State universities are substantially cheaper of course. However the only other suitable universities near me are Princeton, which is around $50k/year, and Columbia which is somehwat less, I'm not sure exactly how much so. Both are obviously private.
There was no state school within commuting distance to where i lived that had an engineering program worth considering, so I simply didn't consider that.
Compared to the schools I did consider, mine was a bargain!
Yes but I think you may be in the minority. If you WANT to do such a thing, nothing forces you to avoid those classes. I take dance classes with my wife, for our enjoyment. I have not found a use for the waltz in the office. However the rest of us felt forced not only into lost time, but money as well. I just wanted a Bachelors to get in the field, and a Masters to earn more money in it. I enjoyed my masters far more because it was concentrated on my subject, short (I was able to take summer courses) and I felt accomplished.
My school was relatively inexpensive and I lived at home. Tuition was $22k/yr (on average), so $88k for a BS. After scholarships etc. we ended up having to pay about 30% of that (which we paid rather than FAFSA'd), however I believe fully half went to "junk" liberal arts I didn't want. It's simply a waste.
I disagree that there is any technical/career advantage in liberal arts courses. Critical thinking isn't taught in ANY course, or derived from it. You can't learn it, you develop it like the ability to speak, or NOT if you are not suitably challenged or curious. I know many engineers who simply lack the ability to solve problems, some with degrees from MIT and Princeton.
What I think many may be saying, unclearly or untruthfully is that you do not want a program that is too focused on specifics. You don't want to teach a kid GCC, Make, and the C language and throw him out the door as a "Computer Scientist". He will be crippled. I think Math, Physics, Digital Logic, Computer Archtiecture and then the core CS courses, in addition to a few practical courses that are not often offered that amount to "training".
There were many good experiences I had in school but the frustration I and my friends had with how much money we were spending and how much time we were spending on stuff that we didn't "sign up for" was at times intense. I guess I wouldn't want my own children to have to endure that (nor do I necessarily want to pay for it).
I was speaking in terms of the design of memory chips themselves, each of them supporting a 10G link for example could today, support compatible bandwidth from the current parallel bus approach and do bank switching in a number of ways.
You could still build a "DIMM-like" module to plug in on board, so your module density would be comparable if you wanted, or throw down your "scrabble" chips singly. Either one works, and has an exact parallel to the more far fetched Sun approach. Let's face it, the real bottleneck in ANY system is memory, they're slow. I can see us building a scrabble board (out of FR4 instead of silicon), where each tile is a big serial link that speaks some common language. You could lay down "words" of memory that take 4 tiles, or one at a time. Each specific technology probably has it's own wants.
128-bit wide and 256-bit wide busses on video cards do exist and are always to memories, and memories these days are 16-bit wide each even for Alpha's, and ganged together in parallel. So the bandwidth to each device is quite replaceable by a serial link.
I did not see a paper attached in the article, and the article itself seemed the usual level of gee-whiz that I'd expect. I saw what seemed to be fact (packaging bottleneck for ASICs) with some things that seemed to be under investigation "there is a noticeable change in voltage levels at the receiver" to wild speculation about scrabble tiles and chips. I can't quite make the leap from what they COULD be talking about that would allow for a multitde of "wireless" links, to "dropping chips in like scrabble tiles". As far as I know if you have two circuits communicating by some electric field between them, you either a) have to use enough power that one can talk to the other through a large area or b) have to place them very close together. When someone says they have two "wireless" circuits communicating by mutual capacitance, I think of two networks placed so closely together that the air (or silicon) between them forms an adequate dielectric that they act like a capacitor has been placed in series between them. To do this in silicon you'd really need to place the CPU "pad" very close to the cache "pad", far more precicely than could be done in the field. The benefits I see are tangible, you could build flavors of a chip with various cache sizes without a complete relayout and requalification of the ASIC, but not revolutionary to us end users. In fact technology to do this also exists today but remains unused (www.sychip.com used to do this, and may still in some form). I'm interested in reading a paper written by an engineer if you know where it is, but the article sounds just way too gee-whiz for me to take seriously and the/. title makes it worse =/
I am aware of a few DARPA projects where they are building these "chip sandwhiches" but they're all made in a semiconductor fab, and the yields aren't too great.
So maybe this technology they're coming up with isn't too bad from the standpoint of helping build a more flexible processor design, I don't see that you'll have a scrabble computer though, or that us end-users will see much benefit. We'll still have to buy a CPU with some amount of cache built on, and still have to have PCBs to connect the CPU to everything else. And they're still going to be beasts to design.
Ok, first I'm insinuating what would be more practical from a sytem design perspective, not proposing a solution on existing technology. It happens existing technology is damn close though.
Second, wireless communication by means of mutual capacitance is reasonably serial in itself, by clever design you may be able to simultaneously communicate with some small number of chips (on the order of 10s), but it's still reasonably serial. I think it would however be harder to get high bandwidth wireless serial links to be as fast and reliable as wired links.
