I knew somebody my freshman year of college that was on the invite list to the big Quake 3 tournaments, was rated somewhere in the top 100 in the country, and was overall amazing at the game.
Guess what he played the game on? Some crappy 8 or 16MB video card with all textures, details, resolution, and everything else all the way down. He had sound through what can only be described as a $2 pair of speakers, but they were enough for him to locate people (which was plain scary). The processor in the box wasn't anything spectacular either. He managed somewhere around 50-60 fps on that thing.
Quake 3 looks terrible at 640x480 with no detail, but it is perfectly playable. Heck, it's even fun multiplayer, because the gameplay is the same, it just isn't so pretty. But it doesn't have to be pretty to be fun. Pac-Man is fun and the graphics on that are terrible. I'm going to guess that Doom 3 will be perfectly playable on minimum specs as well, probably just not as pretty. Also, a $70 GeForce FX 5200 is a DX9 card last I checked. If you want the highest available resolution and textures and want the game for it's glitter then yes, you'll have to shell out cash for it.
That's the way its always been when you want the best right when it's released. I know people who bought a bunch of RAM to play Wolf 3D or Doom or whichever without windowing to a 3" box. You could, however play it in that 3" box. And those RAM upgrades were spendy. Sure, it helped the rest of the system, but when the box was games primarily and other things secondary it hardly matters. Cutting edge has always been expensive.
Fully agreed. I was more pointing out to the original poster that Pixar's task was about as good as it gets when it comes to parallelization, and that nobody needed to call Pixar up and tell them anything about their speedup.
In my explanation I failed to point out that the big point where HPC is going to pull ahead of cheap off the shelf clusters is in the message passing. To my credit I at least mentioned the insane interconnects like Myrinet and do know that fast data passing is a large part of the battle.
Pixar lucked out that the data passing can be hidden during the rendering of the first frame and the only transfer cost their cluster will ever see while not working on the problem is the transfer of the first frame.
And yeah, Cray is really going about their statements wrong. It isn't the linux portion at all, but really the underlying hardware.
Yup, I goofed on the point to pick apart there. The interconnects (which I also touched on) are probably what Cray was taking the linux machines to task on. Pixar's problem is ideal from the standpoint of not having much communication involved in the parallelization of their problem too. It's truly about as good as you can get.
If there's a lot of message passing that's likely to be the point where Crays shine, since the article mentioned something about obscene processor level interconnects.
I guess I got caught up in the point that the original poster was making about how Pixar magically got great speedup out of their clusters and pointed out more that it was the problem than focusing on how a problem might favor a Cray rather than a cluster. Thanks for pointing it out.
Well, it's been a little bit since my Parallel Computing class, but here goes for a problem which you would not want to compute in parallel.
The problem is simple, seed a random number generator with the processor id number and sum up the first N numbers generated from that seed, where N is the number of computers in the cluster. Why are we doing this? I have no idea, it's just my example problem.
If you want to do this in parallel you have to grab the seed from the machine which is initiating the problem, as the seed is tied to its processor. Then you have to scatter that seed. Then each node in the cluster has to generate M random numbers from that seed where M is the computer number of that node in the N node cluster. Then we can gather the values, adding the values that are gathered each step of the gathering for (lg N) time of actual addition. The scatter and gather also take about (lg N) time if you consider the transfer to be on the same order of time as an addition (if I remember right...). At this point we're looking good. Sadly, remember there are N operations for the last node in the cluster to get the Nth number from the random stream. O(3 lg N) + O(N) is on the order of O(N)
Now consider doing this on a single machine. N operations to get the full set of numbers to add plus N-1 additions is O(N). There is zero speedup by doing this in parallel. If you consider that data transfer is hardly on the same time scale as computation (maybe with uber-expensive interconnects, maybe) then you're actually wasting time by sending the problem through the cluster, resulting in a slowdown rather than a speedup. I suppose you can argue that it's parallelizable, but I'd call anybody who wanted to do this in parallel a fool. No matter how you split it, some machine will be forced to do N operations to get the last number out of the stream.
I'll grant you that the example is contrived. However it clearly shows that there exists a problem which cannot be sped up by being run in parallel. It is by no means a stretch to think that there are interesting problems that rarely see a speedup of over 8% when run on a cluster.
Pixar doesn't need telling, their problem breaks up so miraculously well that they'll see the best performance you could possibly expect from a cluster. The big problem, rendering a movie, decomposes into thousands of small problems, rendering a frame. Each machine in their cluster can handle a group of frames at a time with zero need to communicate or worse, share computation, with other machines in the cluster. It's the best case scenario.
