Car audio competitors exceed 154db: * In a very small enclosed space (not a 16 meter room large enough to test spacecraft) * With an acoustic design to focus sound on the microphone (not intended to create a uniform soundfield) * For just a few seconds before the speaker voice coils melt * At a very small range of bass frequencies * Strictly without nobody inside the car to avoid certain injury - or perhaps even death, we have no way to know
154db may not be unusual but what the LEAF facility is doing certainly is unusual.
Henry Newman's response, however, is deeply flawed.
1) Newman complains that average drive age is a "useless statistic." But he seems to prefer "time since product release" which is far worse than useless -- it is an obviously incorrect way to estimate the age of a drive population and is directly contradicted by the average age data reported in the blog post. 2) Newman has questions about Backblaze's burn in. He can find answers by googling "Backblaze burn in" to learn more about the company's remarkably transparent operations. Beach does not go into these details because an effective blog post will focus on its key conclusions rather than discussing every detail of methodology. It is not a research paper. 3) Newman digresses into hard error rate which is unrelated to drive failures. I look forward to a future Backblaze blog post about error rates. In any case since all these drives are consumer drives and all but one have the same specified error rate it is a non-sequiter. 4) Newman points out that Backblaze probably vastly exceeds manufacturer specs for drive throughput. I think this is exactly the point. Is there really enough difference in reliability between commodity and enterprise drives to justify their price difference? Or is it just a form of price discrimination? Does the spec sheet reflect reality or is it a marketing-driven fiction?
Overall this article strikes me as being written by an industry flack: someone who is more interested in parroting jargon and received wisdom rather than indulging in genuine curiosity.
The small version of the logo (for example in the upper left corner of their home page) appears significantly misaligned. The red box will appear to be 2/3 of a pixel to the left of the blue box on most LCD panels due to the structure of the RGB elements in each pixel. The strong vertical lines and narrow gaps in the logo, along with the vertically aligned primary colors, makes this a really glaring problem. How embarassing for Microsoft, inventors of ClearType some 15 years ago, who should have known all about this issue.
It took me about 10 seconds to notice this and another 10 seconds to figure out why. Microsoft spent how long and how much money coming up with this logo?
The students were in audio related fields. They were included because otherwise the data would be age biased. Younger ears can hear higher frequencies. I had no desire to obscure my point by explaining the finer design points of the study to those who couldn't figure it out for themselves. If you want to call that "lying," so be it.
If you take an extremely quiet passage on a properly mastered and dithered CD and amplify it to levels where the quantization noise is audible, a 0db passage with the same gain will either destroy your test equipment or your ears, whichever comes first. This is not a matter of opinion, it is a matter of physics.
A group of sixty audio professionals and audiophiles did a series of controlled double blind trials published in the Journal of the Audio Engineering Society. They found no perceptible degradation caused by a 16-bit/44.1kHz A/D/A.
When you can tell the difference between 44.1/16 and 192/24 in a double blind trial, come back and we'll talk.
Subjective opinions about audio quality, particularly those accompanied by words like "deaf" or "idiot", are worse than useless. Subjective listening is deeply suggestible and unreliable. Claimed differences among any acceptably well designed audio electronics virtually always disappears under rigorous and controlled testing.
To give just one example, listeners reliably prefer the louder source in subjective testing, even if the difference is not consciously perceptible. If a 192/24 D/A is just 0.1db louder than a 44.1/16 source, listeners will tend to describe it in all sorts of subjective terms... "edgier," "richer," "more forward," "cleaner impact," "deeper soundstage" etc when in fact it is simply a little louder.
Did you actually look up the law before you offered your legal opinon? It's not hard, just Google "Nebraska law destructive device." It certainly isn't vague nor does it leave room for interpretation. It explicitly categorizes CO2 "bombs" as destructive devices.
Nebraska Penal Code 28-1213(7)(a) Destructive devices means: (i)(I) vessel or container intentionally caused to rupture or mechanically explode by expanding pressure from any gas, acid, dry ice, or other chemical mixture
At least the Nebraska law requires the intent to "use as a weapon against person or property;" California has a similar law, a felony with a 2 year mandatory minimum sentence, with no such stipulation. Indeed in California, even possessing dry ice and a soda bottle with the intent to put the one inside the other is a felony.
Do not mess around with CO2 "bombs" unless you are very familiar with the applicable law in your state. You could end up in very, very serious, very expensive trouble.