Third I'm not sure about your numbers, but I'm in telecomm not computers. An 800MHz FSB is actually a 200Mhz bus sampled 4x during a single clock period. The bus, I think, is 64-bits wide and is bi-directional. At best, ignoring turn around times etc., I see a bandwidth of 51.2Gbits/sec, which corresponds to the number Intel claims. This can be realized today with about 5 serial links. I'm thinking of a driver on Xilinx FPGAs that I just happen to be using today, I'm sure there are others maybe better, but it supports 10.135Gbit/s as is imminently designable on a PCB even with very long trace lengths.
That 800MHz FSB in fact connects to the north bridge, which aggregates traffic from the DDR Bus, AGP bus and PCI busses that is bound to the CPU. Instead, why not give one serial link to each memory (usually at least 4, often more), one to AGP and another to PCI? You're still perfectly modular and have a far simpler PCB that you can still "mix and match" assuming everyome supports the same protocols.
Fundamentally I think Sun has a good idea but the wrong implementation. It would seem what is needed is a standardized, unburdened serial link standard, and a standardized protocol for data transfer across it for compute oriented applications. Sun however probably wishes to own the IP and make money from it, so this isn't in their interests.
I think this is true, I pirated stuff non-stop as a minor, I had lots of time and the inability to generate money. Piracy takes lots of time: download/driving times, getting bad files and finding new sources, building up karma required for downloads on private sites etc. Pain in the ass.
Now I have money but no time so I never pirate games. I do tend to buy the games I enjoyed way back when where available. Doom somehow I never liked that much, but I look forward to Space Quest 93.
Ground planes alone would just help capture all that leaked RF energy and unhelpfully put it on your power supply return (which also can pick up outside interference and help screw with your chip sandwhich). Stuff can be done about this but you'd still have to have an off chip board with the right set of filters etc.
Instead of screwball stuff I think it'd be more helpful to simply find ways to drastically reduce the number of pins required. Most of these chips are huge because of the 128 signal wide memory/data busses, N control/configuration pins, big address busses etc. Much of this can be replaced with comparatively fewer high speed links.
What I'd rather see is parallel busses being replaced by very high speed serial links (all patented to hell and back of course), perhaps one link per expected peripheral (memory, adjunct processor, i/o bus, video bus, etc.). One could build a very cheap PCB that could almost be hand assembled. The problem is that each peripheral would also have to be compatible with the link. No one builds DDRs with SERDES links for example...
80% of the boards I've designed have been pratically identical in terms of core functionality, but they've been completely different in implementation because of the differing interfaces.
I can't bring myself to actually buy a video game these days since the last 6-7 offline games I've played have sucked horribly. Either yet another FPS copy, a bad perversion on a previous game that was good (i.e. MOO3), or a game that was not at all what it pretended to be on the box.
The best I can come to is trying them out first and buying if it's worth teh money. But most games aren't even worth the download time lately...
All I know is my own situation, but I suspect it's common:
-> I get a great idea for something the world needs
-> I need to prototype it, but have grossly insufficient funds
-> I seek people with money who are willing to loan me some $$$ if I pay them back more than they gave me
-> Now I must build the device on the capital given to me, and figure out how to make more $$$ from it than I put in to it, to pay off my loan AND continue to eat for the next 2 years.
-> I have 2 choices, prototype and sell my invention itself (actually the "IP" behind it), or go in to business as an equipment maker. If I do the former, I can sell the invention and pay off my debts, some other company then makes the $$$ from it. If I do the latter, then I will require even MORE investment, and require even MORE profits to stay afloat.
Enter profit, aka "no good deed goes unpunished".
There is a big gap in the universe between the people with good ideas, and the people with money. Maybe the right thing to do is prohibit companies from owning patents.
But the original poster was correct. SWG *WAS* very much dissapointing and this expansion *IS* only adding content that was originally slated for the initial launch. I was not even on beta but I received regular emails about features for SWG that seemed to DIMINISH every month until launch.
More to the point, I played SWG, it was boring, there was no content and the thrust of the game seemed to be a levelling treadmill that was even MORE dull than EverQuest (if that's possible).
Somehow I read your post and believe the opposite. Games are ultimately designed for the player to win: be it by solving problem, developing spiderman twitch reflexes, etc. Winning or beating a level usually involves an ego massage along with it ("Excellent mighty warrior...") Second, no matter how realisitic a game ultimately looks, suspension of disbelief is voluntary subconcious submission to unreality, and only disturbed people actually think it's real. The rest of us know in our heart that our victory is hollow, but enjoy it for the challenge of it. Do you worry much about dying in a game? Not really, only if it sets you back somehow.
What Semper Fi above was alluding to is that "real life" is quite the opposite. You may not win, your entire purpose in life may be to be blown up. You may be the smartest, fastest and well equipped and STILL eat it. When you kill someone, he stays dead forever. When his brains are splattered on the wall, you are faced with the harsh reality that IT COULD HAVE BEEN YOU, and may yet be. Further your survival may have less to do with native instict & training than by sheer dumb luck.
The trouble is, until you're in one of these situations, I don't think anyone (myself included) really understands it. Games romanticize it, make it sexy etc. I doubt they are a deterrent to anything.
On the other hand I don't really believe many people join based on notions of heroics. I suspect the #1 driver is paying for college or more simply steady work & a familiar institutional atmosphere. During war-time it may be a different story, but then the country may need foolish people in search of glory to fill the ranks, and they will become heroes even if they die unromantically.
