I think AMD merging with ATI will finally make GPUs in the GHz range possible. Not long ago, I read an article on chip design (especially CPUs), and optimizing a chip by hand for various tasks is something reserved to the very big CPU makers Intel, AMD and probably IBM. Others have sub-GHz chips because they don't have the required manpower and financial resources to design entirely custom circuits. This means only critical parts of the chip are optimized by hand and everything else is using blocks from CAD software where you specify what you need and it designs something that works for you.
This is good, but designing by hand is way better: the designer can take into consideration heat dissipation, similar circuit proximity, etc. I'm not a chip designer so I can't say very much on this, but according to the article I read, if AMD or Intel were using the same kind of resources they use for CPU design, but on GPUs, we'd already have like 600 FPS in FEAR on a passively-cooled 65nm GPU (which would be connected through a high-speed bus like HT3 to the CPU, RAM, etc.)
Thinking about it, I really HOPE AMD and ATI merge!
Oh, and AMD chipsets are always the most stable for their chips, so I can't wait to see them produce more and drop VIA et al. (which have exactly the same resource problem I just described).
So, yeah, go AMD and let's kick Conroe's 15-years-old roots to hell. (See TomsHardware's article on the Conroe, a descendant of the Pentium M, which is a descendant of the Pentium 3, which is a descendant of the Pentium Pro, itself a RISC chip with an x86 instruction translator... ugly design).
Oh yeah and I also have to say this: the last time AMD had a huge win (the Athlon64 design), it was because it brought new people in through an acquisition. I have a lot of confidence we'll see this happening again if they grab ATI's hand.
This was because the internal codes used by Microsoft in the document files were language dependent. Ridiculous, isn't it?
Well if you don't mind, I'd really appreciate proof of this statement. I rely mostly on the opinion that MS products are cross-language as their development tools (VS) are prepared to take care of that. I don't see MS using underfeatured development environments where language is part of the code itself. What might have happened to you is that patches are not always available for all languages at the same time. If you upgrade your english systems (which will probably have patches released BEFORE other languages), you might no longer have the compatibility with other language (non-patched) systems.
Oh, and if your network admin had the most basic knowledge of MS products, he would have installed Microsoft Multilingual User Interface packages on English versions of Windows to support non-english workers. That way, he would have been able to manage and update ALL machines at the same time without downloading two versions of all upgrades.
My two canadian cents, which might not be worth anything in your country.
It's a bad idea because what happens when the driver ABI changes? You have to wait umpteen months for the company to get off thier asses and fix it - like nVidia.
That's exactly what Microsoft does when they change the driver model (e.g. from NT4 to 2000/XP, and from 2000/XP to Vista). The key is simple : try not to do it too often, have a very good reason to do so, and assist developpers in learning how to use it (MSDN is pretty damn hard to beat on that).
Besides, WHQL insures that all drivers pass certain criteria and Linux should have the same kind of mechanism. You don't need the source to verify that a driver performs as advertised and doesn't break and/or touch parts of the OS that you specifically ask not to be played with.
Compare that to the Linux driver mess we have today and you'll start to understand why many companies HAVE to stay away from that OS. It's not that it's bad, it's simply that many companies have specific needs regarding hardware compatibility. Where I work, for instance, we have very high-end 3D hardware that simply can't be made to work very well under Linux and we can't justify that to our customers. Cost is not an issue when you need to satisfy your clients.
AMD already let rumors go out about their possible integration of PCI-e. Next, we don't know but I suspect they'll get interrested in GPUs. With a modular multiple-cores architecture, it wouldn't be that hard to insert a GPU core in there. The only problem is heat... The advantages are insane : PCI-e on the cpu means the chipset doesn't need to use the hypertransport bus to route that traffic between cards and cpu! You get a direct connection between whatever cards you have and the main CPU (which is also the memory controller).
AMD really does have an edge with this technology and I hope they get as far as they can with integration. As one poster said earlier, one day we'll have a full x86 system-on-a-chip (SoC), which means cheaper boards, no need for cards, and maybe even no need for any hard drive. I hope AMD gets us to that level.
The one major drawback with OLED monitors at the moment is their lifespan.
