Yet obscurity is considered security, for those purposes. Remember the suit about the universal garage-door opener? The code-rotating chip is a public enough part, it's just the position on the ring that was secret.
Re:what MS funded "study" about Linux isn't FUD?
on
Stallman vs Ken Brown
·
· Score: 1
Er...I should have said, "It's humerous that RMS is easily perceived as saying GNU and Linux are separate entities; Normally, quotations of him suggest that Linux and GNU are the same, since Linux isn't Linux without GNU."
But a computer: reformat HD, install linux: and you can use it all you want, in the privacy of your home.
I doubt they'd let you "own" the hardware. And their contracts would have a minimum term, with an early cancellation fee. And if you don't pay, it goes to the collections department.
He's talking about offloading the raytracing to the GPU. AFAIK, collision, physics and AI aren't dealt with there. (Animation is iffy...you could do some of it on the GPU, but, AFAIK, most is still done on the CPU.)
Yeah, well most people I know still think getting their phone for free is a good deal. And it may be...I don't know that buying your own phone+contract is cheaper.
But then, I wouldn't know...I don't have a cell phone.
Now if they are willing to give me the hardware free of charge...
Then it'l boil down to leasing the hardware. Which doesn't sound all that bad, so long as you know you'll be able to afford it. Neat way to stay on top of the technology.
Unfortunately, you won't be able to accrue old hardware that you can give to friends and family, or to set up for your kid to use. Or to serve as a cheap firewall.
I'm sure people who are interested in "new" computers wouldn't mind leasing them. And people who like to hack their hardware wouldn't mind buying them.
Without Microsoft interfering, I'd wager that people would lease hardware and software as a unit. However, I suspect Microsoft is going to want to charge their own subcription fees.
If you think about it, we have subcription fees already. You periodically pay for the upgrade to the next version of Windows, OSX or some distributions of Linux. The difference is we're currently free to continue using any of those products without being forced into upgrading using contract clauses.
I'm personally reminded of those WebTV boxes of a while back.
IIRC, the software was never in sync with even the commonly used W3C standards. Even in the days when HTML 4.0 was new, WebTV was considered lackluster.
Well, I imagine it serves as a sort of continuous "Who's hot, and who's not" announcement.
People will probably send these maintainers the email equivalent of a slap on the back, and thumbs up.
Also, it draws attention to the developers. Some of these guys might end up hired as a result of these announcements. Tom Lord especially, since two of his projects won.
You'd still need to transfer data through the AGP bus. More data than currently is done, in fact.
Modern graphics cards assemble each frame from a collection of images, or textures, that are provided it. The GPU performs mathematical operations on these textures in order to orient them somewhere in the field of view.
If you performed all of the operations on the CPU, you'd not only be taking up instruction cycles, you'd have to transmit entire frames through the AGP bus. 1600x1200x24bytes works out to about 44Mb per frame. At 24 fps, that's about one gigabit per second. That's an awful low refresh rate. Let's raise it to 56Hz. Now we're at 2.33Gb/s, more than normal PCI. Let's go for a smooth 85Hz: Now we're at 3.54Gb/s. Let's look forwards to higher resolutions, say, 3200x2400@85Hz: 14.17Gb/s. More than the latest HyperTransport revision can handle. By this time, you've already crowded out hard drive and network access. Your sound might be in trouble too.
That's an awful lot of bandwidth. And don't forget the space on the CPU die, and cache pollution for other processes. And Memory latency, not to mention the fact that a lot of that memory could be used for other game data.
That's not to say there wouldn't be advantages. You could also conceivably perform physics calculations like collision detection and simple FEA.
All in all, though, it's more efficient to have a multiprocessor setup where specific tasks are run on specialized hardware.
well, people could do nasty things to the routers and screw with the connections.
And what happens if people use your connection to download child porn? Worse, they do it twice. The second time while police are monitoring traffic on your wireless connection.