I have a MS in EE. I was offered a scholarship to do a PhD. However during my MS thesis work I worked closely with PhD candidates and suffered perhaps a fraction of what they did. Don't get me wrong, I enjoy what I do, but I eat corporate shit for $$$ alone, so I declined. My observations are as follows:
1) PhD is a lot of work for yourself, and 1000x more work doing your professors busy work (papers etc.)
2) PhD slave labor wages are less than those of any given malaysian factory child if you count the total number of hours worked and divide that into your scholarship/stipend/grant/etc.
3) If you are not a US citizen/permanent resident and are on a scholarship to get a PhD in the US, you are fucked. Bring the vasoline and bend over.
4) If your goal is simply to get a degree to get more money, stop at your masters.
5) If your PhD is not in a subject actively investigated by the corporate world be willing to accept an academic position after getting your degree, or find another subject. It's heartbreaking to see people get their degree and realize they are either stuck in academia or worse, take a job in industry doing work outside their expertise making the same they would have as a masters (i.e. degree worthless).
6) If at all possible GET A COMPANY TO FUND YOUR PHD! This is harder now than it used to be, but it is THE way to go. I can't recommend it enough, if I personally thought there was money in a PhD this is what I'd do myself. If your professor administrates whatever finances your degree, and you are above broccoli intelligence, he WILL try to hold you as long as he can (5-7 years in most schools). If your company is paying the bill they are quite good at getting you in and out ASAP. Avg stay of corporate funded PhD students in my experience was 3 years. Do this!
7) Stupid people can get PhD's far easier than smart people. Simply put, professors want stupid people out of their hair, if they can't wash em out, they graduate em. Just like elementary school.
Obviously if gamers see the danger, and in masses go towards it, then clearly that's what they WANT to do. This isn't quantum physics, there is no rewards for genius. The best case you cleverly jump the tripwire and blast the turrets in mid air (massage that ego baby), the worst is you die and start over. Eventually if he can find no way to action hero it, the gamer may try to figure out the door or window.
If I'm playing an FPS type of game (one that requires twitch and speed) then that's all I want to do. Otherwise i'll play an adventure game with a better story and more time to smell the roses.
Anarchy Online was sci-fantasy. It's major failing was insane complexity, and big launch problems that did not give it momentum.
EVE is an amazing game, but not an MMORPG to the best of my knowledge. It's more about trading and empire building. It's quite clear that the victor of such a game will be he who plays it the most...
I haven't seem a good sci-fi MMORPG yet, and it's sad because sci-fi has as much to offer a creative person as fantasy...
The high level game could be quite fun and you could be reasonably casual to play it, but it's more like a sporting event and less like a video game. You gather a bunch of people up in the same place at the same time, get all your gear, then go play for about 4-6 hours. Sorta like a giant golf outing: you all bring exotic equipment and wear funny clothes and swing at a lot of shit.
The high level game consists mostly of "raids" involving 18-72 (or more in some cases) people destroying a dungeon (at a high rate of speed) and tackling a "boss" mob who is typically horribly overpowered, unfair and drops very nice stuff should you kill it.
It actually can be fun, if you can put together 72 of your closest friends. With 72 strangers, I'd rather have my teeth removed through my anus with hot tweezers.
Think Kill Bill Volume 1, then Volume 2. The last expansion "Gates of Discord" (GoD in eq jargon) was by most accounts a dismal failure. It is reasonably innaccessible to non super-guilded types (i.e. the masses), but pathetically easy to the super-guilds on any server. So in other words, it made no one at all happy...
To be honest, it was tiny, expensive ($30), did not appeal to the masses, and was not worthy of discussion and immediately after release (feb '04) made me mad. I spent the most time ever in those expansion zones last night (a whopping 2 hrs) and still don't get it...
As if to throw salt on the wound SOE announced for E3 the next expansion "Omens of War", which seems to have almost an identical plot, promises many of the same ill-defined effects, involves much the same plot-line and characters, and by all appearances appears to be "Gates of Discord - The Missing Content". It honestly seems like an excuse to bilk us out of another $30 for the same content.
I think however it is premature to call EQ dead, or dying. I have played almost all the MMOGs out there, and they're all mostly boring time sinks. I still find EQ to be the least boring of all of them which is why I kept dropping my subscription and returning.
A more fair review as I actually have an HD1000 and have been evaluating it for a real deployment.
The operating system is totally invisible. I can't complain about it, I only know that it's linux from the hype. Most commercial STBs in this market seem to run linux, to argue the box is faulty for that reason exposes some latent bias.
The software application that is the "front end" and user visible portion of the system leaves much to be desired however. For example, by default, it only uses DHCP. There is no way that I have found (short of using the serial port and knowing linux) to program a static IP. There's no easy way to line up or script content, there's no way to organize a "channel lineup" etc. In general to use this in a real system (other than some geek hackfest), you'd have to rewrite the front end.
The software does nice things like recognize windows file shares and auto-mount various storage. However a HUGE failing is that it does not recognize IP Multicast. This is a horrible failing IMHO (at least since that's what our deployment is based around). Pace, Amino, etc. are SD STBs that seem to have no troubles with this, and work quite nicely.