You are correct, but this device is not an OLED monitor, it is a regular LCD panel with a LED-based backlight (LED, not OLED). This means the CCFL (Cold-Cathode Fluorescent Lamp) is replaced by an array of Light-Emitting Diodes that have 1) Greater lifespan, 2) Better color stability, 3) Require less power, 4) Luminosity / color variance.
So MS is sending me spam now and can get away with and get positive credit for doing so?
Can I pleaaase stop reading the same ignorance again and again in these comments? Read TFA and you'll learn that MS had the mails redirected to a fake mail server, effectively blocking everything out.
Microsoft must have hired all the brains 'cause all I read is the same dumb comment over and over... and over again.
Well I didn't RTFA but my guess on this is that it would permit things like an over-the-air Internet. That is simply the future of all wireless communications. Every computer connects to all available network devices around it and then automatically announces and routes packets between peers. All peers doing this means peer A can contact peer C through peer B since A and C both see B and B is announcing to A that it also sees C and vice-versa.
Now, the implications of such a thing on a software level is pretty deep. Instead of connecting to an ISP providing you with an IP address, such a peer would have to be able to route and operate without IP addresses, routing packets based on their MAC addresses alone. IP is still possible, but this is a very different story.
Many scenarios are possible where such a technology would have great value : in-car computers communicating with each others to gather traffic and road-quality data from a few hundred cars in front of you (your in-car computer communicating with the next one in front of you and so on).
I am sure Microsoft is trying to get there very soon as they are pushing their mobile development very hard since a few years. After all, they want to be with people everywhere they are, from their Family Room (Xbox), to the Office (Windows), and everywhere in between (Windows Mobile, Smartphone, etc.)
Wild cats, in Australia at least, need to be shot whenever possible to protect other native species. Its not pretty, its not nice, and it not really fair either, but there is little alternative.
Man, do you seriously think killing ONE of these things will change ANYTHING? I guess not. Well killing them all would change everything, but if we deploy enough forces to track and kill them all, why not track and capture them all for better things like putting them elsewhere or in zoos, or at least, keeping a few alives to prevent the cat's extinction?
You see, killing just isn't the answer anymore. There's so much species near extinction on the planet that your morality is extremely questionable. Hunting for food is on the limit of acceptable. Killing for fun or because the cat is making a mess of the area is just stupid in these times. Go kill something MUCH MORE harmfull to the environment like humans, instead... Man, what a stupid tought.
You're absolutely right, and may I add that Russians had two or three other stations before? The ISS is a laaaaaaarge moneypit, and it's only possibly valuable use would be to actually send it to Mars orbit so the future explorers have all the equipment to study and prepare for Mars landings. Too bad it's in such a low orbit that it would be almost as expensive to boost it enough for such a mission than make a brand new one on Mars...
Well that's nice, just yesterday I submitted that same story and got rejected!! I guess it's slashdot's editors that just get convinced to publish the story when they see the same article twice.
Some songs should sell for $0.01 and some should sell for $10.00.
If you can afford paying 10 bucks for ONE SINGLE SONG, good for you. I certainly CANNOT and will continue staying away from music stores until I can actually afford to buy songs for a one-night party without having to sacrifice beer.
Then the market will decide for itself. That's what he says he wants, right?
Well not really. What he wants is simple : force iTunes (and others) to sell the songs at the price he wants.
While this is normal in some sectors (look at how brand chips all are the same price everywhere, from the grocery store to the automatic dispenser... in fact the price is printed on the bag itself), it is not very good for the public as all retailers will offer the product at the same price. The retailer gets profit not from each sale, but from the volume of sales he is able to sustain each quarter. This is clearly an attempt to cut down on the number of different retailers to help corporations get a stranglehold on the product itself.
Oh, and if you think my example of comparing online music sales to bags of chips, think again : the margin of profit on both products is VERY similar (ie. outrageous).
Well, your phone is crappy! Mine survived a late-night plunge into the pitcher. It was kinda nice to see the phone all lighted-up inside that beer! And it's smell the next morning made me remember how drunk I was.
Sorry for replying to myself, but I forgot to say something. Ever tried to join AND authenticate your users through Active Directory, have Roaming Profiles and Network Home Drive?
