3dfx screwed themselves
on
The Age of Nvidia
·
· Score: 4, Interesting
nVidia may not have climbed as high and fast as it did if 3Dfx didn't hand them the entire market on a silver platter. If you would recall, 3Dfx all of a sudden decided they wanted to pull out of the OEM chip market and manufacture their own boards. So they bought STB and called up Diamond and Creative and all the other major video vendors and went something like: "No more 3Dfx chips for you. All your base are belong to us. Ha ha ha."
And who stepped into their place to fill the void left by them? nVidia.
Now the problem is that STB boards, by my many years experience in building systems, were incredibly low quality. I loved Diamonds, and had never had a problem with them. 3Dfx was gone from the Diamond line now, and I sure as hell wasn't going to start throwing money at STB boards, so I stuck with Diamond and bought a Viper v770...By then many games were supporting OpenGL and DirectX, so compatibility wasn't really an issue, unless I wanted to play Tomb Raider or something, but my old Voodoo would still work with the Viper. As an added bonus, and as you all know, nVidia's chips blew away anything 3Dfx had. nVidia's hardest battle, market acceptance, had been handed to them by 3dfx complete with bow and ribbon.
And then it was with complete irony that nVidia purchased 3Dfx. I love the tech industry.
That's what happens when you get trigger happy on the submit button, the comment was to read as:
This technology has been around for years. The difference with this article on CNET is that it appears it has fallen into the hands of local law enforcement, rather than outfits like the CIA or FBI. It's called TEMPEST.
And it appears I'm smoking something mentioning local law enforcement, I don't know where I got that from, skimmed the article too fast or mixed it up with something else I was reading..Heh, I shouldn't be posting this morning.
This technology has been around for years. The difference with this article on CNET is that it appears it has fallen into the hands of local law enforcement, rather than outfits like the CIA or FBI. It's called
.tv does the same thing, except they base the cost of the domain off of its projected popularity. For instance, maybe I want bill.tv for my domain name. Oh it will only cost $5000/YEAR! I wonder why no one has taken it? Go to www.tv and see for yourself. As far as.pro goes, everyone may think that doctors, lawyers, and accountants have money coming out of their ears, but they also don't tend to buy junk they see on daytime television commercials, which is exactly what this.pro TLD is. Are they really going to shell out 10 times more for.pro when they can get any other one for cheaper? Professionals tend not to be stupid. It seems to me that the folks at.pro have proclaimed themselves as the "professional" registrar before they even have any customers. At $300 a pop the only people who buy these domains will be the same people who watch the home shopping network and reply to spam with their full information.
I am living in a bandwidth-challenged community, and looked into Starband as an alternative, and luckily had a customer who subscribed so I could see it in practical use. My beef with it was as followed:
1. High, unpredictable latency. Forget gaming.
2. $700 price tag. $500 for the gear, and $200 for the FCC monkey to point it at the satellite. Forget doing it yourself, the pointing is simply too precise to be able to eyeball it.
3. USB interface only. This makes it difficult to share access with other computers on your network, short of using the lame Internet Sharing from M$.
4. Flaky connectivity, packet loss, and web page timeouts.
5. All for $70/mo, last I checked.
In other words, no way.
Earlier, I made a post here on Slashdot about a project I was planning on setting up between here and where I work. Well, it works. At a distance of 4.5 miles away I am able to leech off the company's T1 at full speed, with only 2.2ms of latencty point to point between the dishes. It was built entirely from off-the-shelf gear, like Orinoco cards, a couple of junker Pentiums running Linux acting as wireless routers, and two run of the mill 2.4GHz high gain parabolic grid antennas (the same type they use for wireless cable service in rural areas, I hear), all for just under $1000. It has been surprisingly reliable with virtually no dropouts and almost no packet loss (unless a bird lands on my transmitter:), at least no more than your average DSL connection.
If you are within sight of an area that can get DSL or cable, I highly recommend looking into this option. The up front cost may be a little prohibitive, but you would be pretty deep in the hole as it is if you went with satellite.
This is a perfect example of a guy who thinks too much and got totally out of touch with reality. Let's take a look at this, shall we?