Many other computing problems don't decompose nearly so nicely. So there are certainly problems that probably won't see more than 8% of peak performance. If you were particularly inclined you could probably invent a problem that had to be done serially, leaving percent of peak performance equal to what percent of your cluster one box was. Cray is right to that extent and if you're solving a problem that falls into the category of not easily parallelized then perhaps one of their machines is the better tool for the job. But, like you mention there are instances where the cluster is a great tool and cost effective to boot.
Heck, ever check out some of the faster interconnects like Myrinet? They're insane and exist because fast ethernet just doesn't cut it in some places. Just using a slow interconnect is enough to bring real performance down below theoretical peak. Luckily for Pixar off the shelf fast or gigabit ethernet is likely enough.
Anyway, use the best tool available. If your problem falls into the category of trivially parallelizable like rendering a movie is then don't bother wasting your money on a Cray. If your problem isn't suited to a cluster, however, then maybe a cluster isn't the right answer. If you have a big problem that needs serious computation take the time to figure out what you need before taking a marketing drone's spiel for gospel in your situation.
No, while Sim-city is a game it is an incredibly well done game. The reason that people feel this urge to make comments and references to Sim-city is that the game is well enough balanced that it feels like things should work the way they do in the game.
Yeah, things are a tad unrealistic in parts, like $10 for a segment of road or the $15,000 desalination plant. But the balance from income to expense and the balance of population needs vs. budget are well done. I have a hunch that Maxis keeps the costs way down so people don't get intimidated about multi-million dollar budgets every year and instead get excited when there are tens of thousands of dollars flowing in and out of their coffers in a given month. Just a guess though. I don't think anybody that plays the game realistically believes that a coal power plant costs $10,000 or that they can buy 1,000 MW from a nearby city for $20 a month. However, the concept that relying on your neighbors for power means they will raise prices is pretty universal.
Or, if that doesn't sate your rabid urge to scream "It's a game!" then I'll just bother to remind you that the comments referencing Sim-City are jokes. Jokes are all. No matter how well done the game is, it is only a simulation and nobody is seriously saying the world works just like in Sim-City.
In sim-city you have to provide basic necessities to your sim-residents, or they never move in, move in low density and low revenue buildings, or just move out if you stop supplying them. One of the things that you have to provide is water.
One of the nice features is that you can buy things like water and electricity from your neighboring cities for a price. This price tends to be higher per unit of supply than you could provide with a structure like a power plant or water pump, but requires far less up front cost. The not so nice thing is that your neighbors will occassionally renegotiate the price with you, meaning you'll pay more each month if you want to continue getting these supplies.
The joke in the previous post is based on the fact that you could import water (based on the bottled water comment) or that you could build a costly desalination plant (as the article suggests is happening). In sim-city you'll get shafted in time if you don't provide your own facilities, thus the neighbors raising the cost of bottled water is funny.
Now I feel like one of those people that analyzes a joke until it isn't funny. However, I went to the trouble of explaining for the poor non-sim-city player so I'm just going to post it... blah. The interesting thing is that bottled water seems to be pretty expensive anyway, and building one of these big plants is probably well worth the trouble in the long run.
Then why not get a big 10/100 switch that has 1 or 2 gigabit ports on it as well. Since the content is likely coming from a single box and being distributed one or two to each of another boxes this will provide the distribution box with enough bandwidth and not get in the way of the boxes that are receiving the data. If the content isn't coming from a single box then the point is kinda moot since it seems unlikely you'll have many computers each pushing out more than 4 HD streams.
That is, of course, unless you can think of some reason that somebody would be watching more than 4 streams of HD simultaneously on one machine. That seems a little odd though. And if for soem reason it does happen in 5 years there'll likely be even cheaper upgrades then.
This of course, mentions none of the difficulties you might have with the disks required to push this much content all over the place. You'll probably push those before pushing full gigabit capacity.
Good question. I've seen GBA SP's for $90 from Circuit City, but never lower than that. I think they're free shipping at the moment from their website too.
The first gen GBA's are $70 right now. Of course, I've seen them used for around $40 as well. It's the model I have and I'm not a huge fan of the lighting situation, but like the horizontal layout among other things.
It is too late. "You guys want to start a union? Fine. You're all fired, I'll go 100% to India."
Right after we're done training our replacements, right? It isn't that easy to just up and move all of the staff and it goes beyond the big guys that can.
Of course it would take major coordination on the part of every geek in the US, but it probably isn't too late at this point.
I've said it before in my replies to the people who replied and grumped about this before you did, and I guess I'm saying it again. I meant they need to get beyond what people think of them and not actually make non-kid games, since, as you've stated, they've already done that. I'm sorry I didn't word my post perfectly to avoid that bit of confusion, but this marks the third reply to my post that said this and I've already replied to the other two (quite promptly) that how it's being taken is not what I meant.
I mean, c'mon, I actually reply to people that reply to me for a reason. It is a discussion.