In the first case, you'll have a tough time proving it wasn't you. I expect they'd rather have a false positive, and convict, than a false negative, and acquit. In the latter case, you may only need to demonstrate that none of your devices have the MAC addresses that were performing the HTTP requests. (Though I doubt that such a "common carrier" defense would work.)
Still worse, the wireless transfer of illegal data could be done within range of your network, under your DHCP server, but otherwise be completely unrelated to you. Yet you still get pulled in and charged with posession, or trafficking, or whatever.
Re:Just publish the report already!
on
Stallman vs Ken Brown
·
· Score: 2, Insightful
Stop being afraid of reviews and books - the truth will let itself be seen...
It's at public forums like Slashdot where the truth is revealed, in reaction to the anger and other emotions people exhibit when encountering information that starkly contrasts what they believe.
Remember when Microsoft used stock photos for the "I used to be a Mac fan, but I'm better now" ads? It was a Slashdot reader, using Slashdot forums, that pointed out that the photo they used was stock.
Re:what MS funded "study" about Linux isn't FUD?
on
Stallman vs Ken Brown
·
· Score: 1, Interesting
I don't mean to sound trollish, but I think it's funny that, for once, RMS is emphasizing that GNU and Linux really are separate.
(Yes, I'm aware that it's the Distro that he usually insists includes the name "GNU/Linux", but to the uncalibrated ear, it often sounds like he's equating GNU and Linux.)
IBM/everyone did the same when they went from 30 pin to 72 pin SIMMS, then to DIMMS, then to DDR DIMMS.
The switch from 30-pin to 72-pin SIMMS, and then on to DIMMS, occured as a result of the memory bus width increasing.
When the normal width for a memory bank reached 64 bits, they went from 30-pin to 72-pin. (It was easier to insert one 72-pin SIMM than four 30-pin SIMMs.)
When the normal width for a memory bank reached 128 bits, they switched from SIMMs to DIMMS. (Again, it is easier to insert one DIMM than two 72-pin SIMMs. Imagine what would happen if every RAM upgrade meant you had to insert eight 30-pin SIMMs.)
However, the switch from SDRAM DIMMs to DDR SDRAM DIMMs was strictly for bandwidth, not for making memory insertion cheaper/easier. ("Cheaper" because one DIMM socket takes up much less board space than eight 30-pin SIMM sockets.)
No kidding. I went to the TechReport review because it was the first one that's not running slow, and my jaw dropped. AMD's really taking the lead with their recent offerings.
I recall the stories on SlashDot about how the GPUs on modern ATI/nVidia cards are "many times faster than P4s"... well, yes, but you can't run Linux on them...
To elaborate:
GPUs still only run at a couple of hundred of MHz, but their dedicated circuitry allows them to perform certain matrix calculations much faster than x86 chips currently do, even with vector instruction extensions like MMX and SSE/SSE2.
Here are a couple of links to relevant articles. (12)
Speaking of affording it...My grandfather just finished chewing me out for adding RAM to his system. (Scanning photos was taking forever, with lots of swapping.)
Me, I considered it entertainment spending. I was improving the performance of their system, so they probably wouldn't need my help as much.
Unfortunately, it doesn't seem to help. Windows ME must put everything it possibly can in swap space.
Me, I try to use my journal to stimulate discussion. And I link to the individual journal entries in my sig, so people know about it.
My current journal is a little different than usual...it's about an idea for a multiplayer online D&D environment, except that it aids pen-and-paper-style gameplay instead of MMORPG-style gameplay.
My other recent journal entries have been about concepts like brother-sister twin NPC clerics (where one channels positive energy, and the other channels negative energy), using dreams to acquaint players with their characters' strengths, etc.
I get a lot of interesting feedback, averaging six comments per journal entry. (Since I started focusing the topics, anyway.)
Slashdot Mirror
Yet obscurity is considered security, for those purposes. Remember the suit about the universal garage-door opener? The code-rotating chip is a public enough part, it's just the position on the ring that was secret.
Er...I should have said, "It's humerous that RMS is easily perceived as saying GNU and Linux are separate entities; Normally, quotations of him suggest that Linux and GNU are the same, since Linux isn't Linux without GNU."