Well it works, after about two weeks of hacking, and it stopped working as soon as I added DFS to my SAN.
Everything is doable on Linux as it is on Windows. It's just a matter of time. 1 hour of Windows configuration equals about 20 hours of Linux fucking-around. And it isn't even fun because of the damn unreadable manpages.
For starters, why isn't there a shell that emulates the windows CMD? As useful greps / finds / pipe chains are, they can't match the simplicity of a DIR/S. I've been using MS-DOS for more than 15 years, moving to Linux is restarting everything from scratch. Is that really NECESSARY?
As with nearly EVERY piece of code on linux, I feel the programmers are constantly trying to do "better" than what exists on Windows. Now, tell me if I'm wrong, but the HUGE user base of Microsoft's OS can't be anything else than a very powerful indicator of it's user-friendliness.
I can't believe how cryptic some things are on Linux. Try to setup a software RAID array. Now, remove a disk, let it become critical, then put another disk (bigger one, different maker). Watch as you'll be greated by the big emptyness of options. Some fools will tell you "RTFM" or edit this.rc and that.conf, but I tell you : DELETE YOUR PARTITIONS AND INSTALL WINDOWS.
When you manage a hundred workstations with the latest technology in them, you DON'T want to have to deal with thousands of precious configuration files that were hacked for this and that.
On Windows, I have my Group Policies. All computers, be them laptops, workstations or servers, adhere to these policies and I can change some high-level decisions without having to figure out how to implement it in KDE, Gnome and all those dirty piece of crap.
Don't get me wrong, I also maintain many BSD servers, but that's what they are : servers and BSD. I'd NEVER put any Linux/BSD crap in the face of the user unless they need an application that is ABSOLUTELY not available for Windows.
I think this flight has certainly killed the possible flight to repair the HST. Non ISS flights are too risky now.
Well, considering that NASA didn't change anything PAL-ramp-related, I think this mission is a huge success as no dangerous-sized chunk of foam ripped off the tank where they applied modifications. Now, they know that the PAL-ramp might cause trouble so they'll be looking onto it and all other areas that didn't undergo modifications.
That's great news for astronauts and the shuttle. Who cares if it still stays on the ground for another 6 months? If the next flight has no considerably-sized debris peeling off (which is highly likely - every problem has a fix), the shuttle will return to full usage and that would reduce the urge for a new space vehicle.
Reducing that urge is CRITICAL to creating a great space vehicle. Fast designs are good and all, but carefull and planned designs is very important to acheive something really superior. If we are to visit Mars one day, NASA needs to take it's time designing something that could be used safely to stay there for a while! I, for one, would like to see a Mars station being built to complement the ISS. Imagine where that could lead us?
Now, just dreaming but I'd really like to see in my lifetime some kind of biological research being done on Mars to accomodate plants and maybe even small animals to live on it's conditions. Once we have that, we'll be 1) able to modify our own specie to help prevent diseases, 2) able to understand and possibly even create life whenever and wherever we want with a little research.
The implications of long-term space travel are far more interresting than the travel itself!
This means there could be a small hole in the wall of the station, big enough for a cock or a turd, but not big enough for the air to fall out of the space station. Problem solved.
What a stupid idea. First, what about the dangers of dispersed matter floating around in space, hitting satellites and other spacecrafts? Second, do you even have a clue on pressurization? Even the tiniest hole will leak enough air out of the station to pose a serious health risk to it's inhabitants.
No, they aren't taking a risk returning that garbage from space. They are REMOVING one from the 30-billion+ dollars piece of equipment.
OH and by the way, the garbage is not all shit and piss. Much of it's weight is from a damaged gyroscope they plan to check on th see what got wrong with it so they can make better ones in the future.
Slashdot Mirror
I think AMD merging with ATI will finally make GPUs in the GHz range possible. Not long ago, I read an article on chip design (especially CPUs), and optimizing a chip by hand for various tasks is something reserved to the very big CPU makers Intel, AMD and probably IBM. Others have sub-GHz chips because they don't have the required manpower and financial resources to design entirely custom circuits. This means only critical parts of the chip are optimized by hand and everything else is using blocks from CAD software where you specify what you need and it designs something that works for you.