He wants to build solar panels which will somehow instantly generate "harmless" microwaves (according to the article) and beam them to Earth. Unless he developed some kind of revolutionary new kind of solar panel, the ones he's planning on using are the run-of-the-mill photovoltaic variety. Presumably he wants to use all the voltage from these panels to drive a powerful microwave transmitter. The receiver will pick up the alternating magnetic field (i.e. radio frequency) and convert it back into electricity. Is this possible? Yes. Is it practical? Hell no. There are a couple of major things one must consider. First of all, the microwaves would lose an insane amount of energy traveling the 250,000 miles back to Earth. Unless I miss my guess, falloff for electromagnetic waves in space is (1/r*r, where r is distance). Physics geeks, please feel free to correct me if I'm wrong about that. Even if you were able to generate the massive amount of energy in question, turn it into microwaves, send it to Earth, and convert it back into enough electricity to power something more than, say, a lightbulb, the moon isn't geostationary. You would constantly be having to aim the beam. So now you get to power a sophisticated (and quite expensive) tracking system as well, and let's not forget atmospheric diffraction. If the moon supposedly can get 13,000 terawatts from the sun (god knows what kind of convoluted mathematics he used to come up with that), then what does the Earth get, having 13 times more surface area?? It sounds like this fool would be better off taking his $15 billion and planting solar farms all along the equator of the Earth than living out his fantasy of beaming electricity though space.
Wireless could be the way out of bandwidth hell...
on
Cringely's Bank Shot
·
· Score: 5, Informative
I myself live in a bandwidth black hole which I just happen to be in the center of. So, I actually started researching and buying gear to hook into work's T1, which is about 4.8 miles away. The gear I decided on was two Orinoco (or WaveLAN as they used to be called) cards with Linux boxes to match to keep costs down (besides, Linux makes for a great wireless router). My antennas are 24dBi gain Hyperlink parabolic grid antennas. I already have the cards working in my Linux installations and am ready to hook up the antennas soon. The only tricky part is that my path to work is slightly obscured so I'm hoping I have enough power and gain to be able to punch though. Hopefully the bandwidth gods will look favorably upon me. I've never had a high speed connect at home (and probably never will if this doesn't work:/)
One of the coolest projects I found while researching this was the HPWREN project at UCSD. Check out their pictures, it's hella cool. In a nutshell they are running a 45Mbps (802.11a) wireless backbone across the Santa Margarita Ecological Reserve using mostly off-the-shelf equipment, for the purpose of hooking together the facilities strewn across it. They even have remote cameras hooked in that can be remotely controlled through the network, and other testing stations that send data back to them in realtime.
I dropped an email to the project lead and I asked him what kind of gear they used. He said they used a Western Multiplex Tsunami for their backend, Hyperlink for their antennas and WaveLAN and Cisco Aironet for their PCMCIA cards (you can now see how I constructed my parts list:)) I also asked how he got around mountains and such. Well, in certain places they have powered relay stations. Naturally I wondered how they were powered, and he said some of them they could get electricity to, but others they actually have solar panels powering the relays. Damn. For you real hackers he mentioned there was a parts list for the solar power array somewhere on the website, but I never bothered to try and find it.
I've noticed some arguments regarding amplifying 802.11, and thought I'd help clear it up. FCC Part 15.247 governs the unlicensed ISM (Industrial, Scientific, Medical) band, and dictates that you can amplify the signal up to 1 watt (1000mw) This gets tricky when you start using directional antennas >6dBi gain though. You may find more detailed info here..
Technological pundits have been theorizing the crux of any given technology for decades. If you told someone in 1960 that in 40 years everyone would be able to get 40GB drives for $100 in a 3.5" form factor, and 1GHz processors for the same price, and 256MB memory for $40, they'd have you committed. During my time in the computer field I have heard many statements saying that processors will only be able to go so fast, and hard disks will get so big, yadadada. All of these "limits" were imposed by physical limitations just as superparamagnetism is. But then someone (usually IBM) comes out with this whizzbang technology that changes everything, just like the article said. Unlike the article, I don't think that superparamangetism will be any different. A new technology will be out that will address the issue. I trust Moore's Law more than I do the words of doomsayers.
