scare up a copy through Lexis-Nexis) about the millions of dollars poured down the drain by one of the big 5 in an effort to make some chick by the name of Carly Hennessy the next Britney Spears. I think in the end they spent something like $4 or $5 million on her and ended up with roughly 1,000 CD sales and a Wall Street Journal article to show for it. (It may go without saying, but her music SUCKED and I mean was AWFUL; here's a link to her album on CDNow where you, like me, can get a taste of how terrible it is. And I actually LIKE Britney Spears, or at least can appreciate what's well crafted about her good songs.)
Ah; OK, so apparently L-N (or at least the academic version) doesn't get the full text of the Journal, but the piece was syndicated out to the Houston Chronicle in abbreviated form. Although...they don't offer free access to archives either, so I'll just exercise that fair use thing and give some excerpts.
First off, I was wrong; they only wasted $2 mil on her, but then again it was only 378 copies of Ultimate High that got sold. Here's the opening of the story (as excerpted by the Chronicle):
MARINA DEL REY, Calif. - Eighteen-year-old recording artist Carly Hennessy is packing up her small apartment. Her promotional posters will go into storage, and the beige rental couch will be returned. A weight-control message that the slender teen scrawled in marker on the refrigerator - "NO, U R FAT" - will be wiped clean.
For two years, Vivendi Universal SA's MCA Records paid the rent while Hennessy prepared for pop stardom. And that's not all: The label so far has spent about $ 2.2 million to make and market her new album, an upbeat pop recording called Ultimate High. "Some people just struggle," she says. "I was very, very lucky."
Not lucky enough. Ultimate High was released in stores nationwide three months ago. So far, it has sold only 378 copies - amounting to about $ 4,900 at its suggested retail price. In many other industries, this would be considered an extraordinary bomb. But in today's troubled music business, it's routine. Of the thousands of albums released in the United States each year by the five major record companies, fewer than 5 percent become profitable, music executives say.
I mostly included that because the "U R FAT" part is funny. Here's the music industry's sob story (notice how it's more sophisticated than the "Napster and CD burning are eating our sales" PR crap that is trotted out for consumer-oriented press):
Music executives also say it is more difficult to launch new acts. Among the reasons: Deregulation of the radio industry in 1996 has led station owners to consolidate into a few big companies, which are under pressure to maximize profits and pull songs off the air that aren't instant nationwide hits.
Superstores such as Wal-Mart, which stock fewer titles than traditional music stores, are the fastest-growing segment of music retailing, making it costlier and more competitive for record companies to secure prime shelf space.
As a result, industry executives estimate that major-label releases must on average sell about 500,000 copies just to break even. Last year, of the 6,455 new albums distributed in the United States by major labels, only 112 have sold at least that many, according to SoundScan, which monitors music sales.
Golly, if only there were some way for those poor big 5 to break the stranglehold on music distribution caused by the Wall-Mart/radio syndicate oligopoly! If only there were a way to reach customers directly, or to encourage the development of new radio stations, or to allow individuals access to more promotional music I'm sure the record labels would jump all over it.
Overall music sales were down 5 percent last year - the steepest decline in a decade.
After the largest yearly increase in a decade the previous year. What happened last year? Napster was shut down. What happened the previous year? Napster ramped from almost 0 users to 80 million registered users. Hmm.
Hennessy, a native of Dublin, Ireland, released her debut musical effort, Carly's Christmas Album, in Ireland at age 10, after performing in Europe as Little Cosette in Les Miserables...
"The most beautiful voice you'd ever heard - and she would have ended up singing in the bath," says her father, Luke Hennessy, a real estate investor.
More local color.
The executives offered her a six-album contract, under which Hennessy would get a $ 100,000 advance for her first album, plus $ 5,000 a month in living expenses while the album was being made. The label would own the recorded music and would front the cost of recording and promotion.
For Hennessy to make any more money, the label would first have to recoup its advance, its recording costs and half the cost of any music videos, as well as her living expenses - meaning the album would have to sell between 500,000 and 700,000 copies, MCA says.
At that point, Hennessy could collect royalties amounting to 15 percent of sales. But she would still owe a cut to a phalanx of producers and managers, as well as other record-company fees - leaving her with at best about 80 cents to $ 1 per album, MCA says.
Such contracts have drawbacks for both sides. Artists can be unceremoniously dropped if they don't live up to expectations. But if they blossom into superstars, they can use their new leverage to demand that their contracts be rewritten to pay them much more.