The upside is that Roku provides, free of charge, an SDK for adding things to the box, the downside is that I'd have to do it myself...thus this box is not ready to "drop in" any deployment.
As far as hardware goes there is really no interface that is not included. It takes the entire universe of memory types and includes them, no doubt contributing to the $299 price point. As someone who has built STBs in the past, I imagine they are losing a little $ at that price. In my testing however it's rock solid, displays HD video at 1080i on any clip I've used (I am in the process of testing it's MPEG bitrate limitations right now). Everything performs solidly.
As far as bugs go, it seems to have an issue with DHCP in that when the lease expires, it does not attempt to renew. This is problematic for anyone really...who wants to have their box randomly stop playing...
I personally think the Roku is an excellent prototype box. Clearly the designers didn't know what market they wanted to hit, and are looking for a buyer to bring their box to a particular segment (i.e. Cable company, fiber to the home, consumer). By the same token, it's not appropriate yet for any of those markets and while I would and will buy one for my home, I can't figure out what else to recommend it for...
Don't worry, it took them 3.5 years to get sound to work on the PC in a way that encouraged me not to disable it and play mp3's. That they didn't get it right on the Mac should shock no one.
Not having access to higher level content should piss anyone off and they should quit. From time to time people have left EverQuest en masse, and though it takes a while, they eventually solve the problems.
While the liberal arts clowns were having embodied virtual experiences with all the hot chicks, the engineers were forced to read hypertext narrative in between cramming sessions to get off.
I hope you're kidding and your post should have been mod'ed as funny;)
I cannot think of a single job I may have that you would be precluded from because you understand VM swapping algorithms. In fact I think you'd be able to pick up both positional AND containment based widget toolkits, and explain to me exactly why one or the other might be better for any application I ask you to write.
The roll-your-own vs. off the shelf is rarely an engineering problem. Usually it comes down to the classic time vs. money scenario. Depending on which is in least supply you may have to do either one, and you quite often don't get to make that decision. Some brat with an MBA or worse some old ex-"engineer" (who still thinks he's "in touch", and "technical") is gonna make that decision for you. As you can tell, you may learn to hate him for it, but it'll happen.
Of course I hope you'd be able to do more than recite VM swapping algorithms to me. I don't want people who only solve solved problems. I expect, perhaps even in an interview, to present you with a VM swapping-like problem and have you tell me either a) a good/optimal solution or b) that you understand the problem, you have a couple solutions that may or may not be ideal, and explain to me the pro's and con's of each. if you can do that, I trust you'll learn new toolkits in a couple days. O'Reilley makes a fortune on books for that purpose...
If the point was that you understand VM swapping algorithms but don't know how to write good code, then I understand your dilemma. The good news is that prospective employers at good balanced (age wise) companies, won't expect it of you. You can't learn it at school; your professors can't teach it because most of them don't know either simply because they don't write code on a daily basis. That's not what they do, and they don't understand what you want to do (or think it's trivial, which is as equally wrong as your own bombastic statements).
Yeah I was wondering the same thing. Rather than attempt to search their network for people using any of a dozen different filesharing networks, why not just compel everyone to write in and apologize, thus revealing their guilt?
I think the only solution is to require companies to enforce patents for them to be valid, like with trademarks. That way if JPEG is tainted we'll know. Patents are intended to protect individuals who create unique and non-obvious work. Patenting parts of a standard only has the effect of extorting people trying to use standards, which not only defeats the purpose, but hurts everyone as our products cannot interoperate. If was known to be patented, I am 100% sure that it could have been circumvented. Enforcing the patent now is not using the uniqueness and insight of the idea, but instead, the widespread and accepted use of it.
Lots of emerging standards have this issue. There are 2 choices, require litigation or reduce patent lifetime to 5 years.
You have your reasoning and you're entitled to it, however I think our limited jail cell space ought to be used for more significant crimes. To me, taping a movie on your camcorder is a misdemeanor offense, such criminals ought to have to go pick up highway trash for a few months and other "rehabilitating" punishments. Selling copies of said tapes to the public ought to land you in prison for a year or so, that's the real crime.
The ability to force artistic vision on audiences is not a constitutionally protected right, provided it is the choice of the recipient to block material he or she individually does not like. We give parents the right to censor events for their children, perhaps that's not right but I can't bring myself to believe that a young child needs to see Full Metal Jacket raw. This device allows people to choose what they or their children see, there's nothing wrong with that as long as the choice to see it "as intended" is there.
Artists, like scientists and pretty much anyone who makes an item for public distribution must accept that their creation has to stand on it's own once it leaves their hands. If they make an item or statement, no matter how true or valid, that is found to be offensive to the ideals of their audience, some people will choose to blind themselves. That's OK. It's only bad when the choice is made for them by a third party, without their consent.
Sometimes this means that art and pornography are lumped into the same category of "objectionable material". Like it or not, artists and the like must either accept this or maintain a private gallery only for those they deem worthy.
Slashdot Mirror
State universities are substantially cheaper of course. However the only other suitable universities near me are Princeton, which is around $50k/year, and Columbia which is somehwat less, I'm not sure exactly how much so. Both are obviously private.