The linked news article off of this story scared the hell out of me. This has nothing to do with the "gaming causes violence" argument, or any comment having to do with Asians being more apt to get sucked into this kind of thing because their culture is so restrictive. This article is scary because it's about people trading off their real lives for a fantasy game with crap graphics that they watch on a 2D monitor. Never mind that this is in Korea. This happens in the States too, to people who play EQ, UO, and so on and so forth. And that's only the beginning. There's a whole slew of MMPOG's coming out that promise to be even more immersive.
Let's not forget, everyone, where we were 20 years ago. Whoever remembers Infocom, raise your hand. We've gone from "look at ground" and "open mailbox" to games with realtime 3D environments, professionally composed musical scores, voice acting, budgets approaching movies, and the list goes on. But if you throw the Internet into the mix, you can have a persistent world that's contantly evolving. There is no way to "win", so therefore no reason to stop playing. This revolution has just started, and we've already heard stories of people completely unplugging from reality. They quit their jobs, cut off their friends and family, and all that matters is that they have enough money for their net connect and their game subscription. And the games are not yet even close to mimicking reality. The most insane advances in virtual reality are yet to come. With gear like this being developed, how long to you think it's going to be before you're not just watching through a glowing screen, but are actually in the game? And with computers and graphics cards getting ever more powerful, it won't belong before these games will be able to manipulate literally millions of polygons per second. This is in the near future, and it is peanuts compared to what is going to come. Scientists are starting to be able to understand how to interface sensory devices directly with the brain, already I have heard of advances in bionic eyes and ears. After that, feeding artifical sensory signals back into the brain is only a step away. The number of addicts to this kind of technology will be astronomical, and it's already starting to happen even with current tech. Why is this? Well, look at our reality, our sad little lives. We live in our little box-houses and sit in traffic for hours every day driving to jobs we hate. We kiss royal ass at work to hopefully get off the list of the people who are going to be fired that week. Then we spend more time in traffic going home, eat, sleep, repeat cycle until we're about 65, and hopefully have made enough coin spending our entire lives in a cubicle so we can take naps for the rest of our lives. Reminds me of that Police song "Synchronicity II". Vacation? Ha. Your life belongs to the company, buddy. You can be replaced in a heartbeat, and you're a number. Or, if you are still in school and are anything like me back when I was about that age, you get picked on a lot an are off in your own little world. An alternative reality doesn't sound so bad now. Some people use drugs or alcohol to get away, but now MMPOGs can and will be used as an escape, more so because they are, after all, just games. They offer the chance for one to become powerful and respected, not just another Borg in the American Collective. Mark my words, virtual reality will become a social problem, just like any addiction you could name today...And I am not kidding around when I say that it will be noticed when people start dying at their terminals.
Anything in the 10.x.x.x network can be considered private, ie. non-routable. Generally, there should be no problem using a 10 network for your private network, but there are a few exceptions, and one which comes to mind is my T1 provider, @Work. Whether or not it's safe to use a 10 network with your ISP, or any private addressing scheme for that matter, depends on how their routers are configured. IDEALLY, the ISP will configure all interfaces on their routers to use routable IP's. This makes for a nice, clean configuration which in turn makes route summarization a snap. This is much easier on the network in that the routing tables don't get ridiculously huge. @Work, however, for some reason uses IP's out of the 10 network to assign to serial interfaces to their border routers, the ones that connect their customers' T1's. If their customers use their assigned routable IP's then there is not a problem, but if they are set up on a private NAT'ed 10 network that just HAPPENS to be on the same subnet as their border router, it can raise hell. Actually, I had a friend who ran into this issue with Flashcom DSL when he was trying to set up a router to VPN into his work, which was all on a 10 network. It would take the VPN connection but he couldn't get to anything inside of his network, so he telnetted to a known server and ended up getting a Redback in Flashcom's network. Imagine his surprise. Since his DSL was bridged, not routed, he could see a bunch of their numbers out of the 10 network. He ended up having to do something funky with his routing tables to get the packets going to the right places. Anyway, barring all this jabber, in your case you are much better off with the 192.168 network, especially considering you have a pretty small network. Heck, I've set up entire WAN's based on a 192.168 network for all of the offices in it and there are more than enough IP's to go around.