Hennessy says she let her managers, including her father, worry about the financial details. "Pretty much I was like, 'Is this a good contract, or a bad contract? OK, it's a good contract,' " she recalls. She was not even sure how many albums she owed MCA.
Interesting details.
Ok, let's track that $2.2 million, shall we:
Hennessy and her producer, Dorff, spent about three months recording eight songs, including several he had written. The total tab, including studio time, musicians' salaries, producers' fees and Hennessy's living expenses, was about $ 350,000 - typical for a first pop record, MCA says...
In April 2001, MCA decided to try to get Hennessy notice by releasing her first single, a bouncy tune called I'm Gonna Blow Your Mind. Its opening lines:
"I really really, I really really, I really really, I really really, I really really want to kiss you/
But much more than that/
Boy, I'm gonna blow your mind."
It was a risky choice. MCA realized the song's subject matter - oral sex - made it unlikely to get much exposure on youth-oriented outlets deemed important in launching young artists, such as the Radio Disney network of stations.
Tee hee hee.
But executives felt it was Hennessy's catchiest song. MCA spent $ 250,000 on a video that showed Hennessy dancing in a disco and jumping around with pals in their sleepwear...
The label also earmarked about $ 200,000 to hire independent promoters - middlemen who use their influence with radio program directors to secure airplay.
In addition, MCA spent about $ 100,000 on "imaging" for Hennessy, including photos, clothes and makeup artists. It sent Hennessy on a $ 150,000, four-week promotional tour, where she sang at malls over recorded tapes, backed by two dancers, and at station-sponsored concerts.
Um, so so far we're at $1.05 million. Unfortunately, the Chronicle version skips how the other $1.15 mil was wasted. Essentially what happened was they decided to scrap the first cut of the album and bring in all new writers and producers to write and produce her a new one, and that added another $300,000...and the other $800,000, IIRC, were mostly more payoffs for radio brokers (they tried floating a couple different singles) and the like, although a promotional trip to Europe might have been thrown in there too.
Oh wait: the article was also picked up by the Ottawa Citizen, which has a bit more detail. Re-recording the album actually cost another $600,000, meaning almost a million was spend on recording. And $500,000 more was spent trying to buy radio time for the second single, "Beautiful You".
Grand totals:
$1 mil. for recording, throwing away, and re-recording + living expenses
$700,000 to promoters who try to buy radio time (unsuccessfully in this case; the album really sucks)
$500,000 on various promotional tours.
And what's even more amazing than this tremendous tale of industry stupidity and obvious waste is that MCA actually *shared* all this information with the Journal *voluntarily* because they thought it proved their point about how they need to overcharge for CDs and rip off established musicians in order to pay for their flops. The shocking truth is that while the Carly Hennessy story has a few extra funny details, fully 19 out of 20 big 5 releases are complete flops just like hers.
And a huge huge proportion of that lost money, according to record execs, is the lost money they throw at radio stations to play their new single. Except that, as you point out, the record labels actually *make* money from radio, because of licensing on the songs radio *does* play. And, as there's only 5 record labels, it's not like there's any competition for this royalty money. Indeed you might almost think there's something slightly dishonest about spinning this money as losses when it is automatically going to be made up for by royalty gains on other songs.
Ugh. What a sad, sad story, and a sad, sad industry. And the real tragedy is, the way things are going it looks as if the labels will have just enough time before they put themselves out of business to buy laws that will prevent any sustainable models for distributing and creating music in the Internet age from being possible, even after the labels are gone. The end result may just be less music available, *and* the creation of a copyright police state.
This guy is virtually unknown to the general public. Does anyone out there think this system would extend well to established authors (or musicians)? As a counter-point, I'd bring up Stephen King's experiment, where he allowed free download of his book and asked for a tiny donation in return. Very few of the people who downloaded the book paid for it and the project was scrapped.
False. For those who've forgotten the facts (i.e. you) here's what happened: in the fall of 2000, Stephen King offered up every chapter of his book-in-progress "The Plant" for free download, with the caveat that if at least 75% of the people downloading each chapter didn't pay $1, he wouldn't release the next one. (The PDFs were encrypted to prevent uncounted freeloaders.)
He released 7 chapters. That means over 75% of people paid him the buck for 6 times in a row. "Very few" indeed. And this was hundreds of thousands of people, despite the fact that almost no one had a dedicated e-book reader at the time and, well, e-books suck. He made hundreds of thousands of dollars off this half-a-book. (Admittedly, this is less than he makes for his paper novels.) And despite the fact that he RAISED THE PRICE in the middle!! (To $2 AFAICT.) Many, many people paid $13 for half of a serialized mystery novel that there was a very real chance they would never get to finish (as indeed happened), even though they didn't have to pay a cent to get it.