If you give them something that does the job better (which is to say, with a higher profit margin) they'll be all over it.
However, oil companies have a vast investment in oil infrastructure, technology and expertise which give them a significant advantage over potential competitors when it comes to delivering oil at a profit. They do not have an advantage at delivering plastic trees. As a result, they are not quite as unbiased as you imply with regard to the exact form in which energy is delivered to the customer.
There are obvious reasons not to put in too many specific details, in case the buyer reads Slashdot. (Unless of course this whole discussion is a negotiation tactic by the poster, in which case, bravo!)
You seem to want to provide accounting advice. Slashdot is not news for accountants, presumably the poster is better off hiring one of them for their accounting needs.
What Slashdot does have is older more experienced nerds who have been through this before and can give nerdly life advice. That's clearly what the poster is after, and we got exactly the right details to give a reasonable overview of their life outlook.
Your enthusiasm for your current gig puts you in a very nice position. Since you are not especially motivated to sell, you can drive a hard bargain and safely explore the limits of what the buyer will pay. If they get annoyed and walk away you go back to doing what you love. Win-win.
So, I would say, put your heads together and come up with the lowest price at which this decision would become a no-brainer, where you could all move on without regret. Make a counter offer at that price. Then go from there. The very least you will get is additional information about your negotiating position.
Just remember, you have the upper hand in the negotiation because you are happy to walk away.
Access means they can decrypt them. Given enough cycles, encryption can be broken.
What are you talking about? Encryption that can be broken with any feasible level of computing power is worthless. If you're assuming that once the bad guys get your ciphertext they'll be able to decrypt it sooner or later, why encrypt your data at all?
Certainly I'd prefer to have my valuable data stored with both physical security and encryption. But if I had to choose one or the other, I'd definitely choose encryption. If you compare the cost of the security measure with the cost to circumvent it, strong encryption is many orders of magnitude better than physical security.
If you want security, don't put things in untrusted spaces. Period.
Completely, utterly incorrect. It's a sad comment on the ambient understanding of data security that this got modded insightful.
Trust is seldom a good approach to security. Good security is when you can trust nobody and still sleep at night. That means strong encryption. That is exactly the approach implied by the article and it is exactly the right thing to do.
I think it is very unwise to ever assume any level of trust in the storage of backups, certainly offsite backups. The whole idea of backups is that you keep them around for a long time, in several copies and several locations. The more valuable your data, paradoxically, the more copies you need and the more widely dispersed they should be. This is antithetical to maintaining trust. The right way, indeed the only way out of this paradox is strong encryption.
Not necessarily. First, laws differ from state to state. Second, even in states that permit computer professionals to be exempt from overtime, we are not automatically exempt from overtime. Your employer can choose to classify you however they want. Often they will classify you as non-exempt by default, and additional paperwork is required to reclassify you as exempt.
Sometimes this comes about because staffing agencies that handle contract employees have an incentive to pay overtime. The staffing agency is paid as a percentage of payroll, so their financial incentive is aligned with the employee, not the client. Employers who use contract agencies need to watch carefully to make sure their contract employees aren't unnecessarily classified as exempt.
Screw that. I don't want overtime. I'm a contract software engineer and I always request to be exempt from overtime. Overtime is a curse.
I want to be able to work when I want to. I want to be able to work 12 hours today and 4 hours tomorrow. I want to be able to work 60 hours this week and 20 hours next week. My boss generally wants exactly the same thing. Flexibility benefits us both. In return for providing that flexibility, I get paid more every hour of every day than other employees.
If I am paid overtime, I will most likely be restricted in my ability to adjust my hours to the work load and to my own schedule. This harms both myself and my employer, and dilutes the value that I bring as a contract employee. Ultimately I get paid less, not more.
Broadly speaking, highly trained, highly valued professionals are in a sellers market and have no need for overtime. Purely commoditized and unskilled labor are the ones who need overtime laws to protect themselves.
Fortunately most of the stuff that is "slow, plodding, and sequential" isn't particularly challenging anymore, even for a single core on a modern CPU. The stuff that will bring even the fastest hardware to its knees tends to be inherently parallel. Things like video encoding & decoding, graphics and image processing, numerical and scientific computation, or manipulating vast quantities of data.
In part, that's because software developers face the same challenges that hardware developers do. One reason that hardware developers are going multicore is that designing a single custom CPU core to use a billion transistors rises to a superhuman level of complexity. From a project management perspective, it just makes more sense to replicate simpler designs.