Slashdot Mirror
nVidia may not have climbed as high and fast as it did if 3Dfx didn't hand them the entire market on a silver platter. If you would recall, 3Dfx all of a sudden decided they wanted to pull out of the OEM chip market and manufacture their own boards. So they bought STB and called up Diamond and Creative and all the other major video vendors and went something like: "No more 3Dfx chips for you. All your base are belong to us. Ha ha ha."
And who stepped into their place to fill the void left by them? nVidia.
Now the problem is that STB boards, by my many years experience in building systems, were incredibly low quality. I loved Diamonds, and had never had a problem with them. 3Dfx was gone from the Diamond line now, and I sure as hell wasn't going to start throwing money at STB boards, so I stuck with Diamond and bought a Viper v770...By then many games were supporting OpenGL and DirectX, so compatibility wasn't really an issue, unless I wanted to play Tomb Raider or something, but my old Voodoo would still work with the Viper. As an added bonus, and as you all know, nVidia's chips blew away anything 3Dfx had. nVidia's hardest battle, market acceptance, had been handed to them by 3dfx complete with bow and ribbon.
And then it was with complete irony that nVidia purchased 3Dfx. I love the tech industry.
That's what happens when you get trigger happy on the submit button, the comment was to read as: This technology has been around for years. The difference with this article on CNET is that it appears it has fallen into the hands of local law enforcement, rather than outfits like the CIA or FBI. It's called TEMPEST. And it appears I'm smoking something mentioning local law enforcement, I don't know where I got that from, skimmed the article too fast or mixed it up with something else I was reading..Heh, I shouldn't be posting this morning.
This technology has been around for years. The difference with this article on CNET is that it appears it has fallen into the hands of local law enforcement, rather than outfits like the CIA or FBI. It's called
.tv does the same thing, except they base the cost of the domain off of its projected popularity. For instance, maybe I want bill.tv for my domain name. Oh it will only cost $5000/YEAR! I wonder why no one has taken it? Go to www.tv and see for yourself. .pro goes, everyone may think that doctors, lawyers, and accountants have money coming out of their ears, but they also don't tend to buy junk they see on daytime television commercials, which is exactly what this .pro TLD is. Are they really going to shell out 10 times more for .pro when they can get any other one for cheaper? Professionals tend not to be stupid. .pro have proclaimed themselves as the "professional" registrar before they even have any customers. At $300 a pop the only people who buy these domains will be the same people who watch the home shopping network and reply to spam with their full information.
As far as
It seems to me that the folks at
Holy crap. Get that congressman a Slashdot account!
I am living in a bandwidth-challenged community, and looked into Starband as an alternative, and luckily had a customer who subscribed so I could see it in practical use. My beef with it was as followed:
:), at least no more than your average DSL connection.
1. High, unpredictable latency. Forget gaming.
2. $700 price tag. $500 for the gear, and $200 for the FCC monkey to point it at the satellite. Forget doing it yourself, the pointing is simply too precise to be able to eyeball it.
3. USB interface only. This makes it difficult to share access with other computers on your network, short of using the lame Internet Sharing from M$.
4. Flaky connectivity, packet loss, and web page timeouts.
5. All for $70/mo, last I checked.
In other words, no way.
Earlier, I made a post here on Slashdot about a project I was planning on setting up between here and where I work. Well, it works. At a distance of 4.5 miles away I am able to leech off the company's T1 at full speed, with only 2.2ms of latencty point to point between the dishes. It was built entirely from off-the-shelf gear, like Orinoco cards, a couple of junker Pentiums running Linux acting as wireless routers, and two run of the mill 2.4GHz high gain parabolic grid antennas (the same type they use for wireless cable service in rural areas, I hear), all for just under $1000. It has been surprisingly reliable with virtually no dropouts and almost no packet loss (unless a bird lands on my transmitter
If you are within sight of an area that can get DSL or cable, I highly recommend looking into this option. The up front cost may be a little prohibitive, but you would be pretty deep in the hole as it is if you went with satellite.
This is a perfect example of a guy who thinks too much and got totally out of touch with reality. Let's take a look at this, shall we?
