They did have shows on how this was done, back about 1977, all the magic of a Gould computer cracking this stuff out. Pretty impressive, then, for us people contending with IBM 360's and PDP 11's
Why a Gould computer? Surely the Asgard ones have more horsepowerr?
That made me laugh out loud. I pictured something like this -
manufacturing guy - 221mm2 is a big ass chip dude, almost twice the size of a P4 (112mm2 at 90nm). KK -...mystic stuff about the power of two... MG - well 4's also a power of two.
Mind you, people have pointed out the Emotion Engine in the PS2 was about the same size and Sony made money selling that.
Still, it's interesting the manufacturing people obviously thought the chip was too big to be manufactured cheaply.
Just to give you an example: mplayer is technically cool. But its complexity scares people away. It's only cool because it's free. You won't be able to sell it to anybody, because as a software _product_ it sucks. badly. Even with gmplayer.
Hey, but the mplayer team are working on this I'm sure.
2. Data transfer rate. In principle, the thousdands of different tips can all return data at the same time, compared to, say, 4 bits returned at once from a 4-platter HDD. Of course, in real situations, not all 4000 bits will necessarily be of interest, but I think with smart caching and device layout the throughput should be very high (i.e.: contiguous bits in a file are spread out so that the entire file is read by the 4000 tips without anything moving).
4096 bits is a sector isn't it? That's the minimum unit for a read on an IDE disk.
Actually I'd guess most OS's do most IO in response to a page fault, so it's probably multiples of the page size, which is n*4Kbytes or n*32Kbits, so you could actually have a larger read size without much of a penalty.
6. Cost. By using established MEMS technology (i.e.: the same lithography used to make microchips nowadays) I don't think implementation costs (and future scaling) will be too expensive (as compared to some more far-fetched nanotech ideas).
I think the coolest thing is that it's chip-like and gives you more density than flash, and it looks like it will be cheap too.
Flash was always supposed to take over from magnetic disks as far as I can see, but in retrospect that was completely bogus. If you look at the cost per bit, it was too high, and relying on Moore's law to reduce that wouldn't work, since magnetic media has it's own version of Moores where the time to double capacity was actually less than for silicon.
In general, I think any completely solid state technology will be tend to lag behind something with mechanical addressing because making large chips tends to be prohibitively expensive. But this manages to use chip technology and add in some mechanical stuff, which is probably the way to go.
Hope the products based on it are cheap and reliable - they'd be great in an MP3 player. Incidentally, all this talk of nanometre positioning makes me wonder how sensitive it is to vibration - the main problem I have with hard disks in MP3 players/phones etc is that they aren't very robust compared to flash, at least when they are spinning. This technology sounds like it would be worse given the smaller tolerances involved.
Oh and one other worrying thing - they don't say how standard the process used to make it is. I don't know whether it uses some exotic process or whether you could make it in a standard fab plant with some tweaks.
Public dental care is farce because the dentists aren't under enought incentive to treat you.
I'd guess in the States having your kids teeth straightened costs serious cash. On the other hand if all the other kids have it done, not having it done would put them at a disadvantage. People will spend serious money on their kids so it all works out.
Last time I was in the UK I spent ages phoning public dentists all of whom seem to have amazingly unfriendly receptionists / office hours.
Oddly enough, I broke both arms pretty badly on another visit to the UK, and the initial treatment was very good. Waking up in a filthy ward next to people who'd caught drug resistant bugs in the hospital sucked though.
I think for emergency stuff public provision is probably OK, but for the preventative or elective things it just isn't the solution - that needs to be done by private companies for a profit.
Yeah, and if you can't do with a built in function, you can always hack together some VBScript. And if you can't do it in VBScript you can write the C++ and cut'n'paste it into an ATL based ActiveX component and call it from VBScript. And you can email it to non technical people and it will just work, because they _all_ have Excel.
For some reason that last paragraph reminds me of Star Wars -
If you only knew the POWER of the DARK SIDE!!!11!!
Hell, if we look at the fact that strlcat/cpy have been turned down for inclusion multiple times to the GNU libc because it would be "slower" when preventing a buffer vuln, if anything it's getting worse, and will continue down that slope.
In fact the paper they reference on strlcat has this great example of strlcat/cpy vs strncat/cpy
Orginal unsafe code. The path string can be overrun if it is too small to hold the concatenated strings. If it is just before the return address on the stack, you have a possible exploit.
strcpy(path, homedir); strcat(path, "/"); strcat(path, ".foorc"); len = strlen(path);
with strncat / cpy. The code is safe but it's hard to read and thus to check it is safe. Also the calls to strlen require the string size be measured by counting the characters. Note that both strncat and strlcat will need to do this again when working out where to write the string to be concatenated. strncpy also wastes time zero filling the string, which strlcpy doesn't need to do.
