More so if you're licensing the content than shouldn't they provide free replacement CDs? I mean you're not paying for the CD then it's part of the license arrangement:-)
That and I agree with the dozens of other posters. RIAA can say whatever they want, what is actually law (specially in other countries) is another story alltogether.
What a bullshit useless deal. What about mac users who are preintel?
I don't see what this benefits anyone. If anything it just makes more people hate Intel for even thinking of a ridiculous business proposal.
I mean what if Ford struck a deal with Shell and made all oil more expensive [or less useful] for competitor vehicles?
At what point do people just say "fuck you" and find alternatives. Vonage is one of them. There are other VoIP programs out there and this deal is likely to make them even more known then they would through the natural selection process of software.
If you want to get technical early monitors used composite signals which were not RGB. It's only the EGA and above [e.g. VGA] outputs which are RGB.:-)... sorry...
fdisk, install windows, mkfs, install linux on another partition
Leaving the default "Made by $CRAP" install is just a sign of a newb. There are ALWAYS things lingering and it's a huge PITA.
When I got my Dell laptop [which other than the shit software is nice] and powered it up for the first time I had no less than 3 or 4 "please register me" dialogs up on the screen and about 15 tray icons and god only knows how many services loaded.
I wiped that install clean faster then you can say "Dude you got a Dell!" and just dual-booted it. Fortunately the license key on the bottom of the laptop worked with my WinXP OEM cd I used for another box.
And for the record, NAV and McAfee are crap compared to the power of just not running windows insecurely.
This is pure bollocks. Any half-way competent programmer can start a thread or two and pass it information via globals and locks. I wrote in all of 5 minutes a dual-thread RSA accelerator for my cryptolibrary that effectively halved the latency of RSA private key operations.
In many "industrial" applications they're just different processes. E.g. an apache server hardly needs to spawn two threads to handle a single request. But having two threads for two requests make sense.
In desktop cases most of what you do or see can be threads or processes. Sound, graphics, gui, io, networking, etc... So an OS can make good use of multi-core/processor setups too.
Only an mutt thinks it's significantly hard to at least make some use of multi-execution designs. Now I agree making Optimal usage is hard but even then a programming language would hardly help you do something like stripped NUMA...
This sounds funny until you realize that companies like AMD [and Intel I imagine] have fallback plans for 3-core processors in case 1 core is dead on production.
So it's entirely possible when the next series of 6 or 8 core processors are made you can have in theory anywhere from 1 to 8 cores active. Chances are they won't allow chips with >1 defective core to be sold for quality reasons. But it's entirely possible that you could buy a 5-core processor by 2010:-)
I really hope these quad-cores involve 32 or 64-bit HT links to cope with the bandwidth problems. Only so far you can push HT links. Currently with multi processor setups it's a bunch better than Intel's offering [I know this because I work with a 4P dual-core Opteron setup at work...] but even two cores can push an HT link south.... Wants the 4P box for home... drool
Don't mistake the # of cores with their efficiency. An AMD64 core can execute upto 9 instructions per cycle (though really that's just theoretical) and you're looking at 3 at most in practice (do to decoder bandwidth).
Where the AMD64 multi-core shines by comparison to the Niagra chip is the per core cache and more advanced out of order execution engine. Niagra is a very task specific chip whereas AMD and Intel chips are more general purpose.
This whole "4 threads per core" issue is just bollocks anyways. If each core only has one ALU pipe then who cares how many threads you have. Specially since they share caches where you will hit a huge bottleneck.
Ah the american dream lives on! [even if you're not in the states]
The trick really at this point is
Small company: Huge risk, make sure you can street perform something people will pay for.
Midsized: Become barmates with the lads and fit in well.
Large [...where I work...]: Extract all you can, there may not be a tommorow. That or learn to stab people in the back so when cuts roll around you look like a greased wheel.
No one alive in 1983 old enough to remember printed quotes like that types "ROFLMAO"...