This is good, but designing by hand is way better: the designer can take into consideration heat dissipation, similar circuit proximity, etc. I'm not a chip designer so I can't say very much on this, but according to the article I read, if AMD or Intel were using the same kind of resources they use for CPU design, but on GPUs, we'd already have like 600 FPS in FEAR on a passively-cooled 65nm GPU (which would be connected through a high-speed bus like HT3 to the CPU, RAM, etc.)
Thinking about it, I really HOPE AMD and ATI merge!
Oh, and AMD chipsets are always the most stable for their chips, so I can't wait to see them produce more and drop VIA et al. (which have exactly the same resource problem I just described).
So, yeah, go AMD and let's kick Conroe's 15-years-old roots to hell. (See TomsHardware's article on the Conroe, a descendant of the Pentium M, which is a descendant of the Pentium 3, which is a descendant of the Pentium Pro, itself a RISC chip with an x86 instruction translator... ugly design).
Oh yeah and I also have to say this: the last time AMD had a huge win (the Athlon64 design), it was because it brought new people in through an acquisition. I have a lot of confidence we'll see this happening again if they grab ATI's hand.
My two cents.
You just made my day. This is the best comment I've ever read on slashdot. Thanks for posting it.
I've just submitted it again! Wait until they post it a fourth time! ;-)
MOUAHAHAHAHAHAHAHAHA!!! Well the stupidest thing you did this afternoon is also the funniest thing I read this afternoon! :P
Boring....
This was because the internal codes used by Microsoft in the document files were language dependent. Ridiculous, isn't it?
Well if you don't mind, I'd really appreciate proof of this statement. I rely mostly on the opinion that MS products are cross-language as their development tools (VS) are prepared to take care of that. I don't see MS using underfeatured development environments where language is part of the code itself. What might have happened to you is that patches are not always available for all languages at the same time. If you upgrade your english systems (which will probably have patches released BEFORE other languages), you might no longer have the compatibility with other language (non-patched) systems.
Oh, and if your network admin had the most basic knowledge of MS products, he would have installed Microsoft Multilingual User Interface packages on English versions of Windows to support non-english workers. That way, he would have been able to manage and update ALL machines at the same time without downloading two versions of all upgrades.
My two canadian cents, which might not be worth anything in your country.
It's a bad idea because what happens when the driver ABI changes? You have to wait umpteen months for the company to get off thier asses and fix it - like nVidia.
That's exactly what Microsoft does when they change the driver model (e.g. from NT4 to 2000/XP, and from 2000/XP to Vista). The key is simple : try not to do it too often, have a very good reason to do so, and assist developpers in learning how to use it (MSDN is pretty damn hard to beat on that).
Besides, WHQL insures that all drivers pass certain criteria and Linux should have the same kind of mechanism. You don't need the source to verify that a driver performs as advertised and doesn't break and/or touch parts of the OS that you specifically ask not to be played with.
Compare that to the Linux driver mess we have today and you'll start to understand why many companies HAVE to stay away from that OS. It's not that it's bad, it's simply that many companies have specific needs regarding hardware compatibility. Where I work, for instance, we have very high-end 3D hardware that simply can't be made to work very well under Linux and we can't justify that to our customers. Cost is not an issue when you need to satisfy your clients.
My two cents.
AMD already let rumors go out about their possible integration of PCI-e. Next, we don't know but I suspect they'll get interrested in GPUs. With a modular multiple-cores architecture, it wouldn't be that hard to insert a GPU core in there. The only problem is heat... The advantages are insane : PCI-e on the cpu means the chipset doesn't need to use the hypertransport bus to route that traffic between cards and cpu! You get a direct connection between whatever cards you have and the main CPU (which is also the memory controller).
AMD really does have an edge with this technology and I hope they get as far as they can with integration. As one poster said earlier, one day we'll have a full x86 system-on-a-chip (SoC), which means cheaper boards, no need for cards, and maybe even no need for any hard drive. I hope AMD gets us to that level.
The one major drawback with OLED monitors at the moment is their lifespan.