He wants to build solar panels which will somehow instantly generate "harmless" microwaves (according to the article) and beam them to Earth. Unless he developed some kind of revolutionary new kind of solar panel, the ones he's planning on using are the run-of-the-mill photovoltaic variety. Presumably he wants to use all the voltage from these panels to drive a powerful microwave transmitter. The receiver will pick up the alternating magnetic field (i.e. radio frequency) and convert it back into electricity. Is this possible? Yes. Is it practical? Hell no. There are a couple of major things one must consider. First of all, the microwaves would lose an insane amount of energy traveling the 250,000 miles back to Earth. Unless I miss my guess, falloff for electromagnetic waves in space is (1/r*r, where r is distance). Physics geeks, please feel free to correct me if I'm wrong about that. Even if you were able to generate the massive amount of energy in question, turn it into microwaves, send it to Earth, and convert it back into enough electricity to power something more than, say, a lightbulb, the moon isn't geostationary. You would constantly be having to aim the beam. So now you get to power a sophisticated (and quite expensive) tracking system as well, and let's not forget atmospheric diffraction.
If the moon supposedly can get 13,000 terawatts from the sun (god knows what kind of convoluted mathematics he used to come up with that), then what does the Earth get, having 13 times more surface area?? It sounds like this fool would be better off taking his $15 billion and planting solar farms all along the equator of the Earth than living out his fantasy of beaming electricity though space.
I myself live in a bandwidth black hole which I just happen to be in the center of. So, I actually started researching and buying gear to hook into work's T1, which is about 4.8 miles away. The gear I decided on was two Orinoco (or WaveLAN as they used to be called) cards with Linux boxes to match to keep costs down (besides, Linux makes for a great wireless router). My antennas are 24dBi gain Hyperlink parabolic grid antennas. I already have the cards working in my Linux installations and am ready to hook up the antennas soon. The only tricky part is that my path to work is slightly obscured so I'm hoping I have enough power and gain to be able to punch though. Hopefully the bandwidth gods will look favorably upon me. I've never had a high speed connect at home (and probably never will if this doesn't work :/)
:)) I also asked how he got around mountains and such.
One of the coolest projects I found while researching this was the HPWREN project at UCSD. Check out their pictures, it's hella cool. In a nutshell they are running a 45Mbps (802.11a) wireless backbone across the Santa Margarita Ecological Reserve using mostly off-the-shelf equipment, for the purpose of hooking together the facilities strewn across it. They even have remote cameras hooked in that can be remotely controlled through the network, and other testing stations that send data back to them in realtime.
I dropped an email to the project lead and I asked him what kind of gear they used. He said they used a Western Multiplex Tsunami for their backend, Hyperlink for their antennas and WaveLAN and Cisco Aironet for their PCMCIA cards (you can now see how I constructed my parts list
Well, in certain places they have powered relay stations. Naturally I wondered how they were powered, and he said some of them they could get electricity to, but others they actually have solar panels powering the relays. Damn. For you real hackers he mentioned there was a parts list for the solar power array somewhere on the website, but I never bothered to try and find it.
I've noticed some arguments regarding amplifying 802.11, and thought I'd help clear it up. FCC Part 15.247 governs the unlicensed ISM (Industrial, Scientific, Medical) band, and dictates that you can amplify the signal up to 1 watt (1000mw) This gets tricky when you start using directional antennas >6dBi gain though. You may find more detailed info here..
Technological pundits have been theorizing the crux of any given technology for decades. If you told someone in 1960 that in 40 years everyone would be able to get 40GB drives for $100 in a 3.5" form factor, and 1GHz processors for the same price, and 256MB memory for $40, they'd have you committed. During my time in the computer field I have heard many statements saying that processors will only be able to go so fast, and hard disks will get so big, yadadada. All of these "limits" were imposed by physical limitations just as superparamagnetism is. But then someone (usually IBM) comes out with this whizzbang technology that changes everything, just like the article said. Unlike the article, I don't think that superparamangetism will be any different. A new technology will be out that will address the issue. I trust Moore's Law more than I do the words of doomsayers.
