I was kind of wondering about the "modern operating systems" comment... I think he meant "desktop operating systems".
What's a "desktop operating system" these days ? The only mainstream OS that hasn't seen extensive use and development in SMP server environments for a decade plus is OS X. For all the others, "desktop" vs "server" is just a matter of the bundled software and kernel tuning.
Even OS/2 could scale to 1024 processors if I recall correctly.
Yeah. Just like those old PPC Macs were "up to twice as fast" as a PC.
The article isn't worth reading IMO, not unless you're curious as to how much electricity some of the FB datacenters use. Otherwise it's light on the tech details.
Indeed. "All you wanted to know about FaceBook's infrastructure" and little more than a passing mention about their storage ? That's vastly more interesting information than where their datacenters might physically be.
I'm still curious as to what situations - outside of benchmarking - you're in where a x8 PCIe bus is constraining. Or even a 3Gb SATA port for that matter.
Think of the steam engine. It is a single invention that caused and made possible the entire industrial revolution. What it did was it allowed people to have more free time to construct infrastructure, and in made such construction projects easier to accomplish. But fundamentally, we do not know a lot more today than we knew then.
Yes we do.
Likewise, highly industrialized nations routinely get their asses handed to them by less industrialized nations, simply because the people in those nations are fighting for their lives.
Only when they're operating under ridiculous - in the context of outright conquest - constraints on what they can and can't do.
So while an alien civilization may have a much better scientific understanding than us, the ability to travel from star to star does not require this. And even if they were more industrialized, it doesn't mean they'd win in a fight.
What interstellar travel requires is manufacturing and energy generation capacities that are damn near incomprehensible to our society. *That* is why they would win in a fight.
While I don't disagree with the First Post... I do disagree with your supposition that they will significantly superior technologically.
The sheer energy and manufacturing/industrial requirements for meaningful interstellar _travel_, let alone conquest, strongly suggests they'd have to be.
you probably haven't read Turtledove's Worldwar series?
I haven't, but a quick reading of an overview suggests a premise (humanity == teh awesome) with no evidence to support it in a non-fictional setting.
There is a whole cluster of consumer drives today pushing ~275MB/s out of sata 3gb's 300MB/s limit. That's safely within the range of 'sata limited' allowing for a very small amount of controller overhead.
Though it's highly questionable as to whether to is any sort of meaningful "limit" in real-world usage.
In short, a slower PCIe extension cord using existing cables (as opposed to the oddball PCIe external cables). This will probably put pressure on mobo vendors to add more x16 slots. I regularly build storage servers with 16 and 24 drive bays, and it looks like top-end now are Tyan AMD boards with 4 x16 slots. I'd like to see, for instance, a SuperMicro with 6 PCIe x16 slots and dual Intel sockets (though I'm using AMD 12-core more and more lately). PCIe 3.0 is due out in a couple months, so probably it will be there - OCZ could also update to the faster coding rate.
I'm kinda curious, how often are you bus bandwidth constrained, and in what circumstances ?
But I think by concentrating on volume you are missing the big picture:
I'm not concentrating on volume, I'm making the point that your 90% number is a case of lying with statistics. It doesn't really tell you anything useful about the computing market because it's only providing information about a relatively tiny part of it. The cheapest useful Mac you can buy will set you back somewhere between US$800 and $900, whereas the average computer buyer spends $500 or less. That's why despite the 90% number, only something in the 5% region of computers actually out there are Macs.
So what these reports (and others) are saying is Apple has the high-end market.
No, they're not. They're saying is that of the computers sold in brick and mortar stores, Apple sells 90% of the ones that cost more than a grand.
So the 100 dual-CPU, quad-LCD, $3000ea PCs that some Radiology practice just bought directly from Dell aren't counted in that statistic at all. Neither are the 50 $2000 high-end gaming PCs that random people bought online from NewEgg yesterday. Neither are the dozen quad-CPU, 256GB-RAM servers a cloud hosting company just bought to expand their business with.