That said... I don't know why that would be funny in 1983. My vic-20 had 5KB of ram in 1987 and it was fine enough for a 5 yr old to play games and type basic programs. Why would it need a gigabyte of ram to store 500 bytes of basic [at most] or run that amazingly cool rat race game...
Point is they're stupid quotes and the anti-bill zealotry shows through. I hate the guy too but at least I don't pull things he might not even have said out of context in a way to bolster my own ego.
I hate that quote. First off, it's not even provably attributable to Gates in the first place. It's just some oft-quoted jibberish that gets repeated.
Second, it's vastly out of context. Supposing he did say it, in say what, 1983? Then yes, 640kb would be MORE than enough in 1983 for any home computer use given that the average program ran far less than 64KB (most aiming for tiny or small memory models).
So while I agree that people can't just say "this is the market" stop using useless retarded badly quoted lines to prop up an argument.
The dude is talking on behalf of MSFT, that alone makes him a retard, do you really need to say anthing else?
Not for this reason. They jumped the gun by locking themselves into intel only for x86. AMD is a viable contender (hint: TurionX2 is coming out within the year) and shouldn't be discounted specially for the sake of the customers who might be more willing to switch to Apple had they more choice.
There is little reason from a software point of view why other x86 processors are not in the mix. Sure it requires a different motherboard and bios (or whatever you want to call the startup sequence) but if that means you capture even more market share by getting people who don't like Intel offerings I think it would be worth it.
This isn't exactly correct. While externally AMD has a good lead on Intel with the integrated low latency memory controllers and dedicated HT links [e.g. think MP].
Internally however, AMD is behind. Intel has 2 cycle caches... at >3Ghz whereas AMD has 2 [and 3] cycle caches at 2Ghz. I'm talking L1 specifically... So internally Intel can move data much quicker than AMD could ever hope to.
However, AMD has more resources dedicated to execution. The problem with Intel is hardly to "get data to the core" it's "get the core to actually do some work!". There are a lot of pipeline hazards that AMD processors don't have and the AMD processors execute a lot of opcodes faster.
For instance, really trivial but comes up in crypto a lot, "adc" which is add with carry. It takes one cycle on the AMD cores and 1.5 on the intel. "mul" takes 2/6 [2/8 for 64-bit mode] cycles [the slash is throughput/latency] whereas Intels are 10-15.
Then there is the fact you have THREE ALU pipes. So if all is going well you can have an IPC upwards of 3 [the decoder can only issue three opcodes per cycle] but in theory you could retire more [think FPU]. While the IPC over an entire program is likely to be less than 2 certain hotspots could easily get over 2.
Bignum multipliers like those in TomsFastMath [iirc] routinely hit over 1.5 for IPC, I don't have the exact values handy but they're certainly pretty fast.
Intel cores, specially for ALU operations just lack IPC. So the clock may be 1.5x faster but the IPC can be 2x to 3x lower.
Laptop processors suck hard? Get yourself a Turion MT series. They run as fast as 2.2Ghz [though 2Ghz is more common to find] and are rated for 25W TDP which is basically what a PentiumM is.
Oh and the Turions are 64-bit processors. So you can run in a proper long mode.
That said, my last laptop was an Athlon-XP-M and even that was not so bad. It had a TDP of 45W but I could easily eek 4 hours out of it when editing source code. It usually ran 1150mAh in the lowest mode out of a 4.3Ah battery.
My dell with a Dothan 2Ghz runs about 950mAh out of a 7Ah battery. So really the Intel laptop gets a longer run not really because it takes less power but because the battery is more efficient [and physically larger].
The turions are even more efficient than the XP cores. I don't have the current rating yet [hard to figure that out in Windows] but I suspect that it's comparable. I get 6.5 hours with the spare battery in [and this lifebook is still very light]...
Anyways, the reason they went with Intel is because they want some kichy design that the zealots can get up all in a roar over. Both Intel and AMD execute the same ISA so from a users stand point there is no different. A MacOS binary for Intel would work on an AMD cpu.
It does make a difference for the OS/packaging folk but that just means they have to "earn their pay".