[Caveat: I believe King fudged the numbers to allow slightly less than 75% pay rate at one point. OTOH, there were very very widespread reports of corrupted downloads, so many of the "freeloaders" were actually people who downloaded once, paid, and then didn't want to pay again just to download a working copy.]
My basic point is that this guy is getting free advertising by releasing the book for free, which is resulting in some more sales than he would have gotten if nobody had ever heard of him...But the situation is much different when you're talking about an established very well-known author..And the same goes for music. MP3s given away for free by small bands may increase their market..But does anyone hear Britney Spears for the first time on MP3 and think wow, that's great..lets go buy the album? Of course not..And the the RIAA/other publish associations know this, and will quickly discount this guy's story.
First of all, I think you're wrong again; except for a very small number of artists, nobody really has saturation exposure, particularly amongst people who don't listen to pop radio or MTV. Plus you're forgetting that the record labels *pay* very very large amounts of money to *get* these artists songs to be played on pop radio/MTV.
Although no, I don't expect the RIAA will be swayed by this or any other evidence, even the evidence that Napster fueled the largest increase in CD sales in history and that shutting it down directly caused the massive drop they are currently whining about. That's because the big record labels realize that even if they would initially make more money under such a system, they would lose their power. The reason artists sign with the big 5 is that the big 5 have a oligopoly on all the distribution channels; if you add a new distribution channel that cannot be easily controlled, then artists will sign with smaller labels that give them better contracts, or not sign with a label at all.
The sad thing is, even if such a system did decrease the sales of the top few artists (or even the total revenues of the music/publishing industry), that is still not a reason not to move toward it. The only reason we have copyright laws is to encourage as many artists as possible to go into the career of producing valuable art for the rest of us. This motivation is explicitly stated in the Constitution. The copyright laws are *not* there to maximize the earnings of the top artists or of the record labels, only to ensure that the largest amount of music is available to the public.
In other words, if a new system has the effect of increasing the number of people who can make a decent living creating art while decreasing the incomes of the top stars already in the industry, then by the criteria set forth in the Constitution our copyright laws ought to promote that system over the current one. Instead, all the evidence is that the way those laws are currently written prevents rather than promotes their stated purpose.
Works on Win98, IE6 with the latest security updates installed (even went to windowsupdate to double-check). I had a pretty good game of minesweeper at least. Damn, what a great game.
No, AOL has a vested interest in people using the internet. They have nothing to gain from helping the RIAA make music sharing difficult or irritating.
Except for the fact that AOL owns the fourth largest record label in the world, this would be a really insightful point.
Not to mention that that sentence could be read to suggest that AOL has a policy of making the Internet less annoying.
http://www.heise.de/ct/english/02/05/182/
Note that a G4 1Ghz, pIII 1Ghz and duron 1Ghz all perform about the same.
Precisely. I was going to post the same link, but you beat me to it. While one can certainly diddle a bit about the particulars, this is pretty unimpeachable proof that the G4 is around 2 years behind the PC in terms of general purpose performance.
Proof that clock speed == power
Not at all. Check out the SPEC scores for a 1 GHz Alpha EV6--they're ahead of what a 2.2 GHz P4 can pull. Or look at SPEC scores for 500 MHz Alpha, PA-RISC, MIPS R14k, etc, all of which are roughly competitive with these 1 GHz scores.
Instead what it proves is that the nice PPC ISA (i.e. "RISCness", 32 GPRs, etc.) and relatively short pipeline (7 stages) of the G4+ aren't enough to overcome underpowered execution resources, an obsolete (PC-133) memory bus, and "Out of Order" processing resources so meager as to barely qualify for the term.
Of course, the G4's mediocre (but not quite horrendous) general-purpose performance is very much what you would expect once you realize that the G4 is not meant to be a PC CPU, but rather a chip for embedded signal processing systems. These designs represent the strong majority of Moto's 74xx (i.e. G4) sales, and in fact the G4 is a darn good chip for this market. Because of...
AltiVec. AltiVec is apparently miles ahead of the SIMD resources of x86 chips, even though the latter are getting pretty good. Since most signal processing systems use SIMD-friendly algorithms, the PPC7400 family does great there. But whenever you design your chip to maximize one aspect, you're going to hurt the others--in this case, the aggressive AltiVec design hurt the general-purpose capabilities of the G4 by limiting top clock speed and by taking up much vaster amounts of chip real estate than e.g. the SIMD resources on a P4.