Despite a much lower volume they make considerable money. That's interesting for a variety of reasons, and Apple has taken the position that the commodity PC vendors can fight over the scraps.
There's nothing new about Apple's position. They've always been about low-volume, high-profit-margin products. Apple has never, ever demonstrated any ambition to become a majority market share holder the across the whole industry - one of the biggest reasons they've never really seriously courted businesses (outside of specialised niches - and even then it's mostly word of mouth), and why their product line has such gaping holes in it.
If Apple's tiny share of the market makes it far more valuable than 2 of the top PC manufacturers, what happens if it enlarges its share?
Their profit margin goes down because they can only increase marketshare by competing in the markets where the majority of sales are going (ca. $500). Not only that, but they need to retool their whole production and marketing strategy to cater to the things their new customers would expect (eg: predictable and consistent hardware for businesses, constantly dropping prices for home users).
Make no mistake. Apple as they work today could not compete with Dell (or HP) on Dell's turf. It'd be like throwing the best NFL team in the world into a cricket tournament.
And conversely, why should they reduce prices when they have so many buyers at the current prices -- enough to be profitable several times over the competition?
They shouldn't. Unless they want to increase market share, of course, which they can only do by selling lots of cheaper computers.
The problem with a world where every company is like Apple, is that (relatively speaking) hardly anyone has computers, because they can't afford them.
Finally, most of Apple's money isn't made selling computers, it's made selling iPods, iPhones, iPads and associated services. The overall point I'm trying to make here is that Apple's (excellent) financial position cannot be, in any way, mapped to a proportionally commanding position in the market.
Take a look... choose your source... it's been broadcast all over.
The most important thing to keep in mind with that statistic is that most PCs bought cost (well) under $1000, while only two Macs (practically speaking, only one, the MacBook is $999) cost less than a grand. The average price paid for a PC is something like $500, while for a Mac it's something like $1500.
Or, to put it another way, it's a bit like talking about how Bugatti has 90% (number made up, but you get the idea) of the marketshare for cars costing over two million dollars. It's absolutely true, but it doesn't tell you anything about the cars 90% *of all people* are buying.
Also, from memory, that statistic was only for computers bought in physical retail stores. Ie: the thousand PCs $SOME_BUSINESS just bought aren't even counted.
Butt seriously, considering that worldwide mobile phone sales totaled 269.1 million units in the first quarter of 2009, 4 million iPads isn't that big a deal. I know an awful lot of people, and only two of them own iPads.
At VMworld a few weeks back, it seemed to me that about every fourth or fifth person was carrying one around.
There are good reasons for retaining XP on existing systems, not least of which are the facts that upgrading would cost several hundred dollars and force a major change in the user interface. You can't upgrade an XP system to a Vista or Windows 7 kernel with DX11 support while leaving the rest of the system intact. The situations are not comparable.
Can you posit a reason why someone would want/need DX11 support, but where a) cost of upgrading is a meaningful concern and b) the minor changes in the interface are relevant ?
To put it another way, basically the only people who care about DX11 are hardcore gamers, who are happy to spend $hundreds on video cards and somehow manage to deal with the wildly different UIs in every game.
I have a server with a year-old Intel gigabit chipset where only one LAN port works under Solaris, both work under Linux. Last month the Solaris bug was sitting at "3 - Yes, that's a problem". I think the bug was reported about 10 months ago.
'Socially acceptable' doesn't make something right. eg. Keeping slaves used to be socially acceptable (almost required if you wanted to get ahead in society).
That depends on what you mean the competition. It was quite comparable to the Windows 95/98 OSes technically, but was technically far inferior to NT.
Windows 3.1 would be a much more accurate comparison. Co-operative multitasking, no memory protection, static disk cache, etc. Windows 9x was essentially a generation ahead of MacOS, NT another generation again.
