This drive is a performance-oriented drive. If you only care about cost per GB, you won't be buying it. Anyone who is buying it, cares about performance; neglecting the aspect of performance that most desktop users will find most relevant is shoddy reviewing.
FWIW, I mostly agree with you — I care more about cost per GB than raw performance. That said, I still care about performance. Fortunately, most of the good vendors have drives with good performance now.
It's not as simple as using one data set before a specific point, and one after. Modern temperature measurement extends to periods before the divergence began. There's an overlap period that can be used to calibrate the tree growth method.
I thought it was pretty clear that what matters for most desktop users is the random small write speed. See, for example, Anandtech's SSD anthology and later followups.
So, where are the 4 KiB random write benchmarks? They are conspicuously absent from this review. We can see the effect, I think, in the IOMeter results -- the X-25M outperforms the OCZ drive across the board on those, despite the OCZ win in the throughput tests. But, personally, I'd like to see the raw numbers on 4 KiB random writes. Have this many reviewers really learned so little about benchmarking SSDs since they came out?
Lots of people drink out of plastic water bottles without getting cancer.
Most plastics are pretty darn safe. Cigarette smoke residue is pretty darn safe. However, they both have some amount of nasty stuff in them. Neither is absolutely, completely, 100% safe. Because neither one of them is *that* bad, direct comparisons are difficult — and completely impossible without data. Hence, the need for a citation that has some data and analysis of it.
Like I said... nice assurances. I wouldn't be surprised if you're right, but there is nasty crap in smoke residue, so I wouldn't be surprised if you're wrong. Citation?
We live in a world where there are actual fleets of robot assassins patrolling the skies. At some point there, we left the present and entered the future."
This has a lot more to do with detecting mines in old minefields that aren't combat zones, where civilians are routinely killed or maimed. It isn't about tactical detection of mines.
It depends how they use the tracker. (I don't follow KDE development, so this is mostly speculation.)
A lot of projects (open source or otherwise) put lots of things in the tracker. This would include 4.3 bugs, but also bugs introduced and fixed during 4.4 development that never made their way into a release. It might include feature requests for 4.4 that were implemented. It might include random discussion topics ("we should do this" followed by some discussion, closed as "no we shouldn't", or perhaps closed as "not doing this, but doing some other related idea that occurred to us while discussing, see this other bug"). It probably includes bugs about things not being sufficiently documented, or otherwise missing some polish — things users notice, but don't normally classify as "bugs".
I suspect that the 18000 number says more about how they use the tracker than it does about how buggy 4.3 was. That means it has some connection to how much work has been done, but doesn't really distinguish "fixing" from "adding features". A missing feature is just a type of bug of omission, after all.
Sometimes kdawson is the one who posts an article about an actually interesting story. He usually picks a crap, fud-filled summary of it (or edits it to be that...). But then the other editors (sometimes) notice and don't post that story. So if I block kdawson I miss some interesting stories.
Both blocking kdawson stories and reading kdawson stories are unsatisfactory solutions: both result in a lower-quality/. than if they would just fire kdawson.
He was innocent of what he was accused of. Being charged with a crime that the police and prosecutors know you did not commit is being charged with a crime that you are innocent of.
No, if he did in reality go there with the intention of stealing the $20 CD, in that state it would in fact be "Felony Commercial Burglary (Burglary being defined in California Penal Code as entering a premises with the intent to commit larceny)". The police simply dropped it to a smaller petty theft (at the same time making it stick without a costly court case) as it was indeed a $20 CD.
I am probably in a minority here, but I think the police acted in the right way, the person got what they should have gotten. The punishment for petty theft for committing petty theft.
In the abstract I agree with you: petty theft should get the petty theft punishment. The problem, though, is that a smarter person with a good lawyer (aka money) wouldn't say any of those incriminating things, and would probably get a plea bargain on the petty theft charge. I can't see how it's a good thing that knowledge and money matter that much when the crime and the evidence are the same. Taking advantage of the unprepared and the poor to stick them with harsher sentences is not justice.
To complain about this. Asimov himself had begun the work of integrating the Robot stories with his Foundation/Galactic Empire stories. All kinds of prequels and sequels were written by the master himself and by other authors and this is just more of the same. Details here.
Now, here is my question. In the original I Robot stories, the robot's positronic brains were made out of something referred to as Platinum-Iridium sponge. As this is written, Platinum is $1325/troy oz. and Iridium
is $425. Aren't you grateful that real computers are made out of silicon. Was any adjustment of technology made in the subsequent Robot stories?
Huh? High grade manufactured silicon is already worth more than its weight in gold or platinum. If replacing the silicon with gold was useful, people would be doing it already (at least on high end chips).