I've heard Kirby is an interesting sort of racing game which bends traditional rules. Maybe my sources are making stuff up. Haven't really checked review sites for it. Double Dash seems neat, but I'm getting really sick of cart games. They're all practically the same with one or two little twists and what I find to be poor control.
Anyway, you're right, Nintendo is second worldwide. I keep forgetting that the Japanese seem to hate the xbox with a passion. They're behind in the US though, which isn't a good position for them. And if I recall, the race is fairly close, especially considering the gap to the PS2 sales figure. Demand has certainly shot up with the reduction in price again, which is great. The new titles coming out for the cube have it looking good as well.
As for buying which console now I'm still going to speculate that having more cubes in geeks hands would be better. Apart from the people who pirate massively or turn the xbox into a media box of sorts, the geeks are going to want something out of their investment aside from a 10 GB drive that's a pain to unlock, a celeron, some ram, and an inperceptable dent in MS's budget. While some might sit it on a shelf, others would certainly give in to the temptation to play it.
I'm not against buying the xbox and playing it as a console. I think it's an interesting approach and the games are generally well rendered. But I think if you're going out to hurt Microsoft then helping their enemies will be the best route. Even without backwards compatability Microsoft can lure good developers with better sales figures. Developing for the US market? Then MS might be a better option than Nintendo. Buying up Xboxes only strengthens that claim.
Beyond that, I've never been convinced that backwards compatability sells consoles. Honestly I got a PS2 for PS2 games and the backwards compatability has been relatively unused. If I had a huge PS1 library I'd have a PS1. Now I doubt I'll get a PS3 since the PS4 will almost have to be backwards compatable, so Sony's pretty much lost a sale for being backwards compatable. If the Xbox 2 were backwards compatable I might hold off a purchase of one for the second console, again, losing a sale. But if not then no big loss because I can pick up an xbox on the cheap later if I really want it. I can then choose my next console for the next gen hardware. I never expected backwards compatability with my cartridge systems and it never mattered then. Why should it now?
It's definately getting blurred. I know more and more of the people I know are getting into animation. The computer rendered animation is certainly making inroads. Games are also getting more popular with all generations in the US.
Likewise there are plenty of kids animations in Japan, like Pokemon and the others aimed for the young ones. And games have always been popular in Japan as far as I know.
What really strikes me as strange is that geek culture is finally getting out of just geek circles. The rash of comic book movies, anime getting put on adult swim, and just general acceptance of what geeks do for hobbies is kinda amazing.
Sorry, I worded my comment wrong. What I meant was that Nintendo needs to get past all the people who think that it markets only to children, not that it only makes kids games. I know they have gone beyond that with games, but they're stuck with this stigma about them.
And I want to play Kirby's Air Ride actually. I'm a 23 year old male who loves the Kirby games. Nintendo makes some very fun games that are aimed at kids and some very fun games that aren't. Nintendo makes a lot of good games, period.
I sure as heck they don't pull a Sega. Nintendo may still be making a profit now, but what I'm hearing says they're skating pretty thin ice. I mean, their console is selling for the same price as their handheld unit. They've been selling their console for $100 to $50 less than the competition and last I heard they're still in third place. I'd like for them not to pull out of the console market and focus only on handheld games.
Buying Nintendo over Microsoft will ultimately hurt MS far more than it would to buy an Xbox at a loss.
Funny, but even without the consoles out there to buy games for, that's 10,000 more to the gap of sales vs. the Gamecube. Microsoft attracts more developers for the Xbox 2 with great numbers about market penetration. MS continues to dominate the market.
Besides, figuring a $150 loss per sale, that's only a cold $1.5 million on 10,000 units. Now, if those same units never sell there'll be double the pain, because they won't recoup $150 per box and the actual number of people playing legal games on the things hasn't changed.
To get to the $500 million the EU slaped them with you'd have to buy up over 3 million units assuming $150 loss per sale. Microsoft had sold fewer than 14 million units worldwide at the end of 2003. Do we really want to hand them 20% of their total sales to inflict losses similar to those done by the EU?
Okay, and you've found a valid use for them without buying games. Kudos. I don't have time to be messing with the things at the moment (but will later *grin*).
I'm still not sure that increasing their sales numbers is going to hurt them. They've already sunk costs into their inventory. MS has a huge stash of cash just lying around for further market domination. If they can post a big enough sales gap who's to say that Nintendo won't get scared and focus only on their gameboy product? The cube and gba already sell for the same amount! I don't think losing Nintendo would be good for the console wars and I like their games. It would be a win for MS though.
Heck, any favorable market penetration number they can spew about will attract developers. We've seen developers get sucked into xbox only or even away from the PC in the past. They're selling at a loss for a reason. They're not going to be upset if they sell all of their consoles.
Not sure how exactly this or your original reply to my post relates to my post. I'm advising purchase of the cube over the xbox, as at the very least it will help move Nintendo towards the number 2 spot.