I was relying on Apple to prevent personal computing from becoming cable fucking television.
The mind boggles as to why. Apple have been trying to perfect the computing *appliance* for over two and a half decades. All of this:
Here we are in 2010 and Apple Computers is synonymous with walled gardens, dropped calls and no USB ports on their handhelds. Batteries that cannot be changed without voiding the warranty. Planned obsolescence only 1 year out. Apps that have to be "curated" by Apple.
Is exactly the same strategy they've been pursuing since they released the first Mac, way back in 1984.
To understand how they got from 1996 to where they are today you need to remember that, flow of funds aside, it was actually NeXT that acquired Apple. Apple didn't pick up an operating system - NeXT acquired a hardware distribution channel.
Apple got where they are today because of the iPod. Without that, their main product would still be overpriced, underspecced PCs for niche applications.
Recently I was given an apple G3 B&W (I think y'all called them the blueberry) with 400MHz PPC and 3 128Mb RAM sticks along with a couple of hard drives and OSX Tiger installed. My question is this: Are these PPC machines worth sinking any money into as far as upgrades, or are they just old junk like Winboxes?
No Mac without a G5 or better processor in it will run OS X well. A dual G4 will just run it tolerably. A G3 is dismally bad.
According to her (I haven't fired it up yet as I need a PS/2 to USB adapter for my KVM) it ran Tiger just swell. So I was wondering if it was worth picking up some bigger RAM sticks or one of those G4 PPC upgrades, which I've found is around $100.
It will run Tiger about as well as a Pentium (yes, the original) class machine runs XP. My advice is to spend your $100 elsewhere.
A consumer should only have to pay for a movie once on DVD, and then use that copy anywhere - living room TV, streaming across the internal net to a PC or bedroom TV, and on a portable player like an iPod. The only reason they would insist a consumer should pay 3 times is because they are money-hungry Ebenezer Scrooge types.
Further, they should be able to get a higher quality copy when that is available (eg. HD copy in place of SD). After all, the value is in the content, not how it's delivered.
More importantly, how can it be "obviously wrong" when everyone does it ? (Literally, I've never known anyone in my life who hasn't violated copyright in some way, be it as simple as giving someone a mix tape, or something somewhat more illegal like copying rented games for people (back when CDRs were expensive). Heck, even my _aunty_ has some copyright-violating material, and she's well into her 60s and only been "on the internet" for a few years.)
With the exception of some of the brain-dead System 7.5 patches, the classic Mac OS was indistinguishable from OS/2 or WinNT as far as multitasking and stability went.
No it wasn't, not even close. You could bring a Mac to a dead halt simply by holding open a menu, and you'd be lucky to get a few days out of it without a bomb screen.
NT - and even OS/2 - would happily do things like burn CDs (at a blistering 4x) and play games simultaneously, with other stuff like a browser and email client ticking away in the background. The mere *idea* of doing that on MacOS is laughable.
Not to mention taking full advantage of higher-end machines with large amounts of RAM and multiple CPUs (though OS/2 wasn't particularly good at that either).
Neither Dad nor the Windows admin could tell the difference.
He couldn't have been trying very hard then. It was trivial to demonstrate how much MacOS's co-operative multitasking sucked, all you needed to do was start something reasonably large compressing with StuffIT and do something else while that was happening (or the aforementioned click-and-hold to keep a menu open).
Slashdot Mirror
I was kind of wondering about the "modern operating systems" comment... I think he meant "desktop operating systems".
What's a "desktop operating system" these days ? The only mainstream OS that hasn't seen extensive use and development in SMP server environments for a decade plus is OS X. For all the others, "desktop" vs "server" is just a matter of the bundled software and kernel tuning.
Even OS/2 could scale to 1024 processors if I recall correctly.
Yeah. Just like those old PPC Macs were "up to twice as fast" as a PC.