The prospect of retrofitting inductive chargers into every major highway makes the idea of a high-power fast-charge "gas" station look easy in comparison. High capacity batteries already tolerate a fairly rapid charge rate with the correct smarts in the charger (required anyway, really). Today, you only have to sacrifice a little bit of cost / capacity / weight to get 5-minute charge capability. There are improvements coming (LiFePO4, for instance) that should improve on that. Even the 1MW (Tesla Roadster, 100% charge, 3 minutes) charger electronics and cabling doesn't look that bad in comparison. Sure, you need smarts in the connectors to make it safe, and a thick cable -- but we already have things like smart connectors and thick cables for fueling gas cars.
Of course, said post will do a lot better if you start it off with "I'll probably be modded down for this, but..." And unless I'm mistaken, that's been known since at least the days of "The first Slashdot troll post investigation."
Disney's attitude toward digital media has changed a lot for the better since Eisner fucked off out and Steve Jobs became Disneys biggest shareholder after the pixar purchase.;-)
I seriously hate this crowd, When I was to post something about why doesn't Jobs do something at Disney about DRM, I get modded down for him only owning 7% and has no say in the companies activities.
However, when someone tries to put this to Job's name, everyone mods it up as fast as possible. Surely it can't be both ways?
Gee, it's almost like the slashdot moderator pool consists of multiple people, with differing views, only a few of whom actually moderated any individual comment.
Never hearing the term before, it very succinctly communicates the situation. I must say the mental image is also quite pleasant. Well done! ~the chemical engineering student who uses numerical methods to solve large problems
I suspect that the term "blow up" would be just as apt, though a little less British in the degree of understatement.
Rocket engineers are fond of that form of understatement. I've also heard "unscheduled disassembly", and I'm particularly fond of "turbine-rich exhaust".
Most launch vehicles are optimised to the point where they are basically balloons. They can't support themselves unless their tanks are pressurised and then only in one direction.
I read that US engineers watched with amazement when a Russian booster was winched off a truck at an air show supported horizontally by two cables, one at either end.
Actually, that is *not* true in general. It was true for the original Atlas, and is true for the Centaur high-energy upper stage, but most other modern launchers avoid balloon tanks. Most modern designs are very fragile, but self-supporting when unpressurized. That doesn't mean you can hoist them any way you please, but it's still a vast improvement in ease of handling. One of the requirements on the Shuttle External Tank design was that it not be a balloon tank. It was later discovered (to much embarrassment and annoyance) that the ET is self-supporting when empty or full, but that there is a partially-full intermediate range where it isn't, so it has to be filled while pressurized.
Some smaller launchers are assembled horizontally; in particular, SpaceX's Falcon I and Falcon 9 are. They're still fairly fragile, but they're closer to the Russian design approach in a variety of ways. Trading more structural margin, and hence lower payload fraction, for easier operations and hence lower cost per payload mass is one of those.
Given the relative prices of graphics cards and anything Cray sells, why not just run all computations in duplicate on two different graphics cards, and redo any that differ? Even given the performance penalties of regular checkpoints and comparisons on top of needing twice as much hardware, it still ought to be vastly cheaper.
Really? How, pray tell? Do you have a program that can determine whether two non-identical executables do the same thing? Will it also tell me whether those executables have infinite loops?
In general, the attack is considered very subtle and difficult to defend against. The best counter I'm aware of is diverse double compiling, which isn't exactly trivial. Perhaps more importantly, it's not actually in general use. Theoretical attacks are interesting, in part because someone else might actually be using them; theoretical solutions that aren't actually used don't protect against much.
My point was more about what is common practice than theoretically possible: most people don't understand most software they use. Similarly, most people don't verify their compilers through diverse double compiling.
The system will be just fine. It will continue operating at a higher CO2 level. Alongside that, it will have different weather patterns, average temperature, sea level, ocean acidity, rainfall patterns, and other changes. Many of those changes will be incompatible with the survival of some species. For us, though, they're merely incompatible with the way we're used to living -- where we have our major cities, our farmland, etc. Adapting to the changes would be neither easy nor pleasant.
The fact that the Earth will continue onward with a new balance does not mean we would much like that new balance.
There's a common theme here: the *externalities* aren't reflected in the price. That includes environmental effects, in the absence of proper regulation to internalize them. However, things like energy cost of producing the SUV are reflected in the price, at least roughly. Hence the original poster's claim: cost is a viable order of magnitude estimate. If something costs 50% more than something else, it's hard to tell which has the higher impact from that alone. If the cost difference is 10x, it's a good bet the more expensive one has a higher impact.