The grandparent makes the point that buying them up would hurt MS. I agree that it would not, and for more reasons than just sunk inventory costs. Getting to point to larger sales numbers brings people to develop for them, in turn driving profits. It could scare Nintendo off because the sales gap would widen. There's a number of reasons that it could help MS without getting into money, which MS has plenty of even eating the whole console inventory it has, I'm sure.
These games are few and far between and typically have some feature of gameplay that is so compelling that people don't want to let go of their old copies. (Or for Rez, I'd imagine you'd not want the used controller bit anyway...) Anyway, if there's some game that is absolutely being raved about, well, go get it. I don't know exactly how long the production runs on some of these games are, but I'd hope they're not so short that it's completely gone within a couple months.
Say you decide you want a new system, to play KOTOR or Gran Turismo or some console specific game. You go hrm, I need $150 to get it now and I have $50 a month budgeted towards games. Month 1 goes by, you keep $30 in your pocket and finally beat FFX or Halo or whatever. Plus you get a greatest hits game that you've been eyeing for a while but aren't certain of. Month 2 rolls by and you do likewise with another game and buy another greatest hit. Month 3 comes around and there's Chrono Trigger Puzzle Fighter X Ultra or some such that's destined to be great and go for $70 on ebay. Well, dump this month's game money on it and wait another month for the new console. Or if you can't stand the thought of waiting 5 whole months buying only greatest hits games you've underlooked because you have to have every game first, then take $25 each month and buy a new release every other month. It'll take 6 months of that. Simple budgeting.
As an alternate suggestion you could save everything 3 months straight and then buy a couple greatest hits games for the new console the next month, which you know you've been wanting. Then buy three in the second month. Then go back to new releases. If something big comes up after the 3 months then you can get it instead of greatest hits games, you've still got $50 a month. If it happens in the first 3 months just delay the console purchase. This isn't hard.
And if it costs $70 on ebay what's the huge deal? They cost $50 in the stores and you'll never pay less than that. Sure, you spend $20 more on a single title while buying however many good games for the new system at $20 each and still manage to pay off the hardware system with 5 games that are guaranteed to become greatests hits games. It all gets evened out with time. What's the big deal if you don't personally rip the shrink wrap off?
I guess the chunk of my comment was more poorly worded than anything and I really meant that they need to get over the typical mentality of the masses that they make games only for kids. Sorry for the confusion there.
Anyway, not one of those people here. I'm probably gonna pick up a cube soon and get some of the "kiddie" games like Super Monkey Ball and Kirby Air Grind and whatever else looks like fun to play. I'm a big fan of some of the fun and interesting games that Nintendo's made or had on the cube that aren't just rehashes of the same old thing.
Like I said, I recommend getting a Cube rather than an Xbox. You can get games for it and have fun without supporting Microsoft.
I don't know. It seems that the game is played by selling the hardware at a loss and making up for it with good titles and collecting money from 3rd party developers who make money because your console is widespread.
Given that, while the hardware may be taking losses, I don't know how hard the gaming unit at MS or Sony gets hit as a whole. And cripes do they make up for it with accessories. One controller and extras cost $25+. Memory cards going at $20 for 8 MB? The initial console sale is a loss for sure, but there's plenty of room for profit.
So what's the problem with console makers dropping their price horrendously with time? If you look at initial pricing, you can now get an xbox and gamecube for less than the initial price of the xbox or the PS2. If Sony dumps the price again you can get the xbox and the PS2 for the price of either initially.
I don't think this is heading towards a monopoly, I think it's heading towards trying to grab more customers that already have a system and aren't likely to give another a shot without incentives. I have a PS2 which I got at the $200 price point. I'm now considering an Xbox at this new point. And why not, Ninja Gaiden is sweet and there's some other games I'd like to check out as well. Will I stop buying PS2 games? No way. Well, unless they're the all platform games, in that case I'll pick whichever looks the nicest, likely the xbox's version.
Heck, new hardware at $150 means skipping 5 new releases at $50 each and waiting until they're $20 each. I don't have to skip the games, just wait a little bit. And if I have a good selection of "Greatest Hits" or whatever for the console I just got they're as good as new games to me, right?
Microsoft is sitting on billions of dollars. Everybody rushing out and buying an xbox will only serve to skew the results of who's leading in the console wars and make Nintendo look even worse. Besides, who's going to sink $150 into the thing and not get a game or two for it?
I don't want to see Nintendo pull a Sega because of some silly numbers being inflated. Nintendo needs to get over the gaming is for kids only thing, but they've made and continue to make some quality games.
Want to hurt Microsoft's interests here? Everybody rushing out and buying a cube (at $50 less than the xbox no less) and putting the big N into the number 2 spot again will have more of an effect.