There was no state school within commuting distance to where i lived that had an engineering program worth considering, so I simply didn't consider that.
Compared to the schools I did consider, mine was a bargain!
Yes but I think you may be in the minority. If you WANT to do such a thing, nothing forces you to avoid those classes. I take dance classes with my wife, for our enjoyment. I have not found a use for the waltz in the office. However the rest of us felt forced not only into lost time, but money as well. I just wanted a Bachelors to get in the field, and a Masters to earn more money in it. I enjoyed my masters far more because it was concentrated on my subject, short (I was able to take summer courses) and I felt accomplished.
My school was relatively inexpensive and I lived at home. Tuition was $22k/yr (on average), so $88k for a BS. After scholarships etc. we ended up having to pay about 30% of that (which we paid rather than FAFSA'd), however I believe fully half went to "junk" liberal arts I didn't want. It's simply a waste.
I disagree that there is any technical/career advantage in liberal arts courses. Critical thinking isn't taught in ANY course, or derived from it. You can't learn it, you develop it like the ability to speak, or NOT if you are not suitably challenged or curious. I know many engineers who simply lack the ability to solve problems, some with degrees from MIT and Princeton.
What I think many may be saying, unclearly or untruthfully is that you do not want a program that is too focused on specifics. You don't want to teach a kid GCC, Make, and the C language and throw him out the door as a "Computer Scientist". He will be crippled. I think Math, Physics, Digital Logic, Computer Archtiecture and then the core CS courses, in addition to a few practical courses that are not often offered that amount to "training".
There were many good experiences I had in school but the frustration I and my friends had with how much money we were spending and how much time we were spending on stuff that we didn't "sign up for" was at times intense. I guess I wouldn't want my own children to have to endure that (nor do I necessarily want to pay for it).
I was speaking in terms of the design of memory chips themselves, each of them supporting a 10G link for example could today, support compatible bandwidth from the current parallel bus approach and do bank switching in a number of ways.
/. title makes it worse =/
You could still build a "DIMM-like" module to plug in on board, so your module density would be comparable if you wanted, or throw down your "scrabble" chips singly. Either one works, and has an exact parallel to the more far fetched Sun approach. Let's face it, the real bottleneck in ANY system is memory, they're slow. I can see us building a scrabble board (out of FR4 instead of silicon), where each tile is a big serial link that speaks some common language. You could lay down "words" of memory that take 4 tiles, or one at a time. Each specific technology probably has it's own wants.
128-bit wide and 256-bit wide busses on video cards do exist and are always to memories, and memories these days are 16-bit wide each even for Alpha's, and ganged together in parallel. So the bandwidth to each device is quite replaceable by a serial link.
I did not see a paper attached in the article, and the article itself seemed the usual level of gee-whiz that I'd expect. I saw what seemed to be fact (packaging bottleneck for ASICs) with some things that seemed to be under investigation "there is a noticeable change in voltage levels at the receiver" to wild speculation about scrabble tiles and chips. I can't quite make the leap from what they COULD be talking about that would allow for a multitde of "wireless" links, to "dropping chips in like scrabble tiles". As far as I know if you have two circuits communicating by some electric field between them, you either a) have to use enough power that one can talk to the other through a large area or b) have to place them very close together. When someone says they have two "wireless" circuits communicating by mutual capacitance, I think of two networks placed so closely together that the air (or silicon) between them forms an adequate dielectric that they act like a capacitor has been placed in series between them. To do this in silicon you'd really need to place the CPU "pad" very close to the cache "pad", far more precicely than could be done in the field. The benefits I see are tangible, you could build flavors of a chip with various cache sizes without a complete relayout and requalification of the ASIC, but not revolutionary to us end users. In fact technology to do this also exists today but remains unused (www.sychip.com used to do this, and may still in some form). I'm interested in reading a paper written by an engineer if you know where it is, but the article sounds just way too gee-whiz for me to take seriously and the
I am aware of a few DARPA projects where they are building these "chip sandwhiches" but they're all made in a semiconductor fab, and the yields aren't too great.
So maybe this technology they're coming up with isn't too bad from the standpoint of helping build a more flexible processor design, I don't see that you'll have a scrabble computer though, or that us end-users will see much benefit. We'll still have to buy a CPU with some amount of cache built on, and still have to have PCBs to connect the CPU to everything else. And they're still going to be beasts to design.
Ok, first I'm insinuating what would be more practical from a sytem design perspective, not proposing a solution on existing technology. It happens existing technology is damn close though.
Second, wireless communication by means of mutual capacitance is reasonably serial in itself, by clever design you may be able to simultaneously communicate with some small number of chips (on the order of 10s), but it's still reasonably serial. I think it would however be harder to get high bandwidth wireless serial links to be as fast and reliable as wired links.
Third I'm not sure about your numbers, but I'm in telecomm not computers. An 800MHz FSB is actually a 200Mhz bus sampled 4x during a single clock period. The bus, I think, is 64-bits wide and is bi-directional. At best, ignoring turn around times etc., I see a bandwidth of 51.2Gbits/sec, which corresponds to the number Intel claims. This can be realized today with about 5 serial links. I'm thinking of a driver on Xilinx FPGAs that I just happen to be using today, I'm sure there are others maybe better, but it supports 10.135Gbit/s as is imminently designable on a PCB even with very long trace lengths.