You are correct, but this device is not an OLED monitor, it is a regular LCD panel with a LED-based backlight (LED, not OLED). This means the CCFL (Cold-Cathode Fluorescent Lamp) is replaced by an array of Light-Emitting Diodes that have 1) Greater lifespan, 2) Better color stability, 3) Require less power, 4) Luminosity / color variance.
So MS is sending me spam now and can get away with and get positive credit for doing so?
Can I pleaaase stop reading the same ignorance again and again in these comments? Read TFA and you'll learn that MS had the mails redirected to a fake mail server, effectively blocking everything out.
Microsoft must have hired all the brains 'cause all I read is the same dumb comment over and over... and over again.
Just as you didn't install Windows 2000 on that old 486 computer you used to run Windows 95 on.
Well I didn't RTFA but my guess on this is that it would permit things like an over-the-air Internet. That is simply the future of all wireless communications. Every computer connects to all available network devices around it and then automatically announces and routes packets between peers. All peers doing this means peer A can contact peer C through peer B since A and C both see B and B is announcing to A that it also sees C and vice-versa.
Now, the implications of such a thing on a software level is pretty deep. Instead of connecting to an ISP providing you with an IP address, such a peer would have to be able to route and operate without IP addresses, routing packets based on their MAC addresses alone. IP is still possible, but this is a very different story.
Many scenarios are possible where such a technology would have great value : in-car computers communicating with each others to gather traffic and road-quality data from a few hundred cars in front of you (your in-car computer communicating with the next one in front of you and so on).
I am sure Microsoft is trying to get there very soon as they are pushing their mobile development very hard since a few years. After all, they want to be with people everywhere they are, from their Family Room (Xbox), to the Office (Windows), and everywhere in between (Windows Mobile, Smartphone, etc.)
Wild cats, in Australia at least, need to be shot whenever possible to protect other native species. Its not pretty, its not nice, and it not really fair either, but there is little alternative.
Man, do you seriously think killing ONE of these things will change ANYTHING? I guess not. Well killing them all would change everything, but if we deploy enough forces to track and kill them all, why not track and capture them all for better things like putting them elsewhere or in zoos, or at least, keeping a few alives to prevent the cat's extinction?
You see, killing just isn't the answer anymore. There's so much species near extinction on the planet that your morality is extremely questionable. Hunting for food is on the limit of acceptable. Killing for fun or because the cat is making a mess of the area is just stupid in these times. Go kill something MUCH MORE harmfull to the environment like humans, instead... Man, what a stupid tought.
I think the exact same thing. What a fucking shitfull asshole. Makes me sick.
You're absolutely right, and may I add that Russians had two or three other stations before? The ISS is a laaaaaaarge moneypit, and it's only possibly valuable use would be to actually send it to Mars orbit so the future explorers have all the equipment to study and prepare for Mars landings. Too bad it's in such a low orbit that it would be almost as expensive to boost it enough for such a mission than make a brand new one on Mars...
Well that's nice, just yesterday I submitted that same story and got rejected!! I guess it's slashdot's editors that just get convinced to publish the story when they see the same article twice.
Some songs should sell for $0.01 and some should sell for $10.00.
If you can afford paying 10 bucks for ONE SINGLE SONG, good for you. I certainly CANNOT and will continue staying away from music stores until I can actually afford to buy songs for a one-night party without having to sacrifice beer.
Then the market will decide for itself. That's what he says he wants, right?
Well not really. What he wants is simple : force iTunes (and others) to sell the songs at the price he wants.
While this is normal in some sectors (look at how brand chips all are the same price everywhere, from the grocery store to the automatic dispenser... in fact the price is printed on the bag itself), it is not very good for the public as all retailers will offer the product at the same price. The retailer gets profit not from each sale, but from the volume of sales he is able to sustain each quarter. This is clearly an attempt to cut down on the number of different retailers to help corporations get a stranglehold on the product itself.
Oh, and if you think my example of comparing online music sales to bags of chips, think again : the margin of profit on both products is VERY similar (ie. outrageous).
Vincent
Well, your phone is crappy! Mine survived a late-night plunge into the pitcher. It was kinda nice to see the phone all lighted-up inside that beer! And it's smell the next morning made me remember how drunk I was.