Why premature? Why not ask these questions now and get them over with?
Personally, I think it'd be kinda nice to start now as best possible and try to get Mars warmed up for human inhabitants. Think about it. If we could get even something that would thicken up the atmosphere then water might once again run across the surface of Mars. With water there comes the potential to get plants and such up there and turn some CO2, which we know is in one of the ice caps and could sublime into the atmosphere, and convert that into oxygen.
Wouldn't that be nice? An atmosphere that, while not entirely the same as our own might be thick enough to not need pressure suits. An atmosphere that might only require an assisted supply of O2. We don't need to be there personally to try and get started on it. In fact, if we can change the atmosphere significantly it would be best to let the process get to a point that's at least semi-hospitable before trying to engineer housing and the like to withstand conditions so much worse.
Writing in a secure language is all fine and dandy until somebody somewhere finds a way to exploit something that isn't due to language constraints or finds an error in the language which causes an error in the software that's supposed to be secure.
Don't get me wrong, it'd be a great first step and would raise the bar for many attacks, but I think that the way errors are written into anything and the additional complexity it can add, that it's not the end all be all for solutions.
Simple bits of extra hardware can prevent some of this stuff. I don't think that stack inspection classifies as simple, but running a bit of hardware that's purpose is to filter packets removes your machine from being harmed if the firewall is. Sure, there's ugly things like loss of connectivity if somebody successfully attacks your hardware firewall, but it's a little different than arbitrary code execution leading up to destruction of your drives.
I'm a fan of what nVidia's going to do on their next nForce3 board, an onchip (on the nic, or bridge that implements the nic) firewall. I think 3com has done the same sometime in the past. It'll essentially bring something like BlackIce or ZoneAlarm out to a hardware level but still personal to the machine. I'd wager the worst attack on it you could get would to be to lock your system (unlikely) or drop your connectivity (more likely), not arbitrary code execution.
I think the best solution is to get as many of the potential avenues of attack that would get at the OS away from the OS. BlackIce and other security measures are just too close to the processor for me to want to actively use them. Might not be a choice at the moment on a laptop, but for a desktop I'll take more hardware and fewer potentially vulnerable processes anyday.
Slashdot Mirror
Guess what he played the game on? Some crappy 8 or 16MB video card with all textures, details, resolution, and everything else all the way down. He had sound through what can only be described as a $2 pair of speakers, but they were enough for him to locate people (which was plain scary). The processor in the box wasn't anything spectacular either. He managed somewhere around 50-60 fps on that thing.
Quake 3 looks terrible at 640x480 with no detail, but it is perfectly playable. Heck, it's even fun multiplayer, because the gameplay is the same, it just isn't so pretty. But it doesn't have to be pretty to be fun. Pac-Man is fun and the graphics on that are terrible. I'm going to guess that Doom 3 will be perfectly playable on minimum specs as well, probably just not as pretty. Also, a $70 GeForce FX 5200 is a DX9 card last I checked. If you want the highest available resolution and textures and want the game for it's glitter then yes, you'll have to shell out cash for it.
That's the way its always been when you want the best right when it's released. I know people who bought a bunch of RAM to play Wolf 3D or Doom or whichever without windowing to a 3" box. You could, however play it in that 3" box. And those RAM upgrades were spendy. Sure, it helped the rest of the system, but when the box was games primarily and other things secondary it hardly matters. Cutting edge has always been expensive.
In my explanation I failed to point out that the big point where HPC is going to pull ahead of cheap off the shelf clusters is in the message passing. To my credit I at least mentioned the insane interconnects like Myrinet and do know that fast data passing is a large part of the battle.
Pixar lucked out that the data passing can be hidden during the rendering of the first frame and the only transfer cost their cluster will ever see while not working on the problem is the transfer of the first frame.
And yeah, Cray is really going about their statements wrong. It isn't the linux portion at all, but really the underlying hardware.
If there's a lot of message passing that's likely to be the point where Crays shine, since the article mentioned something about obscene processor level interconnects.
I guess I got caught up in the point that the original poster was making about how Pixar magically got great speedup out of their clusters and pointed out more that it was the problem than focusing on how a problem might favor a Cray rather than a cluster. Thanks for pointing it out.
The problem is simple, seed a random number generator with the processor id number and sum up the first N numbers generated from that seed, where N is the number of computers in the cluster. Why are we doing this? I have no idea, it's just my example problem.