...or something to that effect.
But a computer: reformat HD, install linux: and you can use it all you want, in the privacy of your home.
I doubt they'd let you "own" the hardware. And their contracts would have a minimum term, with an early cancellation fee. And if you don't pay, it goes to the collections department.
and there's a lot of important data that can essentially be held hostage.
Ahh...In comes the beauty of USB2 and FireWire800 hard drives.
He's talking about offloading the raytracing to the GPU. AFAIK, collision, physics and AI aren't dealt with there. (Animation is iffy...you could do some of it on the GPU, but, AFAIK, most is still done on the CPU.)
Yeah, well most people I know still think getting their phone for free is a good deal. And it may be...I don't know that buying your own phone+contract is cheaper.
But then, I wouldn't know...I don't have a cell phone.
Now if they are willing to give me the hardware free of charge...
Then it'l boil down to leasing the hardware. Which doesn't sound all that bad, so long as you know you'll be able to afford it. Neat way to stay on top of the technology.
Unfortunately, you won't be able to accrue old hardware that you can give to friends and family, or to set up for your kid to use. Or to serve as a cheap firewall.
I'm sure people who are interested in "new" computers wouldn't mind leasing them. And people who like to hack their hardware wouldn't mind buying them.
Without Microsoft interfering, I'd wager that people would lease hardware and software as a unit. However, I suspect Microsoft is going to want to charge their own subcription fees.
If you think about it, we have subcription fees already. You periodically pay for the upgrade to the next version of Windows, OSX or some distributions of Linux. The difference is we're currently free to continue using any of those products without being forced into upgrading using contract clauses.
I'm personally reminded of those WebTV boxes of a while back.
IIRC, the software was never in sync with even the commonly used W3C standards. Even in the days when HTML 4.0 was new, WebTV was considered lackluster.
More than you'll care to eat.
Trust me on that one.
At risk of being redundant, $500 isn't a whole lot for all the time these guys put into their projects.
On the other hand, the recognition may land them jobs as developers or as managers of a group of developers.
Well, I imagine it serves as a sort of continuous "Who's hot, and who's not" announcement.
People will probably send these maintainers the email equivalent of a slap on the back, and thumbs up.
Also, it draws attention to the developers. Some of these guys might end up hired as a result of these announcements. Tom Lord especially, since two of his projects won.
You'd still need to transfer data through the AGP bus. More data than currently is done, in fact.
Modern graphics cards assemble each frame from a collection of images, or textures, that are provided it. The GPU performs mathematical operations on these textures in order to orient them somewhere in the field of view.
If you performed all of the operations on the CPU, you'd not only be taking up instruction cycles, you'd have to transmit entire frames through the AGP bus. 1600x1200x24bytes works out to about 44Mb per frame. At 24 fps, that's about one gigabit per second. That's an awful low refresh rate. Let's raise it to 56Hz. Now we're at 2.33Gb/s, more than normal PCI. Let's go for a smooth 85Hz: Now we're at 3.54Gb/s. Let's look forwards to higher resolutions, say, 3200x2400@85Hz: 14.17Gb/s. More than the latest HyperTransport revision can handle. By this time, you've already crowded out hard drive and network access. Your sound might be in trouble too.
That's an awful lot of bandwidth. And don't forget the space on the CPU die, and cache pollution for other processes. And Memory latency, not to mention the fact that a lot of that memory could be used for other game data.
That's not to say there wouldn't be advantages. You could also conceivably perform physics calculations like collision detection and simple FEA.
All in all, though, it's more efficient to have a multiprocessor setup where specific tasks are run on specialized hardware.
well, people could do nasty things to the routers and screw with the connections.
And what happens if people use your connection to download child porn? Worse, they do it twice. The second time while police are monitoring traffic on your wireless connection.
In the first case, you'll have a tough time proving it wasn't you. I expect they'd rather have a false positive, and convict, than a false negative, and acquit. In the latter case, you may only need to demonstrate that none of your devices have the MAC addresses that were performing the HTTP requests. (Though I doubt that such a "common carrier" defense would work.)