Apple would be wiser to not deal exclusively. If Intel is the better choice as they claim let the market determin that.
The lag in companies like Intel and AMD is actually about 5-10 years from initial design specs to in your local store.
There is "a lot" of testing that goes into them. So even if HP [or whatever] could get a team together today the earliest we would see a design would be 2011. By then they'd have to plan well ahead for like 45nm, at least 4 cores of OOE processors with at least 1MB of 10 cycle cache each running at at least 2Ghz etc....
And you'll be surprised to learn the newer cores are more of the same.
Where things will be improving are DDR2/DDR3, more cores [e.g. 4 core], more efficient process and other things along those lines.
The actual AMD64 core [excluding the x86 ISA] is actually fairly efficient. It's not the best but it's better than Intel's netburst and [so far] Pentium M line.
Would be cool to see a move away from x86 even if it meant simple O(n) pre-decoding [e.g. no decoding optimizations]. That would make the die simpler but raise the memory bandwidth.
Aside from that AMD could improve the L1 cache. It's currently a 2-way 64KB code/data cache. With say this they may be able to bump that up to say 128KB 4-way code/data. The trick there though is to keep latency to a minimum. 4-way may be too much.
You really ought to read up on the AMD64 design. There is little advantage to ignoring 32-bit opcodes. First off, "64-bit opcodes" are actually extensions of 32-bit ones. That is, 32-bit code can [and often is] smaller than 64-bit code [except when there are a lot of register spills].
Second, the hardware real estate is not that much. If the cpu only did 64-bit opcodes [hint: think something as simple as adding two char's] you'd have MORE overhead as you mask off bytes, words, dwords.
Aside from cache and the ALU decoder space is where there is waste.
Better designs would be to keep the ALU and drop the ISA [or at least re-arrange the opcode map to favour more common opcodes... ala huffman!].
The purpose of the GPL was to prevent lockin not to promote developers.
The idea was born out of the various incompatible UNIX'es which were all proprietary and therefore able to lock people in. Run a Sparc box? Get your OS from Sun. Don't like the OS? Too bad, you can't change it or replace it, etc...
The GPL3 being aggressive against DRM is a sane move I think. As a developer I have a right to license my software as "not for use in Japan" if I wanted. I could just as easily say "not for use with software that uses DRM".
And frankly I don't see anything negative about it. DRM is bullshit anyways, never will protect the producer and only violates the users ability to enjoy the product as they should be able to.
My PC doesn't cost 650$ [what the mini costs here]. It costs about 1200-1300$. It's also a full fledged desktop with an AMD X2 4200+, 1GB ofram [not 256M], a tv tuner, PS/2 and USB and parallel and DB9 serial ports.
I can easily open it up and replace a broken component or even upgrade it. I can add a "super drive" [we call them an LG multi-drive in the real world] for much less. My processor can multitask and work faster while not generating more heat [or taking more power].
Your macmini may be fine for your "I only browse the web and that's it, right? oh and email!" life. And frankly I don't see a problem with that. I just find that setup fairly lame for a person who does "work" with their computer.
And most Macs suffer the "NIH BAD!" syndrome which makes repairs and upgrades costly.
Slashdot Mirror
versions? no ... reasons not to use office!
... whatever fucking use Openoffice already. Microsoft is just satan incarnate.
"oh you mean this file won't open on your copy of office 2007? arrg!"
Or something like that
Tom
More so if you're licensing the content than shouldn't they provide free replacement CDs? I mean you're not paying for the CD then it's part of the license arrangement :-)
That and I agree with the dozens of other posters. RIAA can say whatever they want, what is actually law (specially in other countries) is another story alltogether.
Tom
Simple test: If the code works the answer is no. :-)
Cheapshot but I hate Microsoft so it's well worth it.
Tom
I use vonage...
What a bullshit useless deal. What about mac users who are preintel?
I don't see what this benefits anyone. If anything it just makes more people hate Intel for even thinking of a ridiculous business proposal.
I mean what if Ford struck a deal with Shell and made all oil more expensive [or less useful] for competitor vehicles?