The only acception would be programs like photoshop that are optimised for mac (but how many of these programs really exsist?)
Ah, the $64,000 question. One answer is: surprisingly a lot. In addition to Photoshop, iTunes' MP3 encoding is apparently blazingly fast due to AltiVec, and many of the mathematically intense transforms that make OS X so pretty have recieved AltiVec rewrites as well. (Which is why OS X is nearly unusable on a G3 unless you turn the pretties off.) Presumably voice recognition and video encoding/decoding (although the latter also requires memory bandwidth, something the current G4 platforms sorely lack) can benefit significantly as well.
Then there's 3D games, which can only make limited use of SIMD optimizations (mainly in graphics drivers), but which, on the other hand, are often more dependent on the video card for performance than the CPU. (Although not to the degree that many people think, especially with the newest games.)
At first glance, the above list seems to cover most of the compute-intensive applications that the average consumer might perform on her PC. In other words...sure the G4 is beyond pitiful at the sorts of bandwidth-intensive SISD double-precision float workloads that form the bulk of SPECfp; but so long as nobody's trying to run physics simulations on their G4's (especially without rewriting them to be SIMD if possible), who cares? Sure the 1 GHz G4 isn't going to be any faster than a 1 GHz PIII running SISD programs like Office...but both are more than fast enough for almost all Office tasks. Right?
Well, maybe. Personally, I don't really buy it. For one thing, while there are a number of apps which have been recoded for AltiVec, the vast majority won't be or can't be. In particular, I quite seriously doubt any major open source software projects exist which really take advantage of AltiVec and, unless gcc suddenly grows a (good) auto-vectorization engine really soon, none ever will. Both because most open source programs aren't really amenable to SIMD execution, and because the LinuxPPC community is small and developer resources are tight. At this point, anybody who buys a Macintosh to run Linux on it deserves to have "MORON" tatooed on their forehead.
Moreover, and maybe more importantly, while the number of apps which have some AltiVec routines in them is getting pretty large, it's important to remember that most of the features of these programs can't be rewritten in an SIMD paradigm, and much of what can be rewritten isn't such a great fit that tremendous speed gains are inevitable. Thus, Steve Jobs can always pick 3 or 4 Photoshop filters at which a G4 can run circles around a top-of-the-line P4, but it turns out that if you test the two in a benchmark suite containing *every* Photoshop filter, a 2.2 GHz P4 will win hands down. And then there's that pesky PC-133 and the 1999-style hard drive options shipping with standard Macs (ATA/66 anyone??); add in the time for the application to launch or to open a giant file (100+MB files are of course very common in Photoshop), and the PC blows the Mac away.
Of course the Mac has many strengths: an OS which is gorgeous on top and is UNIX underneath; a suite of strong, free (beer) media manipulation apps like iTunes, iDVD and iMovie; excellent upper-tier media apps like Photoshop and Final Cut Pro; FireWire ports standard; integration with the coolest MP3 player around. Once you take heat/power issues into consideration, the G3/G4 suddenly begin to hold their own with the x86 counterparts; the G4 PowerBook in particular is quite an attractive product. Yes, you can find an x86 laptop as thin and light and fast and with as much battery life for less (but without the big screen), or one with the same screen and a much faster CPU and video chip for less (but without the light weight and long battery life), but it's difficult to find another laptop that beats it all-around. (Although getting close every day; the other disadvantage of Apple products is that their specs stay put for 6 months or more at a time.)
But in terms of raw performance, the G4 is simply outclassed. Perhaps the mythical G5 will arrive soon to even things up (read: lessen the gap). But just because a "G5" based embedded chip is sampling from Moto doesn't mean the Mac is getting a G5 any time soon. Plus there's the rather strong possibility that Moto's semi division will be killed/spun off. (Of course the recent revival in the semi industry makes this less likely.)
Until then, Apple would do well to get DDR mobos here (yesterday, if possible), continue their good work converting anything they can to AltiVec, and...sad to say, but...continue spouting the lie that dual-CPU configurations and AltiVec can actually provide enough performance boost to elevate a new Mac to the performance level of a 6 month-old PC costing half the price (then).