The article isn't worth reading IMO, not unless you're curious as to how much electricity some of the FB datacenters use. Otherwise it's light on the tech details.
Indeed. "All you wanted to know about FaceBook's infrastructure" and little more than a passing mention about their storage ? That's vastly more interesting information than where their datacenters might physically be.
I'm still curious as to what situations - outside of benchmarking - you're in where a x8 PCIe bus is constraining. Or even a 3Gb SATA port for that matter.
It's a limit for me, but admittedly, I'm a software developer, so my usage is a bit different from the conventional.
How, though ? What are you doing where a single drive maxing out at ~300MB/sec actually impacts your productivity ?
Think of the steam engine. It is a single invention that caused and made possible the entire industrial revolution. What it did was it allowed people to have more free time to construct infrastructure, and in made such construction projects easier to accomplish. But fundamentally, we do not know a lot more today than we knew then.
Yes we do.
Likewise, highly industrialized nations routinely get their asses handed to them by less industrialized nations, simply because the people in those nations are fighting for their lives.
Only when they're operating under ridiculous - in the context of outright conquest - constraints on what they can and can't do.
So while an alien civilization may have a much better scientific understanding than us, the ability to travel from star to star does not require this. And even if they were more industrialized, it doesn't mean they'd win in a fight.
What interstellar travel requires is manufacturing and energy generation capacities that are damn near incomprehensible to our society. *That* is why they would win in a fight.
While I don't disagree with the First Post... I do disagree with your supposition that they will significantly superior technologically.
The sheer energy and manufacturing/industrial requirements for meaningful interstellar _travel_, let alone conquest, strongly suggests they'd have to be.
you probably haven't read Turtledove's Worldwar series?
I haven't, but a quick reading of an overview suggests a premise (humanity == teh awesome) with no evidence to support it in a non-fictional setting.
There is a whole cluster of consumer drives today pushing ~275MB/s out of sata 3gb's 300MB/s limit. That's safely within the range of 'sata limited' allowing for a very small amount of controller overhead.
Though it's highly questionable as to whether to is any sort of meaningful "limit" in real-world usage.
In short, a slower PCIe extension cord using existing cables (as opposed to the oddball PCIe external cables). This will probably put pressure on mobo vendors to add more x16 slots. I regularly build storage servers with 16 and 24 drive bays, and it looks like top-end now are Tyan AMD boards with 4 x16 slots. I'd like to see, for instance, a SuperMicro with 6 PCIe x16 slots and dual Intel sockets (though I'm using AMD 12-core more and more lately). PCIe 3.0 is due out in a couple months, so probably it will be there - OCZ could also update to the faster coding rate.
I'm kinda curious, how often are you bus bandwidth constrained, and in what circumstances ?
It's not likely, but maybe the town got all it's revenue from tickets because many people were speeding there.
If large numbers of people are breaking a speed limit, it nearly always means the limit is wrong.
[...] and Democrats have to go insane leftist to win their primaries.
Helpful note for people in the rest of the Western world: "insane leftist" in the USA means "slightly right of center" for you.
But I think by concentrating on volume you are missing the big picture:
I'm not concentrating on volume, I'm making the point that your 90% number is a case of lying with statistics. It doesn't really tell you anything useful about the computing market because it's only providing information about a relatively tiny part of it. The cheapest useful Mac you can buy will set you back somewhere between US$800 and $900, whereas the average computer buyer spends $500 or less. That's why despite the 90% number, only something in the 5% region of computers actually out there are Macs.
So what these reports (and others) are saying is Apple has the high-end market.
No, they're not. They're saying is that of the computers sold in brick and mortar stores, Apple sells 90% of the ones that cost more than a grand.
So the 100 dual-CPU, quad-LCD, $3000ea PCs that some Radiology practice just bought directly from Dell aren't counted in that statistic at all. Neither are the 50 $2000 high-end gaming PCs that random people bought online from NewEgg yesterday. Neither are the dozen quad-CPU, 256GB-RAM servers a cloud hosting company just bought to expand their business with.