At what point do people just say "fuck you" and find alternatives. Vonage is one of them. There are other VoIP programs out there and this deal is likely to make them even more known then they would through the natural selection process of software.
Tom
If you want to get technical early monitors used composite signals which were not RGB. It's only the EGA and above [e.g. VGA] outputs which are RGB. :-) ... sorry ...
Tom
That's a Romero quote if I ever saw one...
/mostly a joke on fark
//don't have fark account
///not cheap just lazy
////slashy
Tom
fdisk, mkfs, install Linux distro.
Stop screwing with stupidity.
Or at least
fdisk, install windows, mkfs, install linux on another partition
Leaving the default "Made by $CRAP" install is just a sign of a newb. There are ALWAYS things lingering and it's a huge PITA.
When I got my Dell laptop [which other than the shit software is nice] and powered it up for the first time I had no less than 3 or 4 "please register me" dialogs up on the screen and about 15 tray icons and god only knows how many services loaded.
I wiped that install clean faster then you can say "Dude you got a Dell!" and just dual-booted it. Fortunately the license key on the bottom of the laptop worked with my WinXP OEM cd I used for another box.
And for the record, NAV and McAfee are crap compared to the power of just not running windows insecurely.
Tom
This is pure bollocks. Any half-way competent programmer can start a thread or two and pass it information via globals and locks. I wrote in all of 5 minutes a dual-thread RSA accelerator for my cryptolibrary that effectively halved the latency of RSA private key operations.
In many "industrial" applications they're just different processes. E.g. an apache server hardly needs to spawn two threads to handle a single request. But having two threads for two requests make sense.
In desktop cases most of what you do or see can be threads or processes. Sound, graphics, gui, io, networking, etc... So an OS can make good use of multi-core/processor setups too.
Only an mutt thinks it's significantly hard to at least make some use of multi-execution designs. Now I agree making Optimal usage is hard but even then a programming language would hardly help you do something like stripped NUMA...
Tom
This sounds funny until you realize that companies like AMD [and Intel I imagine] have fallback plans for 3-core processors in case 1 core is dead on production.
:-)
... Wants the 4P box for home ... drool
So it's entirely possible when the next series of 6 or 8 core processors are made you can have in theory anywhere from 1 to 8 cores active. Chances are they won't allow chips with >1 defective core to be sold for quality reasons. But it's entirely possible that you could buy a 5-core processor by 2010
I really hope these quad-cores involve 32 or 64-bit HT links to cope with the bandwidth problems. Only so far you can push HT links. Currently with multi processor setups it's a bunch better than Intel's offering [I know this because I work with a 4P dual-core Opteron setup at work...] but even two cores can push an HT link south.
Tom
Don't mistake the # of cores with their efficiency. An AMD64 core can execute upto 9 instructions per cycle (though really that's just theoretical) and you're looking at 3 at most in practice (do to decoder bandwidth).
Where the AMD64 multi-core shines by comparison to the Niagra chip is the per core cache and more advanced out of order execution engine. Niagra is a very task specific chip whereas AMD and Intel chips are more general purpose.
This whole "4 threads per core" issue is just bollocks anyways. If each core only has one ALU pipe then who cares how many threads you have. Specially since they share caches where you will hit a huge bottleneck.
Tom
Ah the american dream lives on! [even if you're not in the states]
The trick really at this point is
Small company: Huge risk, make sure you can street perform something people will pay for.
Midsized: Become barmates with the lads and fit in well.
Large [...where I work...]: Extract all you can, there may not be a tommorow. That or learn to stab people in the back so when cuts roll around you look like a greased wheel.
Tom
No one alive in 1983 old enough to remember printed quotes like that types "ROFLMAO" ...
... I don't know why that would be funny in 1983. My vic-20 had 5KB of ram in 1987 and it was fine enough for a 5 yr old to play games and type basic programs. Why would it need a gigabyte of ram to store 500 bytes of basic [at most] or run that amazingly cool rat race game ...