It's as easy to read as the first code, and probably quicker than the second one, since it skips the excess calls to strlen and the zero filling in strncpy. Of course there's a cost to checking for buffer overflows as you copy, so it's slower than the first one.
Yeah, I read about that, it's a cool hack, in the most positive sense of using something in a way that it was not designed for. Intel failed to spot this extra use for the AGP GART when they cloned the chip I suppose.
It's an interesting question though. AMD got to design the chip and the architecture at the same time. Intel had to retrofit AMD's 64 bit stuff to the P4. There are all kinds of reasons why this would be hard - the P4 had a dual speed ALU which needed to be widened to 64 bit for instance.
I wouldn't be surprised at all if the resulting chip had some performance issues, but I haven't seen any benchmarks as to comparative 64 bit performance though. I don't particular like P4's even for 32 bit stuff - it looks like they ultrapipelined the CPU to get higher clock frequencies in a way that reduces performance compared to a similar priced AMD part for instance, whereas AMD seems to have worried about real benchmarks like Spec and ( Doom3:-) )
The Cell is not going to be used in general purpose machines. It has a PowerPC core, but the performance is low compared to a G5. It also has a bunch or SPEs but these are not suitable for general purpose computation either - they lack MMUs for example, and only have access to a small amount of tightly coupled SRAM. The idea is that you split your application into a software pipeline, and implement the stages in different SPEs.
All this stuff doesn't matter for games consoles where you can code all the crucial bits in SPE assembler and the whole machine runs only one application, but it would kill performance on a general purpose OS.
Well, ok, the flu is a virus, but honestly I couldn't come up with an example where a foreign bacteria nearly wiped out a civilization
There is one actually - the development of photosynthesis by cyanobacteria and the resulting oxygenating the atmosphere 2B years was presumably not good news for anerobic bacteria when it happened.
Not saying I can see this happening with Martian bugs, if Mars had life like this we'd detect it easily.
Wouldn't Martian bugs (assuming that they don't share a common ancestor) be different enough to be orthogonal to life on earth.
E.g. right handed molecules instead of left, different DNA coding/components etc.
That's the cool thing about finding truely alien bacteria - how similar they would be. You could tell how much of the choices life makes are frozen accidents or whether they optimal in some way.
x86 outperforms all of your examples on a per-CPU basis
x86(64) rocks for single CPU, but as you look at systems with more CPUs it tends to fall behind.
http://www.aceshardware.com/SPECmine/top.jsp
I'm not saying that x86 is worse at this stuff than Power et al, just that there aren't many people making these sorts of systems.
If AMD have any sense at all, they'll concentrate on that stuff while Intel wastes time and money on the doomed Itanium - the margins must be higher than single CPU desktop stuff. It's be easy too, an AMD chipset for 32 or 64 way to prime the market.
Mind you there's no accounting for the dumbness of big companies. Last time I looked at their site, you could parody^Wsummarise it as
"OMFG we get 150FPS in [fashionable new game] at [ultra high resolution] we are teh l33t Intel only get 120FPS"
as if you can sustain a billion dollar company selling $1000+ CPUs to the richest 1% of PC gamers.
Actually Martian bugs are a win/win situation - if there is no common ancestor then we'd get to see how similar they are - like if they use DNA with a different "instruction set", or a different molecule, or maybe even a completely different mechanism for storing their genes. My guess would be that truly alien life, with no common ancestor would be very different indeed.
As you point out, if there is a common ancestor then panspermia gets a boost. Panspermia between Earth and Mars seems fairly plausible to me - I read articles that calculated the amount of rock transferred by impacts between Earth and Mars in the last few billion years. Even with worst case assumptions, it was possible for viable bacteria to have been transferred.
Plus, you might actually get funding for my favourite mission, one that checks the Europan sea for life. I reckon you'd have a much better chance of finding truly alien life on Europa than Mars - the amount of rock transferred from Mars/Earth to Europa should be fairly minimal. Also, since the seas on Europa are kilometres beneath the surface it should be fairly well protected against 'infection' from meteorites.
Imagine him saying - "After all, when was the last time you heard someone rave about the interaction design of a product that wasn't on the Web? (Okay, besides the iPod.) All the cool, innovative new projects are online."
