I love sci fi books, I'm glad someone is doing this. I have a palm m500 (all other palms suck for pure book reading) set up specifically to read books.
your link doesn't work, try this one: http://rr.informatik.tu-freiberg.de/tmplt/controld esk.controldesk.php?lang=eng&js=1 I tried it and its quite cool. 4 trains and 10 destinations, and you can do an intermedia destination. About 4 live cam views. And best of all it actually works since its not slashdotted:) It has a counter too of the # of ppl using it. A little confusing at first. Click the help (the small pics of trains) for a pop up that describes the proces and shows where to click.
Except its talking about enabling or disabling MMX or SSE type instructions depending on if its intel or amd. If its intel AND pentium 4, they are turned on, otherwise they are not (even pentium 3 has them turned off). Despite the fact that there's a specific register to check to see if the CPU supports MMX and SSE etc. Since they disable them for pentium 3 and earlier and AMD any type, that means they are doing it for performance reasons, to make the P4 look good. Clock for clock the P3 is faster than the P4.
Has nothing to do with scheduling. And thats more of an internal CPU thing anyway.
Many many people are missing the point and thinking the CPU is a magic black box that you can "optimize" for intel, or that this doesn't detect intel only CPU, etc.
Let me make an analogy to explain this a bit.
Its like if you had a motherboard from nvidia, say an nforce 4 motherboard. What if it looked for the presense of an nvidia graphics card, and if so, enabled AGP 8x, fast writes, etc, but if it found an ATI graphics card it only enabled AGP 4x etc, disabled fast writes. REGARDLESS of if the graphics card supported the agp 8x or not. Its ATI's responsibility with thier card to ensure it supports AGP 8x properly if they list it on their box. Its not the motherboard manufacture's job to arbitrarily enable and disable it based on brand alone. Everyone can see where that would be wrong and unfair to the competition (ATI).
RTFA, and some of the comments.
The compiler SPECIFICALLY looks for GenuineIntel as a cpuid. If it doesnt' find it, it doesn't use mmx or sse or sse2 or sse3. Even though many processors support it. And there's a specific register to check to see if the CPU in question supports those extensions.
What it SHOULD do is test to see if the CPU can support those extensions and then use them if supported. As mentioned many times in other comments, you can fool the compiler into thinking everything is an intel and then the stuff runs faster on AMD.
Too many people making similar comments about "optimizing for intel" are missing the issue and have an incorrect perception about processors.
An Intel or AMD cpu is not a magic black box. The way to use processors is by assembly instructions. There are instruction set extensions known as MMX and SSE and SSE2 and SSE3. These are extra instructions that help accomplish things faster that would have taken more instructions to do without them.
A compiler converts an english readable language like C++ to assembly.
Ideally the complier will throw in those extra instructions wherever it can to reduce the number of instructions overall and reduce the time it takes to execute.
Its AMD's responsibility to ensure that standard MMX etc instructions execute fine, fast, and without error. Its the compilers responsibility to ensure that it puts in MMX instructions wherever possible.
The argument is that the compiler inserts something into the.exe that checks for a intel cpu (NOT the ability to run mmx code) and wont run the mmx etc code on anything but intel cpu.
Its like if you had a motherboard from nvidia, say an nforce 4 motherboard. What if it looked for the presense of an nvidia graphics card, and if so, enabled AGP 8x, fast writes, etc, but if it found an ATI grpahics card it only enabled AGP 4x etc. REGARDLESS of if the graphics card supported the agp 8x or not.
Everyone can see where that would be wrong and unfair to the competition (ATI). But for some reason with CPU's its like its a magic black box and ppl cant understand wtf goes on inside it.
I think he's referring to the fact that the memory address width in amd64/opteron is I think 40 bits or 48 bits or some such that only allows something like 64 gigs of RAM instead of 1 TB or whatever is allowed in full 64 bit addressing.
I dont know the exact numbers, I just know the generalities which is that it doesn't have a full 64 bit addressing space for memory.
That being said, does it matter? How many people have more than 4 or 8 gigs of RAM anyway? lol
hahaahahahahah.
$500 wouldn't stop anyone. Companies would figure it was price of doing business.
Make it $50,000 and it might cause companies to pause a bit.
Actuallly it IS that simple.
From the PDF claim form:
1. Model Number of 75GXP you purchased (e.g., DTLA 307-015, DTLA 307-075):
____________________________
2. Approximate Date you purchased the 75GXP (Please provide at least the month and year of
purchase): Month: ____________________ Day: ________ Year:___________
3. If you purchased the 75GXP as a stand-alone component from an authorized seller, please
provide either:
(a) Serial Number of the 75GXP: _________________________, or
(b) Documentation demonstrating your purchase of the 75GXP from an authorized seller
(* attach the documentation to the end of this form).