That said
Point is they're stupid quotes and the anti-bill zealotry shows through. I hate the guy too but at least I don't pull things he might not even have said out of context in a way to bolster my own ego.
Tom
....640kb quote ....
I hate that quote. First off, it's not even provably attributable to Gates in the first place. It's just some oft-quoted jibberish that gets repeated.
Second, it's vastly out of context. Supposing he did say it, in say what, 1983? Then yes, 640kb would be MORE than enough in 1983 for any home computer use given that the average program ran far less than 64KB (most aiming for tiny or small memory models).
So while I agree that people can't just say "this is the market" stop using useless retarded badly quoted lines to prop up an argument.
The dude is talking on behalf of MSFT, that alone makes him a retard, do you really need to say anthing else?
No.
Have a good day sir.
Tom
Not for this reason. They jumped the gun by locking themselves into intel only for x86. AMD is a viable contender (hint: TurionX2 is coming out within the year) and shouldn't be discounted specially for the sake of the customers who might be more willing to switch to Apple had they more choice.
There is little reason from a software point of view why other x86 processors are not in the mix. Sure it requires a different motherboard and bios (or whatever you want to call the startup sequence) but if that means you capture even more market share by getting people who don't like Intel offerings I think it would be worth it.
Tom
It's RSS... not something more important like DNS or whatever.
Who cares if they have their own spin on it. People with compliant RSS readers will be able to see other compliant feeds.
It isn't like Apple is the ONLY source of RSS feeds.
I dunno, I've been out of the "hip tech" for a while, is Apple the only place to get an RSS feed?
Tom
This isn't exactly correct. While externally AMD has a good lead on Intel with the integrated low latency memory controllers and dedicated HT links [e.g. think MP].
... at >3Ghz whereas AMD has 2 [and 3] cycle caches at 2Ghz. I'm talking L1 specifically... So internally Intel can move data much quicker than AMD could ever hope to.
Internally however, AMD is behind. Intel has 2 cycle caches
However, AMD has more resources dedicated to execution. The problem with Intel is hardly to "get data to the core" it's "get the core to actually do some work!". There are a lot of pipeline hazards that AMD processors don't have and the AMD processors execute a lot of opcodes faster.
For instance, really trivial but comes up in crypto a lot, "adc" which is add with carry. It takes one cycle on the AMD cores and 1.5 on the intel. "mul" takes 2/6 [2/8 for 64-bit mode] cycles [the slash is throughput/latency] whereas Intels are 10-15.
Then there is the fact you have THREE ALU pipes. So if all is going well you can have an IPC upwards of 3 [the decoder can only issue three opcodes per cycle] but in theory you could retire more [think FPU]. While the IPC over an entire program is likely to be less than 2 certain hotspots could easily get over 2.
Bignum multipliers like those in TomsFastMath [iirc] routinely hit over 1.5 for IPC, I don't have the exact values handy but they're certainly pretty fast.
Intel cores, specially for ALU operations just lack IPC. So the clock may be 1.5x faster but the IPC can be 2x to 3x lower.
Tom
Laptop processors suck hard? Get yourself a Turion MT series. They run as fast as 2.2Ghz [though 2Ghz is more common to find] and are rated for 25W TDP which is basically what a PentiumM is.
Oh and the Turions are 64-bit processors. So you can run in a proper long mode.
That said, my last laptop was an Athlon-XP-M and even that was not so bad. It had a TDP of 45W but I could easily eek 4 hours out of it when editing source code. It usually ran 1150mAh in the lowest mode out of a 4.3Ah battery.
My dell with a Dothan 2Ghz runs about 950mAh out of a 7Ah battery. So really the Intel laptop gets a longer run not really because it takes less power but because the battery is more efficient [and physically larger].
The turions are even more efficient than the XP cores. I don't have the current rating yet [hard to figure that out in Windows] but I suspect that it's comparable. I get 6.5 hours with the spare battery in [and this lifebook is still very light]...
Anyways, the reason they went with Intel is because they want some kichy design that the zealots can get up all in a roar over. Both Intel and AMD execute the same ISA so from a users stand point there is no different. A MacOS binary for Intel would work on an AMD cpu.