Slashdot Mirror
Not really - the real defense against hacking/piracy is that only Sony can write UMDs.
They did have shows on how this was done, back about 1977, all the magic of a Gould computer cracking this stuff out. Pretty impressive, then, for us people contending with IBM 360's and PDP 11's
Why a Gould computer? Surely the Asgard ones have more horsepowerr?
That made me laugh out loud. I pictured something like this -
...mystic stuff about the power of two...
manufacturing guy - 221mm2 is a big ass chip dude, almost twice the size of a P4 (112mm2 at 90nm).
KK -
MG - well 4's also a power of two.
Mind you, people have pointed out the Emotion Engine in the PS2 was about the same size and Sony made money selling that.
Still, it's interesting the manufacturing people obviously thought the chip was too big to be manufactured cheaply.
Hey, but the mplayer team are working on this I'm sure.
Geez a rant about how the project will CLOSE SOON due to software patents, and a bunch of butt-ugly screenshots on the next page.
4096 bits is a sector isn't it? That's the minimum unit for a read on an IDE disk.
Actually I'd guess most OS's do most IO in response to a page fault, so it's probably multiples of the page size, which is n*4Kbytes or n*32Kbits, so you could actually have a larger read size without much of a penalty.
I think the coolest thing is that it's chip-like and gives you more density than flash, and it looks like it will be cheap too.
Flash was always supposed to take over from magnetic disks as far as I can see, but in retrospect that was completely bogus. If you look at the cost per bit, it was too high, and relying on Moore's law to reduce that wouldn't work, since magnetic media has it's own version of Moores where the time to double capacity was actually less than for silicon.
In general, I think any completely solid state technology will be tend to lag behind something with mechanical addressing because making large chips tends to be prohibitively expensive. But this manages to use chip technology and add in some mechanical stuff, which is probably the way to go.
Hope the products based on it are cheap and reliable - they'd be great in an MP3 player. Incidentally, all this talk of nanometre positioning makes me wonder how sensitive it is to vibration - the main problem I have with hard disks in MP3 players/phones etc is that they aren't very robust compared to flash, at least when they are spinning. This technology sounds like it would be worse given the smaller tolerances involved.
Oh and one other worrying thing - they don't say how standard the process used to make it is. I don't know whether it uses some exotic process or whether you could make it in a standard fab plant with some tweaks.
That deaf, dumb and blind kid sure plays a mean pinball!
Funny, I always thought was the other way around.
If you were a real geek, having your crotch heated to 98C would be a small price to pay for all that computing power.
Don't apologise for your taste in music, apologise for taking the marketing course
Public dental care is farce because the dentists aren't under enought incentive to treat you.
I'd guess in the States having your kids teeth straightened costs serious cash. On the other hand if all the other kids have it done, not having it done would put them at a disadvantage. People will spend serious money on their kids so it all works out.
Last time I was in the UK I spent ages phoning public dentists all of whom seem to have amazingly unfriendly receptionists / office hours.
Oddly enough, I broke both arms pretty badly on another visit to the UK, and the initial treatment was very good. Waking up in a filthy ward next to people who'd caught drug resistant bugs in the hospital sucked though.
I think for emergency stuff public provision is probably OK, but for the preventative or elective things it just isn't the solution - that needs to be done by private companies for a profit.
Very true.
The sad thing is, you'll probably get modded down for saying it.
Yeah, and if you can't do with a built in function, you can always hack together some VBScript. And if you can't do it in VBScript you can write the C++ and cut'n'paste it into an ATL based ActiveX component and call it from VBScript. And you can email it to non technical people and it will just work, because they _all_ have Excel.
For some reason that last paragraph reminds me of Star Wars -
If you only knew the POWER of the DARK SIDE!!!11!!
It's true
http://sources.redhat.com/ml/libc-alpha/2000-08/m
compare and contrast with this
http://weblogs.asp.net/oldnewthing/archive/2005/0
In fact the paper they reference on strlcat has this great example of strlcat/cpy vs strncat/cpy
Orginal unsafe code. The path string can be overrun if it is too small to hold the concatenated strings. If it is just before the return address on the stack, you have a possible exploit.with strncat / cpy. The code is safe but it's hard to read and thus to check it is safe. Also the calls to strlen require the string size be measured by counting the characters. Note that both strncat and strlcat will need to do this again when working out where to write the string to be concatenated. strncpy also wastes time zero filling the string, which strlcpy doesn't need to do.with strlcat / cpyIt's as easy to read as the first code, and probably quicker than the second one, since it skips the excess calls to strlen and the zero filling in strncpy. Of course there's a cost to checking for buffer overflows as you copy, so it's slower than the first one.