Hm, seems like slashdot has some kind of protection for that crap. Oh, well now it works.
Space.com's image script is ridiculously exploitable. Click for yourself.
Sorry for replying to myself, but I forgot to say something. Ever tried to join AND authenticate your users through Active Directory, have Roaming Profiles and Network Home Drive?
Well it works, after about two weeks of hacking, and it stopped working as soon as I added DFS to my SAN.
Everything is doable on Linux as it is on Windows. It's just a matter of time. 1 hour of Windows configuration equals about 20 hours of Linux fucking-around. And it isn't even fun because of the damn unreadable manpages.
Fuck it.
Man, you are SO right!
/S. I've been using MS-DOS for more than 15 years, moving to Linux is restarting everything from scratch. Is that really NECESSARY?
.rc and that .conf, but I tell you : DELETE YOUR PARTITIONS AND INSTALL WINDOWS.
For starters, why isn't there a shell that emulates the windows CMD? As useful greps / finds / pipe chains are, they can't match the simplicity of a DIR
As with nearly EVERY piece of code on linux, I feel the programmers are constantly trying to do "better" than what exists on Windows. Now, tell me if I'm wrong, but the HUGE user base of Microsoft's OS can't be anything else than a very powerful indicator of it's user-friendliness.
I can't believe how cryptic some things are on Linux. Try to setup a software RAID array. Now, remove a disk, let it become critical, then put another disk (bigger one, different maker). Watch as you'll be greated by the big emptyness of options. Some fools will tell you "RTFM" or edit this
When you manage a hundred workstations with the latest technology in them, you DON'T want to have to deal with thousands of precious configuration files that were hacked for this and that.
On Windows, I have my Group Policies. All computers, be them laptops, workstations or servers, adhere to these policies and I can change some high-level decisions without having to figure out how to implement it in KDE, Gnome and all those dirty piece of crap.
Don't get me wrong, I also maintain many BSD servers, but that's what they are : servers and BSD. I'd NEVER put any Linux/BSD crap in the face of the user unless they need an application that is ABSOLUTELY not available for Windows.
Parent, thanks for the GREAT comment.
I think this flight has certainly killed the possible flight to repair the HST. Non ISS flights are too risky now.
Well, considering that NASA didn't change anything PAL-ramp-related, I think this mission is a huge success as no dangerous-sized chunk of foam ripped off the tank where they applied modifications. Now, they know that the PAL-ramp might cause trouble so they'll be looking onto it and all other areas that didn't undergo modifications.
That's great news for astronauts and the shuttle. Who cares if it still stays on the ground for another 6 months? If the next flight has no considerably-sized debris peeling off (which is highly likely - every problem has a fix), the shuttle will return to full usage and that would reduce the urge for a new space vehicle.
Reducing that urge is CRITICAL to creating a great space vehicle. Fast designs are good and all, but carefull and planned designs is very important to acheive something really superior. If we are to visit Mars one day, NASA needs to take it's time designing something that could be used safely to stay there for a while! I, for one, would like to see a Mars station being built to complement the ISS. Imagine where that could lead us?
Now, just dreaming but I'd really like to see in my lifetime some kind of biological research being done on Mars to accomodate plants and maybe even small animals to live on it's conditions. Once we have that, we'll be 1) able to modify our own specie to help prevent diseases, 2) able to understand and possibly even create life whenever and wherever we want with a little research.
The implications of long-term space travel are far more interresting than the travel itself!
This means there could be a small hole in the wall of the station, big enough for a cock or a turd, but not big enough for the air to fall out of the space station. Problem solved.
What a stupid idea. First, what about the dangers of dispersed matter floating around in space, hitting satellites and other spacecrafts? Second, do you even have a clue on pressurization? Even the tiniest hole will leak enough air out of the station to pose a serious health risk to it's inhabitants.
No, they aren't taking a risk returning that garbage from space. They are REMOVING one from the 30-billion+ dollars piece of equipment.
OH and by the way, the garbage is not all shit and piss. Much of it's weight is from a damaged gyroscope they plan to check on th see what got wrong with it so they can make better ones in the future.