If you want to do this in parallel you have to grab the seed from the machine which is initiating the problem, as the seed is tied to its processor. Then you have to scatter that seed. Then each node in the cluster has to generate M random numbers from that seed where M is the computer number of that node in the N node cluster. Then we can gather the values, adding the values that are gathered each step of the gathering for (lg N) time of actual addition. The scatter and gather also take about (lg N) time if you consider the transfer to be on the same order of time as an addition (if I remember right...). At this point we're looking good. Sadly, remember there are N operations for the last node in the cluster to get the Nth number from the random stream. O(3 lg N) + O(N) is on the order of O(N)
Now consider doing this on a single machine. N operations to get the full set of numbers to add plus N-1 additions is O(N). There is zero speedup by doing this in parallel. If you consider that data transfer is hardly on the same time scale as computation (maybe with uber-expensive interconnects, maybe) then you're actually wasting time by sending the problem through the cluster, resulting in a slowdown rather than a speedup. I suppose you can argue that it's parallelizable, but I'd call anybody who wanted to do this in parallel a fool. No matter how you split it, some machine will be forced to do N operations to get the last number out of the stream.
I'll grant you that the example is contrived. However it clearly shows that there exists a problem which cannot be sped up by being run in parallel. It is by no means a stretch to think that there are interesting problems that rarely see a speedup of over 8% when run on a cluster.
Many other computing problems don't decompose nearly so nicely. So there are certainly problems that probably won't see more than 8% of peak performance. If you were particularly inclined you could probably invent a problem that had to be done serially, leaving percent of peak performance equal to what percent of your cluster one box was. Cray is right to that extent and if you're solving a problem that falls into the category of not easily parallelized then perhaps one of their machines is the better tool for the job. But, like you mention there are instances where the cluster is a great tool and cost effective to boot.
Heck, ever check out some of the faster interconnects like Myrinet? They're insane and exist because fast ethernet just doesn't cut it in some places. Just using a slow interconnect is enough to bring real performance down below theoretical peak. Luckily for Pixar off the shelf fast or gigabit ethernet is likely enough.
Anyway, use the best tool available. If your problem falls into the category of trivially parallelizable like rendering a movie is then don't bother wasting your money on a Cray. If your problem isn't suited to a cluster, however, then maybe a cluster isn't the right answer. If you have a big problem that needs serious computation take the time to figure out what you need before taking a marketing drone's spiel for gospel in your situation.
Yeah, things are a tad unrealistic in parts, like $10 for a segment of road or the $15,000 desalination plant. But the balance from income to expense and the balance of population needs vs. budget are well done. I have a hunch that Maxis keeps the costs way down so people don't get intimidated about multi-million dollar budgets every year and instead get excited when there are tens of thousands of dollars flowing in and out of their coffers in a given month. Just a guess though. I don't think anybody that plays the game realistically believes that a coal power plant costs $10,000 or that they can buy 1,000 MW from a nearby city for $20 a month. However, the concept that relying on your neighbors for power means they will raise prices is pretty universal.
Or, if that doesn't sate your rabid urge to scream "It's a game!" then I'll just bother to remind you that the comments referencing Sim-City are jokes. Jokes are all. No matter how well done the game is, it is only a simulation and nobody is seriously saying the world works just like in Sim-City.
One of the nice features is that you can buy things like water and electricity from your neighboring cities for a price. This price tends to be higher per unit of supply than you could provide with a structure like a power plant or water pump, but requires far less up front cost. The not so nice thing is that your neighbors will occassionally renegotiate the price with you, meaning you'll pay more each month if you want to continue getting these supplies.
The joke in the previous post is based on the fact that you could import water (based on the bottled water comment) or that you could build a costly desalination plant (as the article suggests is happening). In sim-city you'll get shafted in time if you don't provide your own facilities, thus the neighbors raising the cost of bottled water is funny.
Now I feel like one of those people that analyzes a joke until it isn't funny. However, I went to the trouble of explaining for the poor non-sim-city player so I'm just going to post it... blah. The interesting thing is that bottled water seems to be pretty expensive anyway, and building one of these big plants is probably well worth the trouble in the long run.
That is, of course, unless you can think of some reason that somebody would be watching more than 4 streams of HD simultaneously on one machine. That seems a little odd though. And if for soem reason it does happen in 5 years there'll likely be even cheaper upgrades then.
This of course, mentions none of the difficulties you might have with the disks required to push this much content all over the place. You'll probably push those before pushing full gigabit capacity.
Good question. I've seen GBA SP's for $90 from Circuit City, but never lower than that. I think they're free shipping at the moment from their website too.
The first gen GBA's are $70 right now. Of course, I've seen them used for around $40 as well. It's the model I have and I'm not a huge fan of the lighting situation, but like the horizontal layout among other things.
Right after we're done training our replacements, right? It isn't that easy to just up and move all of the staff and it goes beyond the big guys that can.
Of course it would take major coordination on the part of every geek in the US, but it probably isn't too late at this point.
Okay, you've got me there. The army of giant Xbox Tuxes is nothing short of a sweet idea. ;)
I mean, c'mon, I actually reply to people that reply to me for a reason. It is a discussion.