That 800MHz FSB in fact connects to the north bridge, which aggregates traffic from the DDR Bus, AGP bus and PCI busses that is bound to the CPU. Instead, why not give one serial link to each memory (usually at least 4, often more), one to AGP and another to PCI? You're still perfectly modular and have a far simpler PCB that you can still "mix and match" assuming everyome supports the same protocols.
Fundamentally I think Sun has a good idea but the wrong implementation. It would seem what is needed is a standardized, unburdened serial link standard, and a standardized protocol for data transfer across it for compute oriented applications. Sun however probably wishes to own the IP and make money from it, so this isn't in their interests.
I think this is true, I pirated stuff non-stop as a minor, I had lots of time and the inability to generate money. Piracy takes lots of time: download/driving times, getting bad files and finding new sources, building up karma required for downloads on private sites etc. Pain in the ass.
Now I have money but no time so I never pirate games. I do tend to buy the games I enjoyed way back when where available. Doom somehow I never liked that much, but I look forward to Space Quest 93.
Ground planes alone would just help capture all that leaked RF energy and unhelpfully put it on your power supply return (which also can pick up outside interference and help screw with your chip sandwhich). Stuff can be done about this but you'd still have to have an off chip board with the right set of filters etc.
Instead of screwball stuff I think it'd be more helpful to simply find ways to drastically reduce the number of pins required. Most of these chips are huge because of the 128 signal wide memory/data busses, N control/configuration pins, big address busses etc. Much of this can be replaced with comparatively fewer high speed links.
What I'd rather see is parallel busses being replaced by very high speed serial links (all patented to hell and back of course), perhaps one link per expected peripheral (memory, adjunct processor, i/o bus, video bus, etc.). One could build a very cheap PCB that could almost be hand assembled. The problem is that each peripheral would also have to be compatible with the link. No one builds DDRs with SERDES links for example...
80% of the boards I've designed have been pratically identical in terms of core functionality, but they've been completely different in implementation because of the differing interfaces.
You can cancel online at ibillcs.com if you are willing to throw your credit card number and email address on that site to do the subscription lookup.
Of course, I had to call the # to find that out.
I can't bring myself to actually buy a video game these days since the last 6-7 offline games I've played have sucked horribly. Either yet another FPS copy, a bad perversion on a previous game that was good (i.e. MOO3), or a game that was not at all what it pretended to be on the box.
The best I can come to is trying them out first and buying if it's worth teh money. But most games aren't even worth the download time lately...
All I know is my own situation, but I suspect it's common:
-> I get a great idea for something the world needs
-> I need to prototype it, but have grossly insufficient funds
-> I seek people with money who are willing to loan me some $$$ if I pay them back more than they gave me
-> Now I must build the device on the capital given to me, and figure out how to make more $$$ from it than I put in to it, to pay off my loan AND continue to eat for the next 2 years.
-> I have 2 choices, prototype and sell my invention itself (actually the "IP" behind it), or go in to business as an equipment maker. If I do the former, I can sell the invention and pay off my debts, some other company then makes the $$$ from it. If I do the latter, then I will require even MORE investment, and require even MORE profits to stay afloat.
Enter profit, aka "no good deed goes unpunished".
There is a big gap in the universe between the people with good ideas, and the people with money. Maybe the right thing to do is prohibit companies from owning patents.
But the original poster was correct. SWG *WAS* very much dissapointing and this expansion *IS* only adding content that was originally slated for the initial launch. I was not even on beta but I received regular emails about features for SWG that seemed to DIMINISH every month until launch.
More to the point, I played SWG, it was boring, there was no content and the thrust of the game seemed to be a levelling treadmill that was even MORE dull than EverQuest (if that's possible).
Somehow I read your post and believe the opposite. Games are ultimately designed for the player to win: be it by solving problem, developing spiderman twitch reflexes, etc. Winning or beating a level usually involves an ego massage along with it ("Excellent mighty warrior...") Second, no matter how realisitic a game ultimately looks, suspension of disbelief is voluntary subconcious submission to unreality, and only disturbed people actually think it's real. The rest of us know in our heart that our victory is hollow, but enjoy it for the challenge of it. Do you worry much about dying in a game? Not really, only if it sets you back somehow.
What Semper Fi above was alluding to is that "real life" is quite the opposite. You may not win, your entire purpose in life may be to be blown up. You may be the smartest, fastest and well equipped and STILL eat it. When you kill someone, he stays dead forever. When his brains are splattered on the wall, you are faced with the harsh reality that IT COULD HAVE BEEN YOU, and may yet be. Further your survival may have less to do with native instict & training than by sheer dumb luck.
The trouble is, until you're in one of these situations, I don't think anyone (myself included) really understands it. Games romanticize it, make it sexy etc. I doubt they are a deterrent to anything.
On the other hand I don't really believe many people join based on notions of heroics. I suspect the #1 driver is paying for college or more simply steady work & a familiar institutional atmosphere. During war-time it may be a different story, but then the country may need foolish people in search of glory to fill the ranks, and they will become heroes even if they die unromantically.