Similarly, if you write a million lines of purely sequential, completely unparallelizable code, then execute it on a modern processor it's going to take a few seconds to run. Obviously that's an extreme example, but even in practice, real world problems which are accessible to merely human programmers are either inherently parallel at some level, or sooner or later become trivial to execute on single threaded hardware.
Nobody's saying that manned spaceflight wouldn't be useful for science. However, it certainly isn't cost effective.
We could invest in lowering launch costs, but we could also invest in improving robotics. Based on the last few decades of experience, in which the economics of spaceflight have barely changed, while robotics capabilities have improved by many orders of magnitude, I know where I'd put my money.
In the meantime, with today's technology and today's budgets, there is not the slightest doubt that unmanned science provides better bang for the buck.
She wants to crack down on violent video games, which, due to the pins and needles the economy is on right now, could devastate the economy if a major sector of the gaming industry would collapse.
You've got to be joking. I didn't know it was possible to be so utterly lacking in perspective.
Even if every single M rated video game were permanently taken off the shelves (which nobody is proposing) the total foregone profits in 2008 would be perhaps one or two billion dollars. In contrast, the subprime mortgage crisis has caused writedowns by major investment banks on the order of $10 billion -- per bank -- in the last quarter alone.
Or to look at it another way, the entire game industry makes no more than $10 billion in annual profits in the US, whereas total mortgage losses are likely to be well in excess of $200 billion -- with some estimates ranging beyond $1 trillion.
Both Sony and Microsoft have a well established, and documented, history of announcing shipped instead of sold numbers!
You can announce whatever you want, the question is what you recognize as revenue. For that you need to ask an accountant.
I'm not an accountant, but here's the basic principle. The important question is who is the customer and when a sale is made. If you sell a product to another company, such as a distributor or retail store, that's revenue, even if it hasn't been sold to their customer. If you ship a product to your own retail store, that's not revenue until it's sold to a customer.
Then you get complexities like, what happens if the distributor has an agreement where they can require you to buy back unsold product? Does that mean the distributor's inventory should also be treated as your own inventory?
That's why there's a genuinely interesting question about how many iPhones have really been sold to customers, and the truth may not be a simple matter of reading quarterly press releases from Apple and AT&T.
Last I checked both are listed on the NYSE. Do MSFT and SNE ring any bells for you?
Actually, MSFT is listed on Nasdaq, Sony is listed in Tokyo. SNE is a secondary listing (American Depository Receipt).
I'm all for a less belligerent America and negotiation with our rivals and enemies. However, "security by making people love you" has never worked and will never work.
Putin didn't lift anyone from poverty, commodities prices did. If oil were still $12/bbl, as it was when Putin took office, instead of $88 as it is today, he would be a footnote to history. Your gas guzzling SUV gave Vladimir Putin the opportunity to do shit like this.
It is not a coincidence that countries rich in natural resources tend to have the least democratic governments.
So explain this to me, again, one more time: Why exactly should I spend $2,000 on a new SSD in 2008?
You shouldn't, obviously.
But:
1) You won't have to. The SSDs you are referring to will drop in price by at least a factor of 5 by the end of 2008, for a variety of reasons. Read and write throughput will almost certainly rise to SATA2 bus speeds. People like Torvalds are excited about flash SSDs because they are aware of near-future hardware trends that you apparently are not.
3) Adding more RAM is inherently expensive from an architecture standpoint. I looked on Newegg and I can't find a motherboard that lets you add more than 8G of RAM with available RAM densities, so while the first 8G may be cheap, the next 8G isn't. If you have to upgrade to a 64 bit OS, going beyond 4G is pretty damn expensive. (Are you running XP Pro 32 or 64?) Whereas, flash SSDs are expensive today, but they are not inherently expensive.
4) Even for your relatively mundane use, a 32G swap partition, if properly managed as a persistent buffer cache, by a real operating system, would still have a huge impact on performance. Virtually your entire file system working set could be in the buffer cache. If you think it is in RAM right now, try quitting and restarting Firefox while watching the hard drive activity LED. Does it flash? OK, your working set is not in RAM.:)
5) Lots of people do much more challenging things on their computers than you do. I routinely do data analysis that strides across 30-40G databases. It is completely IO bound, because of seek latency. A proper SSD would deliver at least 10x speedup -- even if the throughput were no faster than a hard drive.