Despite a much lower volume they make considerable money. That's interesting for a variety of reasons, and Apple has taken the position that the commodity PC vendors can fight over the scraps.
There's nothing new about Apple's position. They've always been about low-volume, high-profit-margin products. Apple has never, ever demonstrated any ambition to become a majority market share holder the across the whole industry - one of the biggest reasons they've never really seriously courted businesses (outside of specialised niches - and even then it's mostly word of mouth), and why their product line has such gaping holes in it.
If Apple's tiny share of the market makes it far more valuable than 2 of the top PC manufacturers, what happens if it enlarges its share?
Their profit margin goes down because they can only increase marketshare by competing in the markets where the majority of sales are going (ca. $500). Not only that, but they need to retool their whole production and marketing strategy to cater to the things their new customers would expect (eg: predictable and consistent hardware for businesses, constantly dropping prices for home users).
Make no mistake. Apple as they work today could not compete with Dell (or HP) on Dell's turf. It'd be like throwing the best NFL team in the world into a cricket tournament.
And conversely, why should they reduce prices when they have so many buyers at the current prices -- enough to be profitable several times over the competition?
They shouldn't. Unless they want to increase market share, of course, which they can only do by selling lots of cheaper computers.
The problem with a world where every company is like Apple, is that (relatively speaking) hardly anyone has computers, because they can't afford them.
Finally, most of Apple's money isn't made selling computers, it's made selling iPods, iPhones, iPads and associated services. The overall point I'm trying to make here is that Apple's (excellent) financial position cannot be, in any way, mapped to a proportionally commanding position in the market.
Take a look... choose your source... it's been broadcast all over.
The most important thing to keep in mind with that statistic is that most PCs bought cost (well) under $1000, while only two Macs (practically speaking, only one, the MacBook is $999) cost less than a grand. The average price paid for a PC is something like $500, while for a Mac it's something like $1500.
Or, to put it another way, it's a bit like talking about how Bugatti has 90% (number made up, but you get the idea) of the marketshare for cars costing over two million dollars. It's absolutely true, but it doesn't tell you anything about the cars 90% *of all people* are buying.
Also, from memory, that statistic was only for computers bought in physical retail stores. Ie: the thousand PCs $SOME_BUSINESS just bought aren't even counted.
Butt seriously, considering that worldwide mobile phone sales totaled 269.1 million units in the first quarter of 2009, 4 million iPads isn't that big a deal. I know an awful lot of people, and only two of them own iPads.
At VMworld a few weeks back, it seemed to me that about every fourth or fifth person was carrying one around.
There are good reasons for retaining XP on existing systems, not least of which are the facts that upgrading would cost several hundred dollars and force a major change in the user interface. You can't upgrade an XP system to a Vista or Windows 7 kernel with DX11 support while leaving the rest of the system intact. The situations are not comparable.
Can you posit a reason why someone would want/need DX11 support, but where a) cost of upgrading is a meaningful concern and b) the minor changes in the interface are relevant ?
To put it another way, basically the only people who care about DX11 are hardcore gamers, who are happy to spend $hundreds on video cards and somehow manage to deal with the wildly different UIs in every game.
I have a server with a year-old Intel gigabit chipset where only one LAN port works under Solaris, both work under Linux. Last month the Solaris bug was sitting at "3 - Yes, that's a problem". I think the bug was reported about 10 months ago.
What's the server vendor and model ?
'Socially acceptable' doesn't make something right. eg. Keeping slaves used to be socially acceptable (almost required if you wanted to get ahead in society).
That didn't answer either of the questions.
That depends on what you mean the competition. It was quite comparable to the Windows 95/98 OSes technically, but was technically far inferior to NT.