The linked news article off of this story scared the hell out of me. This has nothing to do with the "gaming causes violence" argument, or any comment having to do with Asians being more apt to get sucked into this kind of thing because their culture is so restrictive. This article is scary because it's about people trading off their real lives for a fantasy game with crap graphics that they watch on a 2D monitor. Never mind that this is in Korea. This happens in the States too, to people who play EQ, UO, and so on and so forth. And that's only the beginning. There's a whole slew of MMPOG's coming out that promise to be even more immersive.
Let's not forget, everyone, where we were 20 years ago. Whoever remembers Infocom, raise your hand. We've gone from "look at ground" and "open mailbox" to games with realtime 3D environments, professionally composed musical scores, voice acting, budgets approaching movies, and the list goes on. But if you throw the Internet into the mix, you can have a persistent world that's contantly evolving. There is no way to "win", so therefore no reason to stop playing. This revolution has just started, and we've already heard stories of people completely unplugging from reality. They quit their jobs, cut off their friends and family, and all that matters is that they have enough money for their net connect and their game subscription. And the games are not yet even close to mimicking reality. The most insane advances in virtual reality are yet to come. With gear like this being developed, how long to you think it's going to be before you're not just watching through a glowing screen, but are actually in the game? And with computers and graphics cards getting ever more powerful, it won't belong before these games will be able to manipulate literally millions of polygons per second. This is in the near future, and it is peanuts compared to what is going to come. Scientists are starting to be able to understand how to interface sensory devices directly with the brain, already I have heard of advances in bionic eyes and ears. After that, feeding artifical sensory signals back into the brain is only a step away. The number of addicts to this kind of technology will be astronomical, and it's already starting to happen even with current tech.
Why is this? Well, look at our reality, our sad little lives. We live in our little box-houses and sit in traffic for hours every day driving to jobs we hate. We kiss royal ass at work to hopefully get off the list of the people who are going to be fired that week. Then we spend more time in traffic going home, eat, sleep, repeat cycle until we're about 65, and hopefully have made enough coin spending our entire lives in a cubicle so we can take naps for the rest of our lives. Reminds me of that Police song "Synchronicity II". Vacation? Ha. Your life belongs to the company, buddy. You can be replaced in a heartbeat, and you're a number. Or, if you are still in school and are anything like me back when I was about that age, you get picked on a lot an are off in your own little world. An alternative reality doesn't sound so bad now. Some people use drugs or alcohol to get away, but now MMPOGs can and will be used as an escape, more so because they are, after all, just games. They offer the chance for one to become powerful and respected, not just another Borg in the American Collective. Mark my words, virtual reality will become a social problem, just like any addiction you could name today...And I am not kidding around when I say that it will be noticed when people start dying at their terminals.
Anything in the 10.x.x.x network can be considered private, ie. non-routable. Generally, there should be no problem using a 10 network for your private network, but there are a few exceptions, and one which comes to mind is my T1 provider, @Work. Whether or not it's safe to use a 10 network with your ISP, or any private addressing scheme for that matter, depends on how their routers are configured. IDEALLY, the ISP will configure all interfaces on their routers to use routable IP's. This makes for a nice, clean configuration which in turn makes route summarization a snap. This is much easier on the network in that the routing tables don't get ridiculously huge. @Work, however, for some reason uses IP's out of the 10 network to assign to serial interfaces to their border routers, the ones that connect their customers' T1's. If their customers use their assigned routable IP's then there is not a problem, but if they are set up on a private NAT'ed 10 network that just HAPPENS to be on the same subnet as their border router, it can raise hell. Actually, I had a friend who ran into this issue with Flashcom DSL when he was trying to set up a router to VPN into his work, which was all on a 10 network. It would take the VPN connection but he couldn't get to anything inside of his network, so he telnetted to a known server and ended up getting a Redback in Flashcom's network. Imagine his surprise. Since his DSL was bridged, not routed, he could see a bunch of their numbers out of the 10 network. He ended up having to do something funky with his routing tables to get the packets going to the right places. Anyway, barring all this jabber, in your case you are much better off with the 192.168 network, especially considering you have a pretty small network. Heck, I've set up entire WAN's based on a 192.168 network for all of the offices in it and there are more than enough IP's to go around.