Slashdot Mirror
This drive is a performance-oriented drive. If you only care about cost per GB, you won't be buying it. Anyone who is buying it, cares about performance; neglecting the aspect of performance that most desktop users will find most relevant is shoddy reviewing.
FWIW, I mostly agree with you — I care more about cost per GB than raw performance. That said, I still care about performance. Fortunately, most of the good vendors have drives with good performance now.
It's not as simple as using one data set before a specific point, and one after. Modern temperature measurement extends to periods before the divergence began. There's an overlap period that can be used to calibrate the tree growth method.
I thought it was pretty clear that what matters for most desktop users is the random small write speed. See, for example, Anandtech's SSD anthology and later followups.
So, where are the 4 KiB random write benchmarks? They are conspicuously absent from this review. We can see the effect, I think, in the IOMeter results -- the X-25M outperforms the OCZ drive across the board on those, despite the OCZ win in the throughput tests. But, personally, I'd like to see the raw numbers on 4 KiB random writes. Have this many reviewers really learned so little about benchmarking SSDs since they came out?
Lots of people drink out of plastic water bottles without getting cancer.
Most plastics are pretty darn safe. Cigarette smoke residue is pretty darn safe. However, they both have some amount of nasty stuff in them. Neither is absolutely, completely, 100% safe. Because neither one of them is *that* bad, direct comparisons are difficult — and completely impossible without data. Hence, the need for a citation that has some data and analysis of it.
Like I said... nice assurances. I wouldn't be surprised if you're right, but there is nasty crap in smoke residue, so I wouldn't be surprised if you're wrong. Citation?
Yeah, plastics can be toxic. Do you have a citation that compares it to cigarette smoke?
Your assurances are very nice to have, I'm sure, but do you have any citations to go with them?
Give credit where credit is due.
We live in a world where there are actual fleets of robot assassins patrolling the skies. At some point there, we left the present and entered the future."
This has a lot more to do with detecting mines in old minefields that aren't combat zones, where civilians are routinely killed or maimed. It isn't about tactical detection of mines.
It depends how they use the tracker. (I don't follow KDE development, so this is mostly speculation.)
A lot of projects (open source or otherwise) put lots of things in the tracker. This would include 4.3 bugs, but also bugs introduced and fixed during 4.4 development that never made their way into a release. It might include feature requests for 4.4 that were implemented. It might include random discussion topics ("we should do this" followed by some discussion, closed as "no we shouldn't", or perhaps closed as "not doing this, but doing some other related idea that occurred to us while discussing, see this other bug"). It probably includes bugs about things not being sufficiently documented, or otherwise missing some polish — things users notice, but don't normally classify as "bugs".
I suspect that the 18000 number says more about how they use the tracker than it does about how buggy 4.3 was. That means it has some connection to how much work has been done, but doesn't really distinguish "fixing" from "adding features". A missing feature is just a type of bug of omission, after all.
Lots of things designed to operate at one temperature have problems when you quadruple their temperature.
Sometimes kdawson is the one who posts an article about an actually interesting story. He usually picks a crap, fud-filled summary of it (or edits it to be that...). But then the other editors (sometimes) notice and don't post that story. So if I block kdawson I miss some interesting stories.
Both blocking kdawson stories and reading kdawson stories are unsatisfactory solutions: both result in a lower-quality /. than if they would just fire kdawson.
He was innocent of what he was accused of. Being charged with a crime that the police and prosecutors know you did not commit is being charged with a crime that you are innocent of.
No, if he did in reality go there with the intention of stealing the $20 CD, in that state it would in fact be "Felony Commercial Burglary (Burglary being defined in California Penal Code as entering a premises with the intent to commit larceny)". The police simply dropped it to a smaller petty theft (at the same time making it stick without a costly court case) as it was indeed a $20 CD. I am probably in a minority here, but I think the police acted in the right way, the person got what they should have gotten. The punishment for petty theft for committing petty theft.
In the abstract I agree with you: petty theft should get the petty theft punishment. The problem, though, is that a smarter person with a good lawyer (aka money) wouldn't say any of those incriminating things, and would probably get a plea bargain on the petty theft charge. I can't see how it's a good thing that knowledge and money matter that much when the crime and the evidence are the same. Taking advantage of the unprepared and the poor to stick them with harsher sentences is not justice.
To complain about this. Asimov himself had begun the work of integrating the Robot stories with his Foundation/Galactic Empire stories. All kinds of prequels and sequels were written by the master himself and by other authors and this is just more of the same. Details here.
Now, here is my question. In the original I Robot stories, the robot's positronic brains were made out of something referred to as Platinum-Iridium sponge. As this is written, Platinum is $1325/troy oz. and Iridium is $425. Aren't you grateful that real computers are made out of silicon. Was any adjustment of technology made in the subsequent Robot stories?