I've got to say, your clarification was itself more wrong than right.
1) With properly designed controllers, bank interleaving, etc., Flash based SSDs are rapidly approaching RAM based SSDs in performance. In any case, the performance (particularly in latency) of either one will be so much better than hard disks that minor performance differences are virtually irrelevant. The major differences is that RAM based SSDs are expensive, power hungry, volatile, and have poor packaging density. As a result, RAM based SSDs are, or soon will be, dead. Hence SSD is correctly becoming synonymous with "flash-based SSD."
2) Your perception that swap is no longer necessary is due to your own thinking being stuck in the past -- exactly what you accused the original poster of. SSDs will permit swap to be much faster than disk based backing store. It certainly does make sense to swap out hard disk buffer cache into SSD swap space, because the latter is so much faster than the former. This is basically the idea of hybrid hard drives, but it makes far more sense to have this process mediated by the VM system rather than some opaque controller on a hard drive.
3) Incidentally, even ignoring using SSDs for swap, eliminating swap is probably a bad idea. In many situations, a swap partition will improve performance even if you have a lot of RAM. It permits stale application data to get swapped to disk so that RAM can be used for buffer cache, speeding up currently running applications. If you want to know more, Google "vfs_cache_pressure" and "swappiness." Also, it's likely that in 2008 and 2009, many people who are still running 32 bit kernels will find themselves running into the 4G memory limit, and swap will serve as a temporary solution -- so don't write off swap as "magical thinking" just yet.
4) Few people properly appreciate how devastating the hard disk seek latency is to computer performance. A 32G swap partition on SSD serving as a large buffer cache would do more for perceived performance than almost any other upgrade. This is probably what Torvalds is referring to when he says that SSD is going to have a big impact on Linux in 2008.
Slashdot Mirror
"because we where[sic] peaking at 163db"
No actually you weren't.
Just because the meter says it doesn't mean its true.
Martin
Car audio competitors exceed 154db:
* In a very small enclosed space (not a 16 meter room large enough to test spacecraft)
* With an acoustic design to focus sound on the microphone (not intended to create a uniform soundfield)
* For just a few seconds before the speaker voice coils melt
* At a very small range of bass frequencies
* Strictly without nobody inside the car to avoid certain injury - or perhaps even death, we have no way to know
154db may not be unusual but what the LEAF facility is doing certainly is unusual.
Martin
No.
Henry Newman's response, however, is deeply flawed.
1) Newman complains that average drive age is a "useless statistic." But he seems to prefer "time since product release" which is far worse than useless -- it is an obviously incorrect way to estimate the age of a drive population and is directly contradicted by the average age data reported in the blog post.
2) Newman has questions about Backblaze's burn in. He can find answers by googling "Backblaze burn in" to learn more about the company's remarkably transparent operations. Beach does not go into these details because an effective blog post will focus on its key conclusions rather than discussing every detail of methodology. It is not a research paper.
3) Newman digresses into hard error rate which is unrelated to drive failures. I look forward to a future Backblaze blog post about error rates. In any case since all these drives are consumer drives and all but one have the same specified error rate it is a non-sequiter.
4) Newman points out that Backblaze probably vastly exceeds manufacturer specs for drive throughput. I think this is exactly the point. Is there really enough difference in reliability between commodity and enterprise drives to justify their price difference? Or is it just a form of price discrimination? Does the spec sheet reflect reality or is it a marketing-driven fiction?
Overall this article strikes me as being written by an industry flack: someone who is more interested in parroting jargon and received wisdom rather than indulging in genuine curiosity.
The small version of the logo (for example in the upper left corner of their home page) appears significantly misaligned. The red box will appear to be 2/3 of a pixel to the left of the blue box on most LCD panels due to the structure of the RGB elements in each pixel. The strong vertical lines and narrow gaps in the logo, along with the vertically aligned primary colors, makes this a really glaring problem. How embarassing for Microsoft, inventors of ClearType some 15 years ago, who should have known all about this issue.
It took me about 10 seconds to notice this and another 10 seconds to figure out why. Microsoft spent how long and how much money coming up with this logo?
You're welcome!
The students were in audio related fields. They were included because otherwise the data would be age biased. Younger ears can hear higher frequencies. I had no desire to obscure my point by explaining the finer design points of the study to those who couldn't figure it out for themselves. If you want to call that "lying," so be it.