Yeah, I read about that, it's a cool hack, in the most positive sense of using something in a way that it was not designed for. Intel failed to spot this extra use for the AGP GART when they cloned the chip I suppose.
:-) )
It's an interesting question though. AMD got to design the chip and the architecture at the same time. Intel had to retrofit AMD's 64 bit stuff to the P4. There are all kinds of reasons why this would be hard - the P4 had a dual speed ALU which needed to be widened to 64 bit for instance.
I wouldn't be surprised at all if the resulting chip had some performance issues, but I haven't seen any benchmarks as to comparative 64 bit performance though. I don't particular like P4's even for 32 bit stuff - it looks like they ultrapipelined the CPU to get higher clock frequencies in a way that reduces performance compared to a similar priced AMD part for instance, whereas AMD seems to have worried about real benchmarks like Spec and ( Doom3
Did retrofitting AMD^H^H^HEMT64 make it worse?
The Cell is not going to be used in general purpose machines. It has a PowerPC core, but the performance is low compared to a G5. It also has a bunch or SPEs but these are not suitable for general purpose computation either - they lack MMUs for example, and only have access to a small amount of tightly coupled SRAM. The idea is that you split your application into a software pipeline, and implement the stages in different SPEs.
. ars
All this stuff doesn't matter for games consoles where you can code all the crucial bits in SPE assembler and the whole machine runs only one application, but it would kill performance on a general purpose OS.
http://arstechnica.com/articles/paedia/cpu/cell-1
In what way exactly?
Yup, both IAPX-432 and IA64 had architects like a dog has fleas.
There is one actually - the development of photosynthesis by cyanobacteria and the resulting oxygenating the atmosphere 2B years was presumably not good news for anerobic bacteria when it happened.
Not saying I can see this happening with Martian bugs, if Mars had life like this we'd detect it easily.
Wouldn't Martian bugs (assuming that they don't share a common ancestor) be different enough to be orthogonal to life on earth.
E.g. right handed molecules instead of left, different DNA coding/components etc.
That's the cool thing about finding truely alien bacteria - how similar they would be. You could tell how much of the choices life makes are frozen accidents or whether they optimal in some way.
MS don't give away T Shirts anymore, they give away body armour.
--email-c-code-to-china-to-be-hand-translated-into -assembler is 0.05% quicker.
x86(64) rocks for single CPU, but as you look at systems with more CPUs it tends to fall behind.
http://www.aceshardware.com/SPECmine/top.jsp
I'm not saying that x86 is worse at this stuff than Power et al, just that there aren't many people making these sorts of systems.
If AMD have any sense at all, they'll concentrate on that stuff while Intel wastes time and money on the doomed Itanium - the margins must be higher than single CPU desktop stuff. It's be easy too, an AMD chipset for 32 or 64 way to prime the market.
Mind you there's no accounting for the dumbness of big companies. Last time I looked at their site, you could parody^Wsummarise it as
"OMFG we get 150FPS in [fashionable new game] at [ultra high resolution] we are teh l33t Intel only get 120FPS"
as if you can sustain a billion dollar company selling $1000+ CPUs to the richest 1% of PC gamers.
Actually Martian bugs are a win/win situation - if there is no common ancestor then we'd get to see how similar they are - like if they use DNA with a different "instruction set", or a different molecule, or maybe even a completely different mechanism for storing their genes. My guess would be that truly alien life, with no common ancestor would be very different indeed.
As you point out, if there is a common ancestor then panspermia gets a boost. Panspermia between Earth and Mars seems fairly plausible to me - I read articles that calculated the amount of rock transferred by impacts between Earth and Mars in the last few billion years. Even with worst case assumptions, it was possible for viable bacteria to have been transferred.
Plus, you might actually get funding for my favourite mission, one that checks the Europan sea for life. I reckon you'd have a much better chance of finding truly alien life on Europa than Mars - the amount of rock transferred from Mars/Earth to Europa should be fairly minimal. Also, since the seas on Europa are kilometres beneath the surface it should be fairly well protected against 'infection' from meteorites.
"There can't be life on Earth, there's too much oxygen there"
Martian Chronicles
http://www.adaptivepath.com/images/team/headshot_g arrett.jpg
Imagine him saying - "After all, when was the last time you heard someone rave about the interaction design of a product that wasn't on the Web? (Okay, besides the iPod.) All the cool, innovative new projects are online."
Spooky, innit?