Huh? High grade manufactured silicon is already worth more than its weight in gold or platinum. If replacing the silicon with gold was useful, people would be doing it already (at least on high end chips).
The prospect of retrofitting inductive chargers into every major highway makes the idea of a high-power fast-charge "gas" station look easy in comparison. High capacity batteries already tolerate a fairly rapid charge rate with the correct smarts in the charger (required anyway, really). Today, you only have to sacrifice a little bit of cost / capacity / weight to get 5-minute charge capability. There are improvements coming (LiFePO4, for instance) that should improve on that. Even the 1MW (Tesla Roadster, 100% charge, 3 minutes) charger electronics and cabling doesn't look that bad in comparison. Sure, you need smarts in the connectors to make it safe, and a thick cable -- but we already have things like smart connectors and thick cables for fueling gas cars.
Of course, said post will do a lot better if you start it off with "I'll probably be modded down for this, but..." And unless I'm mistaken, that's been known since at least the days of "The first Slashdot troll post investigation."
Disney's attitude toward digital media has changed a lot for the better since Eisner fucked off out and Steve Jobs became Disneys biggest shareholder after the pixar purchase. ;-)
I seriously hate this crowd, When I was to post something about why doesn't Jobs do something at Disney about DRM, I get modded down for him only owning 7% and has no say in the companies activities. However, when someone tries to put this to Job's name, everyone mods it up as fast as possible. Surely it can't be both ways?
Gee, it's almost like the slashdot moderator pool consists of multiple people, with differing views, only a few of whom actually moderated any individual comment.
I suspect that the term "blow up" would be just as apt, though a little less British in the degree of understatement.
Rocket engineers are fond of that form of understatement. I've also heard "unscheduled disassembly", and I'm particularly fond of "turbine-rich exhaust".
Most launch vehicles are optimised to the point where they are basically balloons. They can't support themselves unless their tanks are pressurised and then only in one direction.
I read that US engineers watched with amazement when a Russian booster was winched off a truck at an air show supported horizontally by two cables, one at either end.
Actually, that is *not* true in general. It was true for the original Atlas, and is true for the Centaur high-energy upper stage, but most other modern launchers avoid balloon tanks. Most modern designs are very fragile, but self-supporting when unpressurized. That doesn't mean you can hoist them any way you please, but it's still a vast improvement in ease of handling. One of the requirements on the Shuttle External Tank design was that it not be a balloon tank. It was later discovered (to much embarrassment and annoyance) that the ET is self-supporting when empty or full, but that there is a partially-full intermediate range where it isn't, so it has to be filled while pressurized.
Some smaller launchers are assembled horizontally; in particular, SpaceX's Falcon I and Falcon 9 are. They're still fairly fragile, but they're closer to the Russian design approach in a variety of ways. Trading more structural margin, and hence lower payload fraction, for easier operations and hence lower cost per payload mass is one of those.
Given the relative prices of graphics cards and anything Cray sells, why not just run all computations in duplicate on two different graphics cards, and redo any that differ? Even given the performance penalties of regular checkpoints and comparisons on top of needing twice as much hardware, it still ought to be vastly cheaper.
Really? How, pray tell? Do you have a program that can determine whether two non-identical executables do the same thing? Will it also tell me whether those executables have infinite loops?
In general, the attack is considered very subtle and difficult to defend against. The best counter I'm aware of is diverse double compiling, which isn't exactly trivial. Perhaps more importantly, it's not actually in general use. Theoretical attacks are interesting, in part because someone else might actually be using them; theoretical solutions that aren't actually used don't protect against much.
My point was more about what is common practice than theoretically possible: most people don't understand most software they use. Similarly, most people don't verify their compilers through diverse double compiling.
How many programs on your system do you *fully* understand? How certain are you about that?
The system will be just fine. It will continue operating at a higher CO2 level. Alongside that, it will have different weather patterns, average temperature, sea level, ocean acidity, rainfall patterns, and other changes. Many of those changes will be incompatible with the survival of some species. For us, though, they're merely incompatible with the way we're used to living -- where we have our major cities, our farmland, etc. Adapting to the changes would be neither easy nor pleasant.
The fact that the Earth will continue onward with a new balance does not mean we would much like that new balance.
He's pseudonymous, not anonymous.
There's a common theme here: the *externalities* aren't reflected in the price. That includes environmental effects, in the absence of proper regulation to internalize them. However, things like energy cost of producing the SUV are reflected in the price, at least roughly. Hence the original poster's claim: cost is a viable order of magnitude estimate. If something costs 50% more than something else, it's hard to tell which has the higher impact from that alone. If the cost difference is 10x, it's a good bet the more expensive one has a higher impact.