If you take an extremely quiet passage on a properly mastered and dithered CD and amplify it to levels where the quantization noise is audible, a 0db passage with the same gain will either destroy your test equipment or your ears, whichever comes first. This is not a matter of opinion, it is a matter of physics.
A group of sixty audio professionals and audiophiles did a series of controlled double blind trials published in the Journal of the Audio Engineering Society. They found no perceptible degradation caused by a 16-bit/44.1kHz A/D/A.
http://www.aes.org/e-lib/browse.cfm?elib=14195
When you can tell the difference between 44.1/16 and 192/24 in a double blind trial, come back and we'll talk.
Subjective opinions about audio quality, particularly those accompanied by words like "deaf" or "idiot", are worse than useless. Subjective listening is deeply suggestible and unreliable. Claimed differences among any acceptably well designed audio electronics virtually always disappears under rigorous and controlled testing.
To give just one example, listeners reliably prefer the louder source in subjective testing, even if the difference is not consciously perceptible. If a 192/24 D/A is just 0.1db louder than a 44.1/16 source, listeners will tend to describe it in all sorts of subjective terms... "edgier," "richer," "more forward," "cleaner impact," "deeper soundstage" etc when in fact it is simply a little louder.
trust me - we count nanoseconds
Incidentally, light travels a bit more than one foot in a nanosecond. I trust you that you think you're counting nanoseconds.
Martin
Did you actually look up the law before you offered your legal opinon? It's not hard, just Google "Nebraska law destructive device." It certainly isn't vague nor does it leave room for interpretation. It explicitly categorizes CO2 "bombs" as destructive devices.
Nebraska Penal Code 28-1213(7)(a) Destructive devices means: (i)(I) vessel or container intentionally caused to rupture or mechanically explode by expanding pressure from any gas, acid, dry ice, or other chemical mixture
http://law.justia.com/nebraska/codes/s28index/s2812013000.html
At least the Nebraska law requires the intent to "use as a weapon against person or property;" California has a similar law, a felony with a 2 year mandatory minimum sentence, with no such stipulation. Indeed in California, even possessing dry ice and a soda bottle with the intent to put the one inside the other is a felony.
Do not mess around with CO2 "bombs" unless you are very familiar with the applicable law in your state. You could end up in very, very serious, very expensive trouble.
If you give them something that does the job better (which is to say, with a higher profit margin) they'll be all over it.
However, oil companies have a vast investment in oil infrastructure, technology and expertise which give them a significant advantage over potential competitors when it comes to delivering oil at a profit. They do not have an advantage at delivering plastic trees. As a result, they are not quite as unbiased as you imply with regard to the exact form in which energy is delivered to the customer.
I disagree.
There are obvious reasons not to put in too many specific details, in case the buyer reads Slashdot. (Unless of course this whole discussion is a negotiation tactic by the poster, in which case, bravo!)
You seem to want to provide accounting advice. Slashdot is not news for accountants, presumably the poster is better off hiring one of them for their accounting needs.
What Slashdot does have is older more experienced nerds who have been through this before and can give nerdly life advice. That's clearly what the poster is after, and we got exactly the right details to give a reasonable overview of their life outlook.
Your enthusiasm for your current gig puts you in a very nice position. Since you are not especially motivated to sell, you can drive a hard bargain and safely explore the limits of what the buyer will pay. If they get annoyed and walk away you go back to doing what you love. Win-win.
So, I would say, put your heads together and come up with the lowest price at which this decision would become a no-brainer, where you could all move on without regret. Make a counter offer at that price. Then go from there. The very least you will get is additional information about your negotiating position.
Just remember, you have the upper hand in the negotiation because you are happy to walk away.
Access means they can decrypt them. Given enough cycles, encryption can be broken.
What are you talking about? Encryption that can be broken with any feasible level of computing power is worthless. If you're assuming that once the bad guys get your ciphertext they'll be able to decrypt it sooner or later, why encrypt your data at all?
Certainly I'd prefer to have my valuable data stored with both physical security and encryption. But if I had to choose one or the other, I'd definitely choose encryption. If you compare the cost of the security measure with the cost to circumvent it, strong encryption is many orders of magnitude better than physical security.
If you want security, don't put things in untrusted spaces. Period.
Completely, utterly incorrect. It's a sad comment on the ambient understanding of data security that this got modded insightful.