Anyway, you're right, Nintendo is second worldwide. I keep forgetting that the Japanese seem to hate the xbox with a passion. They're behind in the US though, which isn't a good position for them. And if I recall, the race is fairly close, especially considering the gap to the PS2 sales figure. Demand has certainly shot up with the reduction in price again, which is great. The new titles coming out for the cube have it looking good as well.
As for buying which console now I'm still going to speculate that having more cubes in geeks hands would be better. Apart from the people who pirate massively or turn the xbox into a media box of sorts, the geeks are going to want something out of their investment aside from a 10 GB drive that's a pain to unlock, a celeron, some ram, and an inperceptable dent in MS's budget. While some might sit it on a shelf, others would certainly give in to the temptation to play it.
I'm not against buying the xbox and playing it as a console. I think it's an interesting approach and the games are generally well rendered. But I think if you're going out to hurt Microsoft then helping their enemies will be the best route. Even without backwards compatability Microsoft can lure good developers with better sales figures. Developing for the US market? Then MS might be a better option than Nintendo. Buying up Xboxes only strengthens that claim.
Beyond that, I've never been convinced that backwards compatability sells consoles. Honestly I got a PS2 for PS2 games and the backwards compatability has been relatively unused. If I had a huge PS1 library I'd have a PS1. Now I doubt I'll get a PS3 since the PS4 will almost have to be backwards compatable, so Sony's pretty much lost a sale for being backwards compatable. If the Xbox 2 were backwards compatable I might hold off a purchase of one for the second console, again, losing a sale. But if not then no big loss because I can pick up an xbox on the cheap later if I really want it. I can then choose my next console for the next gen hardware. I never expected backwards compatability with my cartridge systems and it never mattered then. Why should it now?
Likewise there are plenty of kids animations in Japan, like Pokemon and the others aimed for the young ones. And games have always been popular in Japan as far as I know.
What really strikes me as strange is that geek culture is finally getting out of just geek circles. The rash of comic book movies, anime getting put on adult swim, and just general acceptance of what geeks do for hobbies is kinda amazing.
And I want to play Kirby's Air Ride actually. I'm a 23 year old male who loves the Kirby games. Nintendo makes some very fun games that are aimed at kids and some very fun games that aren't. Nintendo makes a lot of good games, period.
I sure as heck they don't pull a Sega. Nintendo may still be making a profit now, but what I'm hearing says they're skating pretty thin ice. I mean, their console is selling for the same price as their handheld unit. They've been selling their console for $100 to $50 less than the competition and last I heard they're still in third place. I'd like for them not to pull out of the console market and focus only on handheld games.
Buying Nintendo over Microsoft will ultimately hurt MS far more than it would to buy an Xbox at a loss.
Not a problem. It's a good point I neglected to make in my original post anyway.
Besides, figuring a $150 loss per sale, that's only a cold $1.5 million on 10,000 units. Now, if those same units never sell there'll be double the pain, because they won't recoup $150 per box and the actual number of people playing legal games on the things hasn't changed.
To get to the $500 million the EU slaped them with you'd have to buy up over 3 million units assuming $150 loss per sale. Microsoft had sold fewer than 14 million units worldwide at the end of 2003. Do we really want to hand them 20% of their total sales to inflict losses similar to those done by the EU?
I'm still not sure that increasing their sales numbers is going to hurt them. They've already sunk costs into their inventory. MS has a huge stash of cash just lying around for further market domination. If they can post a big enough sales gap who's to say that Nintendo won't get scared and focus only on their gameboy product? The cube and gba already sell for the same amount! I don't think losing Nintendo would be good for the console wars and I like their games. It would be a win for MS though.
Heck, any favorable market penetration number they can spew about will attract developers. We've seen developers get sucked into xbox only or even away from the PC in the past. They're selling at a loss for a reason. They're not going to be upset if they sell all of their consoles.
The grandparent makes the point that buying them up would hurt MS. I agree that it would not, and for more reasons than just sunk inventory costs. Getting to point to larger sales numbers brings people to develop for them, in turn driving profits. It could scare Nintendo off because the sales gap would widen. There's a number of reasons that it could help MS without getting into money, which MS has plenty of even eating the whole console inventory it has, I'm sure.
Say you decide you want a new system, to play KOTOR or Gran Turismo or some console specific game. You go hrm, I need $150 to get it now and I have $50 a month budgeted towards games. Month 1 goes by, you keep $30 in your pocket and finally beat FFX or Halo or whatever. Plus you get a greatest hits game that you've been eyeing for a while but aren't certain of. Month 2 rolls by and you do likewise with another game and buy another greatest hit. Month 3 comes around and there's Chrono Trigger Puzzle Fighter X Ultra or some such that's destined to be great and go for $70 on ebay. Well, dump this month's game money on it and wait another month for the new console. Or if you can't stand the thought of waiting 5 whole months buying only greatest hits games you've underlooked because you have to have every game first, then take $25 each month and buy a new release every other month. It'll take 6 months of that. Simple budgeting.