I have a MS in EE. I was offered a scholarship to do a PhD. However during my MS thesis work I worked closely with PhD candidates and suffered perhaps a fraction of what they did. Don't get me wrong, I enjoy what I do, but I eat corporate shit for $$$ alone, so I declined. My observations are as follows:
1) PhD is a lot of work for yourself, and 1000x more work doing your professors busy work (papers etc.)
2) PhD slave labor wages are less than those of any given malaysian factory child if you count the total number of hours worked and divide that into your scholarship/stipend/grant/etc.
3) If you are not a US citizen/permanent resident and are on a scholarship to get a PhD in the US, you are fucked. Bring the vasoline and bend over.
4) If your goal is simply to get a degree to get more money, stop at your masters.
5) If your PhD is not in a subject actively investigated by the corporate world be willing to accept an academic position after getting your degree, or find another subject. It's heartbreaking to see people get their degree and realize they are either stuck in academia or worse, take a job in industry doing work outside their expertise making the same they would have as a masters (i.e. degree worthless).
6) If at all possible GET A COMPANY TO FUND YOUR PHD! This is harder now than it used to be, but it is THE way to go. I can't recommend it enough, if I personally thought there was money in a PhD this is what I'd do myself. If your professor administrates whatever finances your degree, and you are above broccoli intelligence, he WILL try to hold you as long as he can (5-7 years in most schools). If your company is paying the bill they are quite good at getting you in and out ASAP. Avg stay of corporate funded PhD students in my experience was 3 years. Do this!
7) Stupid people can get PhD's far easier than smart people. Simply put, professors want stupid people out of their hair, if they can't wash em out, they graduate em. Just like elementary school.
Obviously if gamers see the danger, and in masses go towards it, then clearly that's what they WANT to do. This isn't quantum physics, there is no rewards for genius. The best case you cleverly jump the tripwire and blast the turrets in mid air (massage that ego baby), the worst is you die and start over. Eventually if he can find no way to action hero it, the gamer may try to figure out the door or window.
If I'm playing an FPS type of game (one that requires twitch and speed) then that's all I want to do. Otherwise i'll play an adventure game with a better story and more time to smell the roses.
Life kills. It's morbid but a fact. The question is what to do with what time you have.
Anarchy Online was sci-fantasy. It's major failing was insane complexity, and big launch problems that did not give it momentum.
EVE is an amazing game, but not an MMORPG to the best of my knowledge. It's more about trading and empire building. It's quite clear that the victor of such a game will be he who plays it the most...
I haven't seem a good sci-fi MMORPG yet, and it's sad because sci-fi has as much to offer a creative person as fantasy...
The high level game could be quite fun and you could be reasonably casual to play it, but it's more like a sporting event and less like a video game. You gather a bunch of people up in the same place at the same time, get all your gear, then go play for about 4-6 hours. Sorta like a giant golf outing: you all bring exotic equipment and wear funny clothes and swing at a lot of shit.
The high level game consists mostly of "raids" involving 18-72 (or more in some cases) people destroying a dungeon (at a high rate of speed) and tackling a "boss" mob who is typically horribly overpowered, unfair and drops very nice stuff should you kill it.
It actually can be fun, if you can put together 72 of your closest friends. With 72 strangers, I'd rather have my teeth removed through my anus with hot tweezers.
Think Kill Bill Volume 1, then Volume 2. The last expansion "Gates of Discord" (GoD in eq jargon) was by most accounts a dismal failure. It is reasonably innaccessible to non super-guilded types (i.e. the masses), but pathetically easy to the super-guilds on any server. So in other words, it made no one at all happy...
To be honest, it was tiny, expensive ($30), did not appeal to the masses, and was not worthy of discussion and immediately after release (feb '04) made me mad. I spent the most time ever in those expansion zones last night (a whopping 2 hrs) and still don't get it...
As if to throw salt on the wound SOE announced for E3 the next expansion "Omens of War", which seems to have almost an identical plot, promises many of the same ill-defined effects, involves much the same plot-line and characters, and by all appearances appears to be "Gates of Discord - The Missing Content". It honestly seems like an excuse to bilk us out of another $30 for the same content.
I think however it is premature to call EQ dead, or dying. I have played almost all the MMOGs out there, and they're all mostly boring time sinks. I still find EQ to be the least boring of all of them which is why I kept dropping my subscription and returning.
A more fair review as I actually have an HD1000 and have been evaluating it for a real deployment.
The operating system is totally invisible. I can't complain about it, I only know that it's linux from the hype. Most commercial STBs in this market seem to run linux, to argue the box is faulty for that reason exposes some latent bias.
The software application that is the "front end" and user visible portion of the system leaves much to be desired however. For example, by default, it only uses DHCP. There is no way that I have found (short of using the serial port and knowing linux) to program a static IP. There's no easy way to line up or script content, there's no way to organize a "channel lineup" etc. In general to use this in a real system (other than some geek hackfest), you'd have to rewrite the front end.