Trust is seldom a good approach to security. Good security is when you can trust nobody and still sleep at night. That means strong encryption. That is exactly the approach implied by the article and it is exactly the right thing to do.
I think it is very unwise to ever assume any level of trust in the storage of backups, certainly offsite backups. The whole idea of backups is that you keep them around for a long time, in several copies and several locations. The more valuable your data, paradoxically, the more copies you need and the more widely dispersed they should be. This is antithetical to maintaining trust. The right way, indeed the only way out of this paradox is strong encryption.
Not necessarily. First, laws differ from state to state. Second, even in states that permit computer professionals to be exempt from overtime, we are not automatically exempt from overtime. Your employer can choose to classify you however they want. Often they will classify you as non-exempt by default, and additional paperwork is required to reclassify you as exempt.
Sometimes this comes about because staffing agencies that handle contract employees have an incentive to pay overtime. The staffing agency is paid as a percentage of payroll, so their financial incentive is aligned with the employee, not the client. Employers who use contract agencies need to watch carefully to make sure their contract employees aren't unnecessarily classified as exempt.
All non-managers go by the clock.
Screw that. I don't want overtime. I'm a contract software engineer and I always request to be exempt from overtime. Overtime is a curse.
I want to be able to work when I want to. I want to be able to work 12 hours today and 4 hours tomorrow. I want to be able to work 60 hours this week and 20 hours next week. My boss generally wants exactly the same thing. Flexibility benefits us both. In return for providing that flexibility, I get paid more every hour of every day than other employees.
If I am paid overtime, I will most likely be restricted in my ability to adjust my hours to the work load and to my own schedule. This harms both myself and my employer, and dilutes the value that I bring as a contract employee. Ultimately I get paid less, not more.
Broadly speaking, highly trained, highly valued professionals are in a sellers market and have no need for overtime. Purely commoditized and unskilled labor are the ones who need overtime laws to protect themselves.
Martin
Fortunately most of the stuff that is "slow, plodding, and sequential" isn't particularly challenging anymore, even for a single core on a modern CPU. The stuff that will bring even the fastest hardware to its knees tends to be inherently parallel. Things like video encoding & decoding, graphics and image processing, numerical and scientific computation, or manipulating vast quantities of data.
In part, that's because software developers face the same challenges that hardware developers do. One reason that hardware developers are going multicore is that designing a single custom CPU core to use a billion transistors rises to a superhuman level of complexity. From a project management perspective, it just makes more sense to replicate simpler designs.
Similarly, if you write a million lines of purely sequential, completely unparallelizable code, then execute it on a modern processor it's going to take a few seconds to run. Obviously that's an extreme example, but even in practice, real world problems which are accessible to merely human programmers are either inherently parallel at some level, or sooner or later become trivial to execute on single threaded hardware.
Martin
Nobody's saying that manned spaceflight wouldn't be useful for science. However, it certainly isn't cost effective.
We could invest in lowering launch costs, but we could also invest in improving robotics. Based on the last few decades of experience, in which the economics of spaceflight have barely changed, while robotics capabilities have improved by many orders of magnitude, I know where I'd put my money.
In the meantime, with today's technology and today's budgets, there is not the slightest doubt that unmanned science provides better bang for the buck.
Martin
She wants to crack down on violent video games, which, due to the pins and needles the economy is on right now, could devastate the economy if a major sector of the gaming industry would collapse.
You've got to be joking. I didn't know it was possible to be so utterly lacking in perspective.
Even if every single M rated video game were permanently taken off the shelves (which nobody is proposing) the total foregone profits in 2008 would be perhaps one or two billion dollars. In contrast, the subprime mortgage crisis has caused writedowns by major investment banks on the order of $10 billion -- per bank -- in the last quarter alone.
Or to look at it another way, the entire game industry makes no more than $10 billion in annual profits in the US, whereas total mortgage losses are likely to be well in excess of $200 billion -- with some estimates ranging beyond $1 trillion.
Martin
Both Sony and Microsoft have a well established, and documented, history of announcing shipped instead of sold numbers!
You can announce whatever you want, the question is what you recognize as revenue. For that you need to ask an accountant.
I'm not an accountant, but here's the basic principle. The important question is who is the customer and when a sale is made. If you sell a product to another company, such as a distributor or retail store, that's revenue, even if it hasn't been sold to their customer. If you ship a product to your own retail store, that's not revenue until it's sold to a customer.