4. If you purchased the 75GXP as a pre-installed component in a computer system, please
provide the Serial Number of the 75GXP: _______________________________.
Notice that if you bought it standalone, you need EITHER the serial OR the receipt. That means serial is enough. 5 Year old receipts not nessasary. If you bought it already installed you only need serial.
The problem would be that if you bought from ebay odds are the $100 for that serial was already claimed. Unless you got some ppl that were clueless and selling them on ebay and didn't know about this.
And buying on ebay specifically to get the money, I dunno, IBM might be able to argue fraud on that and not send anyway.
OMG! Yes I hated that! I stopped reading the entire series of books that happened after that point when they killed him off. I have like 4 unread books (some hardback) starwars ones that I haven't read because I hated that so much.
I didn't realize that there was debates here on/. about it. Linkage?
I took this test when they were doing a pilot test at my college (they did the same at several colleges) a couple months ago. My college offered a $25 amazon gift card and a chance to win ipod. It took a couple hours to go through the whole thing, and the interface was kinda klunky. Plus at the end when I tried to fill out the comment part about the test it crashed the browser so I couldn't send my comments in lol.
What was really cool was that they ended up sending out TWO amazon gift certificates (I think they accidently sent out an extra to everyone, perhaps some ppl complained because they entered wrong email to send the cards to or something) to me for $25. So I got paid $50 for a couple hours hehehe.
But the problem is the phone companies would never go for it. They rely upon most people doing email and stuff and not using much of the bandwidth. If everyone used thier maximum, they would end up raising prices.
Its kinda like rebates. They offer them because most people never send them in.
From what I understand the 6mbit at 3mbit advertised was because they didn't have the 3mbit profiles in their system yet and marketing went ahead and offered the packages.
The people who got 6mbit will be able to keep it.
I'm sure they will be offering 6mbit again (not sure when, or for what price). Best place to get details on this stuff is dslreports.com forums. Official SBC techs post there.
Not true. There's been criminals awarded tons of money from lawsuits because they got hurt or shot breaking in. There's some lady in jail because her dog held onto the leg of a burgler until cops got there. I say if someone breaks in, make sure to shoot them dead. Of course then you get families suing you.
And yes, cases like this HAVE been successfully done by the criminals and families. Try google.
I checked the firmware site briefly. There were a lot of tools on the download page to do MD5 signature hacks. Perhaps this gets around that? I dont know I'm just guessing, perhaps someone with more knowledge can say.
Seems like most of you dont even click on the other links in the story:) The link "uncappers" leads to an older slashdot story about FBI raiding uncappers. With today's paranoia you could probably have secret warrant too.
Plus, who doesn't have at least one illegal MP3 or movie or game or application? Once they find that they could expand scope of warrant and get you for that even if they dont get you for uncapping.
I think the dangers of uncapping far outweigh the benifits, especially as DSL companies like SBC (and people who resell thier lines like sonic and dslextreme) are currently offering 6.0/607 lines for $45ish a month.
Repeat of the link in the story:
http://slashdot.org/article.pl?sid=02/11/22/013226 &tid=123
Guess no one has read the article about the master/slave thing in LA county. They mentioned that they were calling it "Primary" and "Secondary" instead of master/slave. Pretty stupid if you ask me, whats the drive connected on the slave on secondary IDE channel? Secondary Secondary? How about Primary Primary?
Looks like most of the speculation about it just being one guy who way overreacted are true according to the article. Heh IBM even said "we'll look into it" about the terminology.
Yes but they were suing them for SHARING not for d/ling. All the lawsuits by the RIAA so far have been for sharing.
BTW, did you catch the article I submitted about RIAA trying to get exemption for getting sued?
Thats not the point, read closer. The point is RIAA got an IP (supposedly hers) from someone sharing files on KAZAA. Thats it. Mac doesn't have kazaa. Therefore she could not have possibly done what they said. More likely someone screwed up somewhere with the IP process, or RIAA gave wrong tiems or something, they just didn't want to admit mistake.
ALL of the court cases from RIAA have been from file sharing p2p services, and I think only from kazaa networks so far.
Slashdot Mirror
If I had a ps2 I might. Too bad I dont. Anyone want to send me one? :)
I love sci fi books, I'm glad someone is doing this. I have a palm m500 (all other palms suck for pure book reading) set up specifically to read books.
/.ed it took about 5 mins to load here so I'm going to paste the direct bit torrent links that are on his site:t f.zip.torrent t ml.zip.torrent
Anyway I think the site is
RTF text format:
http://files.machinima.com/torrents/accelerando-r
HTML:
http://files.machinima.com/torrents/accelerando-h
He has like almost a dozen formats (pdf, palm doc, ascii, etc etc), so check out his page for more.