The software does nice things like recognize windows file shares and auto-mount various storage. However a HUGE failing is that it does not recognize IP Multicast. This is a horrible failing IMHO (at least since that's what our deployment is based around). Pace, Amino, etc. are SD STBs that seem to have no troubles with this, and work quite nicely.
The upside is that Roku provides, free of charge, an SDK for adding things to the box, the downside is that I'd have to do it myself...thus this box is not ready to "drop in" any deployment.
As far as hardware goes there is really no interface that is not included. It takes the entire universe of memory types and includes them, no doubt contributing to the $299 price point. As someone who has built STBs in the past, I imagine they are losing a little $ at that price. In my testing however it's rock solid, displays HD video at 1080i on any clip I've used (I am in the process of testing it's MPEG bitrate limitations right now). Everything performs solidly.
As far as bugs go, it seems to have an issue with DHCP in that when the lease expires, it does not attempt to renew. This is problematic for anyone really...who wants to have their box randomly stop playing...
I personally think the Roku is an excellent prototype box. Clearly the designers didn't know what market they wanted to hit, and are looking for a buyer to bring their box to a particular segment (i.e. Cable company, fiber to the home, consumer). By the same token, it's not appropriate yet for any of those markets and while I would and will buy one for my home, I can't figure out what else to recommend it for...
Don't worry, it took them 3.5 years to get sound to work on the PC in a way that encouraged me not to disable it and play mp3's. That they didn't get it right on the Mac should shock no one.
Not having access to higher level content should piss anyone off and they should quit. From time to time people have left EverQuest en masse, and though it takes a while, they eventually solve the problems.
The gap between science and art is fundamental:
While the liberal arts clowns were having embodied virtual experiences with all the hot chicks, the engineers were forced to read hypertext narrative in between cramming sessions to get off.
I hope you're kidding and your post should have been mod'ed as funny ;)
I cannot think of a single job I may have that you would be precluded from because you understand VM swapping algorithms. In fact I think you'd be able to pick up both positional AND containment based widget toolkits, and explain to me exactly why one or the other might be better for any application I ask you to write.
The roll-your-own vs. off the shelf is rarely an engineering problem. Usually it comes down to the classic time vs. money scenario. Depending on which is in least supply you may have to do either one, and you quite often don't get to make that decision. Some brat with an MBA or worse some old ex-"engineer" (who still thinks he's "in touch", and "technical") is gonna make that decision for you. As you can tell, you may learn to hate him for it, but it'll happen.
Of course I hope you'd be able to do more than recite VM swapping algorithms to me. I don't want people who only solve solved problems. I expect, perhaps even in an interview, to present you with a VM swapping-like problem and have you tell me either a) a good/optimal solution or b) that you understand the problem, you have a couple solutions that may or may not be ideal, and explain to me the pro's and con's of each. if you can do that, I trust you'll learn new toolkits in a couple days. O'Reilley makes a fortune on books for that purpose...
If the point was that you understand VM swapping algorithms but don't know how to write good code, then I understand your dilemma. The good news is that prospective employers at good balanced (age wise) companies, won't expect it of you. You can't learn it at school; your professors can't teach it because most of them don't know either simply because they don't write code on a daily basis. That's not what they do, and they don't understand what you want to do (or think it's trivial, which is as equally wrong as your own bombastic statements).
Yeah I was wondering the same thing. Rather than attempt to search their network for people using any of a dozen different filesharing networks, why not just compel everyone to write in and apologize, thus revealing their guilt?
I think the only solution is to require companies to enforce patents for them to be valid, like with trademarks. That way if JPEG is tainted we'll know. Patents are intended to protect individuals who create unique and non-obvious work. Patenting parts of a standard only has the effect of extorting people trying to use standards, which not only defeats the purpose, but hurts everyone as our products cannot interoperate. If was known to be patented, I am 100% sure that it could have been circumvented. Enforcing the patent now is not using the uniqueness and insight of the idea, but instead, the widespread and accepted use of it.
Lots of emerging standards have this issue. There are 2 choices, require litigation or reduce patent lifetime to 5 years.
You have your reasoning and you're entitled to it, however I think our limited jail cell space ought to be used for more significant crimes. To me, taping a movie on your camcorder is a misdemeanor offense, such criminals ought to have to go pick up highway trash for a few months and other "rehabilitating" punishments. Selling copies of said tapes to the public ought to land you in prison for a year or so, that's the real crime.
The ability to force artistic vision on audiences is not a constitutionally protected right, provided it is the choice of the recipient to block material he or she individually does not like. We give parents the right to censor events for their children, perhaps that's not right but I can't bring myself to believe that a young child needs to see Full Metal Jacket raw. This device allows people to choose what they or their children see, there's nothing wrong with that as long as the choice to see it "as intended" is there.
Artists, like scientists and pretty much anyone who makes an item for public distribution must accept that their creation has to stand on it's own once it leaves their hands. If they make an item or statement, no matter how true or valid, that is found to be offensive to the ideals of their audience, some people will choose to blind themselves. That's OK. It's only bad when the choice is made for them by a third party, without their consent.
Sometimes this means that art and pornography are lumped into the same category of "objectionable material". Like it or not, artists and the like must either accept this or maintain a private gallery only for those they deem worthy.