As an alternate suggestion you could save everything 3 months straight and then buy a couple greatest hits games for the new console the next month, which you know you've been wanting. Then buy three in the second month. Then go back to new releases. If something big comes up after the 3 months then you can get it instead of greatest hits games, you've still got $50 a month. If it happens in the first 3 months just delay the console purchase. This isn't hard.
And if it costs $70 on ebay what's the huge deal? They cost $50 in the stores and you'll never pay less than that. Sure, you spend $20 more on a single title while buying however many good games for the new system at $20 each and still manage to pay off the hardware system with 5 games that are guaranteed to become greatests hits games. It all gets evened out with time. What's the big deal if you don't personally rip the shrink wrap off?
Anyway, not one of those people here. I'm probably gonna pick up a cube soon and get some of the "kiddie" games like Super Monkey Ball and Kirby Air Grind and whatever else looks like fun to play. I'm a big fan of some of the fun and interesting games that Nintendo's made or had on the cube that aren't just rehashes of the same old thing.
Like I said, I recommend getting a Cube rather than an Xbox. You can get games for it and have fun without supporting Microsoft.
Given that, while the hardware may be taking losses, I don't know how hard the gaming unit at MS or Sony gets hit as a whole. And cripes do they make up for it with accessories. One controller and extras cost $25+. Memory cards going at $20 for 8 MB? The initial console sale is a loss for sure, but there's plenty of room for profit.
So what's the problem with console makers dropping their price horrendously with time? If you look at initial pricing, you can now get an xbox and gamecube for less than the initial price of the xbox or the PS2. If Sony dumps the price again you can get the xbox and the PS2 for the price of either initially.
I don't think this is heading towards a monopoly, I think it's heading towards trying to grab more customers that already have a system and aren't likely to give another a shot without incentives. I have a PS2 which I got at the $200 price point. I'm now considering an Xbox at this new point. And why not, Ninja Gaiden is sweet and there's some other games I'd like to check out as well. Will I stop buying PS2 games? No way. Well, unless they're the all platform games, in that case I'll pick whichever looks the nicest, likely the xbox's version.
Heck, new hardware at $150 means skipping 5 new releases at $50 each and waiting until they're $20 each. I don't have to skip the games, just wait a little bit. And if I have a good selection of "Greatest Hits" or whatever for the console I just got they're as good as new games to me, right?
I don't want to see Nintendo pull a Sega because of some silly numbers being inflated. Nintendo needs to get over the gaming is for kids only thing, but they've made and continue to make some quality games.
Want to hurt Microsoft's interests here? Everybody rushing out and buying a cube (at $50 less than the xbox no less) and putting the big N into the number 2 spot again will have more of an effect.
Personally, I think it'd be kinda nice to start now as best possible and try to get Mars warmed up for human inhabitants. Think about it. If we could get even something that would thicken up the atmosphere then water might once again run across the surface of Mars. With water there comes the potential to get plants and such up there and turn some CO2, which we know is in one of the ice caps and could sublime into the atmosphere, and convert that into oxygen.
Wouldn't that be nice? An atmosphere that, while not entirely the same as our own might be thick enough to not need pressure suits. An atmosphere that might only require an assisted supply of O2. We don't need to be there personally to try and get started on it. In fact, if we can change the atmosphere significantly it would be best to let the process get to a point that's at least semi-hospitable before trying to engineer housing and the like to withstand conditions so much worse.
Don't get me wrong, it'd be a great first step and would raise the bar for many attacks, but I think that the way errors are written into anything and the additional complexity it can add, that it's not the end all be all for solutions.
Simple bits of extra hardware can prevent some of this stuff. I don't think that stack inspection classifies as simple, but running a bit of hardware that's purpose is to filter packets removes your machine from being harmed if the firewall is. Sure, there's ugly things like loss of connectivity if somebody successfully attacks your hardware firewall, but it's a little different than arbitrary code execution leading up to destruction of your drives.
I'm a fan of what nVidia's going to do on their next nForce3 board, an onchip (on the nic, or bridge that implements the nic) firewall. I think 3com has done the same sometime in the past. It'll essentially bring something like BlackIce or ZoneAlarm out to a hardware level but still personal to the machine. I'd wager the worst attack on it you could get would to be to lock your system (unlikely) or drop your connectivity (more likely), not arbitrary code execution.
I think the best solution is to get as many of the potential avenues of attack that would get at the OS away from the OS. BlackIce and other security measures are just too close to the processor for me to want to actively use them. Might not be a choice at the moment on a laptop, but for a desktop I'll take more hardware and fewer potentially vulnerable processes anyday.