Still worse, the wireless transfer of illegal data could be done within range of your network, under your DHCP server, but otherwise be completely unrelated to you. Yet you still get pulled in and charged with posession, or trafficking, or whatever.
Stop being afraid of reviews and books - the truth will let itself be seen...
It's at public forums like Slashdot where the truth is revealed, in reaction to the anger and other emotions people exhibit when encountering information that starkly contrasts what they believe.
Remember when Microsoft used stock photos for the "I used to be a Mac fan, but I'm better now" ads? It was a Slashdot reader, using Slashdot forums, that pointed out that the photo they used was stock.
I don't mean to sound trollish, but I think it's funny that, for once, RMS is emphasizing that GNU and Linux really are separate.
(Yes, I'm aware that it's the Distro that he usually insists includes the name "GNU/Linux", but to the uncalibrated ear, it often sounds like he's equating GNU and Linux.)
IBM/everyone did the same when they went from 30 pin to 72 pin SIMMS, then to DIMMS, then to DDR DIMMS.
The switch from 30-pin to 72-pin SIMMS, and then on to DIMMS, occured as a result of the memory bus width increasing.
When the normal width for a memory bank reached 64 bits, they went from 30-pin to 72-pin. (It was easier to insert one 72-pin SIMM than four 30-pin SIMMs.)
When the normal width for a memory bank reached 128 bits, they switched from SIMMs to DIMMS. (Again, it is easier to insert one DIMM than two 72-pin SIMMs. Imagine what would happen if every RAM upgrade meant you had to insert eight 30-pin SIMMs.)
However, the switch from SDRAM DIMMs to DDR SDRAM DIMMs was strictly for bandwidth, not for making memory insertion cheaper/easier. ("Cheaper" because one DIMM socket takes up much less board space than eight 30-pin SIMM sockets.)
SGI's Origin supercomputer's are essentially clusters of Itanium 2 boxes in a NUMA architecture...And they're still supporting them.
Who still does this?
I would. I'm running a Duron 750MHz, and I'd like to upgrade it to an Athlon 1.4GHz. (The only problem is finding the darn part.)
I can't wait for cheap, desktop-oriented 10k and 15krpm units.
Don't hold your breath. 10k drives generate a lot of heat.
No kidding. I went to the TechReport review because it was the first one that's not running slow, and my jaw dropped. AMD's really taking the lead with their recent offerings.
Well, they did run ten of these things simultaneously, each one working on a different chunk of the problem.
It probably would be...if you tried to crack the same ciphers that the Colossus Mk2 was intended to break.
I recall the stories on SlashDot about how the GPUs on modern ATI/nVidia cards are "many times faster than P4s"... well, yes, but you can't run Linux on them...
To elaborate:
GPUs still only run at a couple of hundred of MHz, but their dedicated circuitry allows them to perform certain matrix calculations much faster than x86 chips currently do, even with vector instruction extensions like MMX and SSE/SSE2.
Here are a couple of links to relevant articles. (1 2)
Speaking of affording it...My grandfather just finished chewing me out for adding RAM to his system. (Scanning photos was taking forever, with lots of swapping.)
Me, I considered it entertainment spending. I was improving the performance of their system, so they probably wouldn't need my help as much.
Unfortunately, it doesn't seem to help. Windows ME must put everything it possibly can in swap space.
Me, I try to use my journal to stimulate discussion. And I link to the individual journal entries in my sig, so people know about it.
My current journal is a little different than usual...it's about an idea for a multiplayer online D&D environment, except that it aids pen-and-paper-style gameplay instead of MMORPG-style gameplay.
My other recent journal entries have been about concepts like brother-sister twin NPC clerics (where one channels positive energy, and the other channels negative energy), using dreams to acquaint players with their characters' strengths, etc.
I get a lot of interesting feedback, averaging six comments per journal entry. (Since I started focusing the topics, anyway.)