This was MEDICAL studies in peer reviewed journals. As such, this is a pretty high percentage.
Now think of all the other studies that are funded by someone with an axe to grind, like those MS funded studies that say linux sux.
If 1/3 of peer reviewed medical studies are innaccurate, I'm betting a MINIMUM of 50% of all other studies are the same.
Yeah thats the one that got sued in california I think.
your link doesn't work, try this one:d esk.controldesk.php?lang=eng&js=1 :) It has a counter too of the # of ppl using it.
http://rr.informatik.tu-freiberg.de/tmplt/control
I tried it and its quite cool. 4 trains and 10 destinations, and you can do an intermedia destination. About 4 live cam views. And best of all it actually works since its not slashdotted
A little confusing at first. Click the help (the small pics of trains) for a pop up that describes the proces and shows where to click.
Except its talking about enabling or disabling MMX or SSE type instructions depending on if its intel or amd. If its intel AND pentium 4, they are turned on, otherwise they are not (even pentium 3 has them turned off). Despite the fact that there's a specific register to check to see if the CPU supports MMX and SSE etc.
Since they disable them for pentium 3 and earlier and AMD any type, that means they are doing it for performance reasons, to make the P4 look good. Clock for clock the P3 is faster than the P4.
Has nothing to do with scheduling. And thats more of an internal CPU thing anyway.
Let me make an analogy to explain this a bit. Its like if you had a motherboard from nvidia, say an nforce 4 motherboard. What if it looked for the presense of an nvidia graphics card, and if so, enabled AGP 8x, fast writes, etc, but if it found an ATI graphics card it only enabled AGP 4x etc, disabled fast writes. REGARDLESS of if the graphics card supported the agp 8x or not. Its ATI's responsibility with thier card to ensure it supports AGP 8x properly if they list it on their box. Its not the motherboard manufacture's job to arbitrarily enable and disable it based on brand alone. Everyone can see where that would be wrong and unfair to the competition (ATI).
RTFA, and some of the comments. The compiler SPECIFICALLY looks for GenuineIntel as a cpuid. If it doesnt' find it, it doesn't use mmx or sse or sse2 or sse3. Even though many processors support it. And there's a specific register to check to see if the CPU in question supports those extensions. What it SHOULD do is test to see if the CPU can support those extensions and then use them if supported. As mentioned many times in other comments, you can fool the compiler into thinking everything is an intel and then the stuff runs faster on AMD. Too many people making similar comments about "optimizing for intel" are missing the issue and have an incorrect perception about processors. An Intel or AMD cpu is not a magic black box. The way to use processors is by assembly instructions. There are instruction set extensions known as MMX and SSE and SSE2 and SSE3. These are extra instructions that help accomplish things faster that would have taken more instructions to do without them. A compiler converts an english readable language like C++ to assembly. Ideally the complier will throw in those extra instructions wherever it can to reduce the number of instructions overall and reduce the time it takes to execute. Its AMD's responsibility to ensure that standard MMX etc instructions execute fine, fast, and without error. Its the compilers responsibility to ensure that it puts in MMX instructions wherever possible. The argument is that the compiler inserts something into the .exe that checks for a intel cpu (NOT the ability to run mmx code) and wont run the mmx etc code on anything but intel cpu.
Its like if you had a motherboard from nvidia, say an nforce 4 motherboard. What if it looked for the presense of an nvidia graphics card, and if so, enabled AGP 8x, fast writes, etc, but if it found an ATI grpahics card it only enabled AGP 4x etc. REGARDLESS of if the graphics card supported the agp 8x or not.
Everyone can see where that would be wrong and unfair to the competition (ATI). But for some reason with CPU's its like its a magic black box and ppl cant understand wtf goes on inside it.
I think he's referring to the fact that the memory address width in amd64/opteron is I think 40 bits or 48 bits or some such that only allows something like 64 gigs of RAM instead of 1 TB or whatever is allowed in full 64 bit addressing. I dont know the exact numbers, I just know the generalities which is that it doesn't have a full 64 bit addressing space for memory. That being said, does it matter? How many people have more than 4 or 8 gigs of RAM anyway? lol
hahaahahahahah. $500 wouldn't stop anyone. Companies would figure it was price of doing business. Make it $50,000 and it might cause companies to pause a bit.