Then you get complexities like, what happens if the distributor has an agreement where they can require you to buy back unsold product? Does that mean the distributor's inventory should also be treated as your own inventory?
That's why there's a genuinely interesting question about how many iPhones have really been sold to customers, and the truth may not be a simple matter of reading quarterly press releases from Apple and AT&T.
Last I checked both are listed on the NYSE. Do MSFT and SNE ring any bells for you?
Actually, MSFT is listed on Nasdaq, Sony is listed in Tokyo. SNE is a secondary listing (American Depository Receipt).
Martin
I'm all for a less belligerent America and negotiation with our rivals and enemies. However, "security by making people love you" has never worked and will never work.
Putin didn't lift anyone from poverty, commodities prices did. If oil were still $12/bbl, as it was when Putin took office, instead of $88 as it is today, he would be a footnote to history. Your gas guzzling SUV gave Vladimir Putin the opportunity to do shit like this.
It is not a coincidence that countries rich in natural resources tend to have the least democratic governments.
So explain this to me, again, one more time: Why exactly should I spend $2,000 on a new SSD in 2008?
:)
You shouldn't, obviously.
But:
1) You won't have to. The SSDs you are referring to will drop in price by at least a factor of 5 by the end of 2008, for a variety of reasons. Read and write throughput will almost certainly rise to SATA2 bus speeds. People like Torvalds are excited about flash SSDs because they are aware of near-future hardware trends that you apparently are not.
2) There are flash SSDs which are already much faster than any SATA hard drive: http://www.fastsilicon.com/index.php?option=com_content&task=view&id=333&Itemid=60
3) Adding more RAM is inherently expensive from an architecture standpoint. I looked on Newegg and I can't find a motherboard that lets you add more than 8G of RAM with available RAM densities, so while the first 8G may be cheap, the next 8G isn't. If you have to upgrade to a 64 bit OS, going beyond 4G is pretty damn expensive. (Are you running XP Pro 32 or 64?) Whereas, flash SSDs are expensive today, but they are not inherently expensive.
4) Even for your relatively mundane use, a 32G swap partition, if properly managed as a persistent buffer cache, by a real operating system, would still have a huge impact on performance. Virtually your entire file system working set could be in the buffer cache. If you think it is in RAM right now, try quitting and restarting Firefox while watching the hard drive activity LED. Does it flash? OK, your working set is not in RAM.
5) Lots of people do much more challenging things on their computers than you do. I routinely do data analysis that strides across 30-40G databases. It is completely IO bound, because of seek latency. A proper SSD would deliver at least 10x speedup -- even if the throughput were no faster than a hard drive.
Martin
I've got to say, your clarification was itself more wrong than right.
1) With properly designed controllers, bank interleaving, etc., Flash based SSDs are rapidly approaching RAM based SSDs in performance. In any case, the performance (particularly in latency) of either one will be so much better than hard disks that minor performance differences are virtually irrelevant. The major differences is that RAM based SSDs are expensive, power hungry, volatile, and have poor packaging density. As a result, RAM based SSDs are, or soon will be, dead. Hence SSD is correctly becoming synonymous with "flash-based SSD."
2) Your perception that swap is no longer necessary is due to your own thinking being stuck in the past -- exactly what you accused the original poster of. SSDs will permit swap to be much faster than disk based backing store. It certainly does make sense to swap out hard disk buffer cache into SSD swap space, because the latter is so much faster than the former. This is basically the idea of hybrid hard drives, but it makes far more sense to have this process mediated by the VM system rather than some opaque controller on a hard drive.
3) Incidentally, even ignoring using SSDs for swap, eliminating swap is probably a bad idea. In many situations, a swap partition will improve performance even if you have a lot of RAM. It permits stale application data to get swapped to disk so that RAM can be used for buffer cache, speeding up currently running applications. If you want to know more, Google "vfs_cache_pressure" and "swappiness." Also, it's likely that in 2008 and 2009, many people who are still running 32 bit kernels will find themselves running into the 4G memory limit, and swap will serve as a temporary solution -- so don't write off swap as "magical thinking" just yet.
4) Few people properly appreciate how devastating the hard disk seek latency is to computer performance. A 32G swap partition on SSD serving as a large buffer cache would do more for perceived performance than almost any other upgrade. This is probably what Torvalds is referring to when he says that SSD is going to have a big impact on Linux in 2008.
How long is a piece of string?