Windows 3.1 would be a much more accurate comparison. Co-operative multitasking, no memory protection, static disk cache, etc. Windows 9x was essentially a generation ahead of MacOS, NT another generation again.
I was relying on Apple to prevent personal computing from becoming cable fucking television.
The mind boggles as to why. Apple have been trying to perfect the computing *appliance* for over two and a half decades. All of this:
Here we are in 2010 and Apple Computers is synonymous with walled gardens, dropped calls and no USB ports on their handhelds. Batteries that cannot be changed without voiding the warranty. Planned obsolescence only 1 year out. Apps that have to be "curated" by Apple.
Is exactly the same strategy they've been pursuing since they released the first Mac, way back in 1984.
All this from Microsoft allowed Apple the ability to bring itself out of the red and be still alive for when they finally released OSX in 2001.
Er, no. Apple were several billion in the black when Microsoft made their little investment.
To understand how they got from 1996 to where they are today you need to remember that, flow of funds aside, it was actually NeXT that acquired Apple. Apple didn't pick up an operating system - NeXT acquired a hardware distribution channel.
Apple got where they are today because of the iPod. Without that, their main product would still be overpriced, underspecced PCs for niche applications.
Recently I was given an apple G3 B&W (I think y'all called them the blueberry) with 400MHz PPC and 3 128Mb RAM sticks along with a couple of hard drives and OSX Tiger installed. My question is this: Are these PPC machines worth sinking any money into as far as upgrades, or are they just old junk like Winboxes?
No Mac without a G5 or better processor in it will run OS X well. A dual G4 will just run it tolerably. A G3 is dismally bad.
According to her (I haven't fired it up yet as I need a PS/2 to USB adapter for my KVM) it ran Tiger just swell. So I was wondering if it was worth picking up some bigger RAM sticks or one of those G4 PPC upgrades, which I've found is around $100.
It will run Tiger about as well as a Pentium (yes, the original) class machine runs XP. My advice is to spend your $100 elsewhere.
A consumer should only have to pay for a movie once on DVD, and then use that copy anywhere - living room TV, streaming across the internal net to a PC or bedroom TV, and on a portable player like an iPod. The only reason they would insist a consumer should pay 3 times is because they are money-hungry Ebenezer Scrooge types.
Further, they should be able to get a higher quality copy when that is available (eg. HD copy in place of SD). After all, the value is in the content, not how it's delivered.
Copying something is obviously wrong [...]
Why ?
More importantly, how can it be "obviously wrong" when everyone does it ? (Literally, I've never known anyone in my life who hasn't violated copyright in some way, be it as simple as giving someone a mix tape, or something somewhat more illegal like copying rented games for people (back when CDRs were expensive). Heck, even my _aunty_ has some copyright-violating material, and she's well into her 60s and only been "on the internet" for a few years.)
Apple gave us, in terms of innovation, Grand Central, Webkit, Aqua and made the decision to make WiFi ubiquitous in it's offerings.
How are any of these things innovative ?
With the exception of some of the brain-dead System 7.5 patches, the classic Mac OS was indistinguishable from OS/2 or WinNT as far as multitasking and stability went.
No it wasn't, not even close. You could bring a Mac to a dead halt simply by holding open a menu, and you'd be lucky to get a few days out of it without a bomb screen.
NT - and even OS/2 - would happily do things like burn CDs (at a blistering 4x) and play games simultaneously, with other stuff like a browser and email client ticking away in the background. The mere *idea* of doing that on MacOS is laughable.
Not to mention taking full advantage of higher-end machines with large amounts of RAM and multiple CPUs (though OS/2 wasn't particularly good at that either).
Neither Dad nor the Windows admin could tell the difference.
He couldn't have been trying very hard then. It was trivial to demonstrate how much MacOS's co-operative multitasking sucked, all you needed to do was start something reasonably large compressing with StuffIT and do something else while that was happening (or the aforementioned click-and-hold to keep a menu open).