It does make a difference for the OS/packaging folk but that just means they have to "earn their pay".
Apple would be wiser to not deal exclusively. If Intel is the better choice as they claim let the market determin that.
Tom
The lag in companies like Intel and AMD is actually about 5-10 years from initial design specs to in your local store.
There is "a lot" of testing that goes into them. So even if HP [or whatever] could get a team together today the earliest we would see a design would be 2011. By then they'd have to plan well ahead for like 45nm, at least 4 cores of OOE processors with at least 1MB of 10 cycle cache each running at at least 2Ghz etc....
Tom
And you'll be surprised to learn the newer cores are more of the same.
:-)
Where things will be improving are DDR2/DDR3, more cores [e.g. 4 core], more efficient process and other things along those lines.
The actual AMD64 core [excluding the x86 ISA] is actually fairly efficient. It's not the best but it's better than Intel's netburst and [so far] Pentium M line.
Would be cool to see a move away from x86 even if it meant simple O(n) pre-decoding [e.g. no decoding optimizations]. That would make the die simpler but raise the memory bandwidth.
Aside from that AMD could improve the L1 cache. It's currently a 2-way 64KB code/data cache. With say this they may be able to bump that up to say 128KB 4-way code/data. The trick there though is to keep latency to a minimum. 4-way may be too much.
But even 128KB of 2-way would be nicer.
Or even say 2MB L2s
Tom
You really ought to read up on the AMD64 design. There is little advantage to ignoring 32-bit opcodes. First off, "64-bit opcodes" are actually extensions of 32-bit ones. That is, 32-bit code can [and often is] smaller than 64-bit code [except when there are a lot of register spills].
... ala huffman!].
Second, the hardware real estate is not that much. If the cpu only did 64-bit opcodes [hint: think something as simple as adding two char's] you'd have MORE overhead as you mask off bytes, words, dwords.
Aside from cache and the ALU decoder space is where there is waste.
Better designs would be to keep the ALU and drop the ISA [or at least re-arrange the opcode map to favour more common opcodes
Tom
The purpose of the GPL was to prevent lockin not to promote developers.
The idea was born out of the various incompatible UNIX'es which were all proprietary and therefore able to lock people in. Run a Sparc box? Get your OS from Sun. Don't like the OS? Too bad, you can't change it or replace it, etc...
The GPL3 being aggressive against DRM is a sane move I think. As a developer I have a right to license my software as "not for use in Japan" if I wanted. I could just as easily say "not for use with software that uses DRM".
And frankly I don't see anything negative about it. DRM is bullshit anyways, never will protect the producer and only violates the users ability to enjoy the product as they should be able to.
Tom
That's information I could use BEFORE I made my mac mini unbootable.
Fortunately I bought it at a slave-wage best-buy and I got a full refund. Woohoo!
Tom
Yeah, except OF didn't work on my Mac Mini [at least the the way I found on the web].
May I take this time to mention there was
NO FUCKING MANUAL
With the mac mini?
At least my PC tells me "hit F2 to go into setup" [usually in engrish].
Tom
My PC doesn't cost 650$ [what the mini costs here]. It costs about 1200-1300$. It's also a full fledged desktop with an AMD X2 4200+, 1GB ofram [not 256M], a tv tuner, PS/2 and USB and parallel and DB9 serial ports.
I can easily open it up and replace a broken component or even upgrade it. I can add a "super drive" [we call them an LG multi-drive in the real world] for much less. My processor can multitask and work faster while not generating more heat [or taking more power].
Your macmini may be fine for your "I only browse the web and that's it, right? oh and email!" life. And frankly I don't see a problem with that. I just find that setup fairly lame for a person who does "work" with their computer.
And most Macs suffer the "NIH BAD!" syndrome which makes repairs and upgrades costly.
Enjoy your Mac.
Tom
I haven't gone anywhere. Maybe you're just too busy fucking your sister to realize life goes on like normal for the rest of us.
Tom