Actuallly it IS that simple. From the PDF claim form: 1. Model Number of 75GXP you purchased (e.g., DTLA 307-015, DTLA 307-075): ____________________________ 2. Approximate Date you purchased the 75GXP (Please provide at least the month and year of purchase): Month: ____________________ Day: ________ Year:___________ 3. If you purchased the 75GXP as a stand-alone component from an authorized seller, please provide either: (a) Serial Number of the 75GXP: _________________________, or (b) Documentation demonstrating your purchase of the 75GXP from an authorized seller (* attach the documentation to the end of this form). 4. If you purchased the 75GXP as a pre-installed component in a computer system, please provide the Serial Number of the 75GXP: _______________________________. Notice that if you bought it standalone, you need EITHER the serial OR the receipt. That means serial is enough. 5 Year old receipts not nessasary. If you bought it already installed you only need serial. The problem would be that if you bought from ebay odds are the $100 for that serial was already claimed. Unless you got some ppl that were clueless and selling them on ebay and didn't know about this. And buying on ebay specifically to get the money, I dunno, IBM might be able to argue fraud on that and not send anyway.
OMG! Yes I hated that! I stopped reading the entire series of books that happened after that point when they killed him off. I have like 4 unread books (some hardback) starwars ones that I haven't read because I hated that so much. I didn't realize that there was debates here on /. about it. Linkage?
They say you are supposed to run it on durons, celerons, VIA cpus.
I took this test when they were doing a pilot test at my college (they did the same at several colleges) a couple months ago. My college offered a $25 amazon gift card and a chance to win ipod. It took a couple hours to go through the whole thing, and the interface was kinda klunky. Plus at the end when I tried to fill out the comment part about the test it crashed the browser so I couldn't send my comments in lol. What was really cool was that they ended up sending out TWO amazon gift certificates (I think they accidently sent out an extra to everyone, perhaps some ppl complained because they entered wrong email to send the cards to or something) to me for $25. So I got paid $50 for a couple hours hehehe.
If that was so, why did the other commisioners write the dissent the way they did? That quote was from THEM.
But the problem is the phone companies would never go for it. They rely upon most people doing email and stuff and not using much of the bandwidth. If everyone used thier maximum, they would end up raising prices. Its kinda like rebates. They offer them because most people never send them in.
From what I understand the 6mbit at 3mbit advertised was because they didn't have the 3mbit profiles in their system yet and marketing went ahead and offered the packages. The people who got 6mbit will be able to keep it. I'm sure they will be offering 6mbit again (not sure when, or for what price). Best place to get details on this stuff is dslreports.com forums. Official SBC techs post there.
Not true. There's been criminals awarded tons of money from lawsuits because they got hurt or shot breaking in. There's some lady in jail because her dog held onto the leg of a burgler until cops got there. I say if someone breaks in, make sure to shoot them dead. Of course then you get families suing you. And yes, cases like this HAVE been successfully done by the criminals and families. Try google.
Yes, but what about the CHICKEN? What happened to it after two weeks? Ants? Mold? Smell?
I checked the firmware site briefly. There were a lot of tools on the download page to do MD5 signature hacks. Perhaps this gets around that? I dont know I'm just guessing, perhaps someone with more knowledge can say.
You must have missed the story yesterday about the Linksys Wireless G switch/router :)
Seems like most of you dont even click on the other links in the story :) The link "uncappers" leads to an older slashdot story about FBI raiding uncappers. With today's paranoia you could probably have secret warrant too.
Plus, who doesn't have at least one illegal MP3 or movie or game or application? Once they find that they could expand scope of warrant and get you for that even if they dont get you for uncapping.
I think the dangers of uncapping far outweigh the benifits, especially as DSL companies like SBC (and people who resell thier lines like sonic and dslextreme) are currently offering 6.0/607 lines for $45ish a month.
Repeat of the link in the story:
http://slashdot.org/article.pl?sid=02/11/22/013226 &tid=123
Guess no one has read the article about the master/slave thing in LA county. They mentioned that they were calling it "Primary" and "Secondary" instead of master/slave. Pretty stupid if you ask me, whats the drive connected on the slave on secondary IDE channel? Secondary Secondary? How about Primary Primary? Looks like most of the speculation about it just being one guy who way overreacted are true according to the article. Heh IBM even said "we'll look into it" about the terminology.
Yes but they were suing them for SHARING not for d/ling. All the lawsuits by the RIAA so far have been for sharing. BTW, did you catch the article I submitted about RIAA trying to get exemption for getting sued?
Thats not the point, read closer. The point is RIAA got an IP (supposedly hers) from someone sharing files on KAZAA. Thats it. Mac doesn't have kazaa. Therefore she could not have possibly done what they said. More likely someone screwed up somewhere with the IP process, or RIAA gave wrong tiems or something, they just didn't want to admit mistake. ALL of the court cases from RIAA have been from file sharing p2p services, and I think only from kazaa networks so far.