One nitpick:
When you say,
(There are problems that Turing machines can't solve - the classic one is the halting problem. Can you write a program that can look at any other programs (not just most programs, ANY program in the infinite universe, including itself) and say if it'll lock up or not? With a Turing-equivalent machine, it's fundamentally impossible to do that - not just improbable or hard-to-implement, but impossible. The solution is to create a machine that's more capable than a Turing-equivalent machine, but that's rather tricky with macroscopic physics.)
it seems as if you're implying that a quantum computer would be more than Turing complete, and/or might be able to solve the Halting problem. This is wrong. It *may* be the case (AFAIK this hasn't been proven yet) that a quantum computer of arbitrary size could be used to solve all problems in NP in polynomial time/space in the length of the inputs (at least, if you don't believe the many-worlds interpretation of quantum theory) even if NP != P. In other words, a quantum computer of arbitrary size may be able to serve as the oracle that you need for the naive way to solve any problem in NP in polynomial time, even if a Turing machine could never solve NP-Complete problems in poly time in the length of the inputs.
This is *not* the same as being able to solve an incomputable problem--i.e. those which a Turing machine may take infinite time to solve (the original example being the Halting problem)--even though it would represent the first greater-than-polynomial improvement on the Turing machine yet. (Assuming that a quantum computer is, in fact, equivalent to a nondeterministic Turing machine.)
Sara-
If your company has "Product A", and another company attempts to create "Product B" to replace your product you're not going to be happy.
Yes, but trademark law is not about making the owner of the trademark happy. It's about making society as a whole happy.
Competition is the basis of any free market system and thus our laws are (theoretically) always set up to encourage more fair and transparent competition. Indeed, the Constitution specifically states that copyrights, trademarks, patents, etc. are provided only for their benefits to society, not (unlike, e.g. the rights to free speech or physical property) because they are inherently good in and of themselves.
To this end, the basic rule of thumb with trademarks is that Product B is violating the trademark on Product A if and only if potential customers are likely to be confused into thinking that Product B *is* Product A, or the company that makes Product A makes or has given their support to Product B. That's it. The company that makes Product A might wish for more, but their wishes would be at odds with the wishes of society as a whole.
And if you think about it, this is a very sensible rule, given that the intent of trademarks is to increase fair and transparent competition, not to offer barriers to competition with the trademark-holding company. Of course, if there were no trademarks at all, then "competition" would be increased, but it would not be transparent--the customer would not know whether they were getting the genuine article or a rip-off. Thus things like fake Gucci handbags are violations of trademark law (especially if they say "Gucci" on them or copy the Gucci logo or a logo which has clearly been designed to confuse customers into thinking it's the Gucci logo). But copying the overall design, color scheme, attitude, etc. of Gucci bags, or of any fashion designer, is legal and encouraged.
I think your confusion stems from the widespread use of the term "intellectual property" to describe copyright, trademarks, patents, etc. The term is semantically dishonest, because it makes it seem as if copyrights, trademarks, patents, etc. confer on their owners the same privileges that ownership of property does, when in fact nothing could be farther from the truth. The privileges conferred by copyrights, trademarks, patents, etc. are quite limited, and explicitly delineated (as opposed to the privileges conferred by owning physical property, which are expansive and only the *limitations* of which are explicitly delineated).
When you start thinking of copyrights, patents, trademarks, etc. as the (intellectual) "property" of the individuals or companies involved, you start thinking that they get to limit the way that their patented/trademarked/copyrighted product can be used to just those uses which they approve of. (Sort of like real property.) But, again, that's not what the law says, nor has it ever been the intent of the law; instead, the intent of the law is to prevent certain specific uses of the product or ideas associated with the product, such that the people who come up with those ideas are given a decent financial incentive to continue coming up with ideas, and such that the market remains more transparent to the public. (Patents do allow the patent-holder pretty wide lattitude to decide the uses to which their invention can be applied, but in exchange for that, patents are much more difficult to get than copyrights or trademarks, and last for a shorter time.)
Of course, if you start asking yourself "well, if that's the case, how on earth could laws like the Sonny Bono Copyright Extension Act"--which extends copyright to the life of the artist + 75 years, even though a much shorter term would never stop any reasonable person from creating and releasing a copyrightable work--"or the DMCA"--which, under the banner of copyright grants to copyright holders the ability to restrict the uses of their products to a degree much greater than the restrictions granted by copyrights, and even makes it illegal to attempt to use the product in ways which are legal but which the copyright holder does not like--"be constitutional?" The answer is that they're probably not, and simply haven't been around long enough to recieve a challenge at the Supreme Court level (IIRC there's a Sonny Bono case appearing before the Court now). What's scary is that the fact that these laws and other exist has confused many people into thinking that they must be constitutional, and thus into thinking that copyright (and, by extention, patents, trademarks, etc.) confers many more priviliges than it actually does. And, worst of all, thinking that it "ought to".