While Ubuntu was good at handling system upgrades with the Nvidia drivers, it was abysmal at handling the circumstance where I temporarily removed the Nvidia card and started using the onboard Intel graphics - I had to uninstall the Nvidia driver before X would start with the Intel driver. Gentoo gave the option to switch which library was active while both were currently installed. Granted, not a common situation but annoying nonetheless.
...by bundling the Nvidia package and its libraries to a package matching the Mesa components and install one or the other, but no one has yet done so.
Gentoo does an excellent job of managing the upgrades - the Mesa and Nvidia drivers are installed to different locations and symlinks are used to choose the right one, with a nice wrapper script to make it easy to choose what one you want with eselect.
I wonder what effect it would have on McDonalds to say to everyone in the US, "Is 2009 finally the year people stop stuffing themselves with fattening poison?"
I doubt it would have any effect unless you made it explicit what the fattening poison was.
They did that a while ago (XP or earlier) when transisioning the mainstream OS line from 9x to NT. They aren't bolting any legacy cruft on top of Vista to create 7.
it's adding all the legacy support that seems to slow things down (and add security holes).
or have i got this wrong?
You have it wrong. They don't start from scratch and then add the legacy stuff back in. They start with the old code and incrementally improve it. Thus, the "legacy stuff slowing it down" is there from the beginning.
While all of them work well with SSDs as they write data more efficiently or run fewer applications in the background than XP
They also didn't really do a good job of comparing Vista to XP. Vista is by no means an OS that "runs fewer applications in the background". It also thrashes the hard drive a lot more than XP due to SuperFetch and the search indexer.
I didn't say they were king of servers. What I was suggesting is that the lead that Linux might take away in the server world in terms of job prospects probably isn't sufficient to offset the lead that Microsoft has in the desktop world. I could easily be wrong, but it would surprise me.
That's true.
A fifth of our company are programmers (~20/100) and we do nothing but Linux and web-based work.
And your company must be the only place that employs programmers. Where I work, a rough guestimate would put our programmer count to around 1000 (out of perhaps 3000) and we almost exclusively deal with Microsoft platforms.
I'm willing to agree with the statement that there are more job opportunities on Microsoft platforms even without the conclusive evidence because it just seems to make more sense. Windows has a much larger market share of desktop computing, and much of the Linux software is open source software written by skilled enthusiasts in their free time. There are a lot of Linux servers out there and clearly there must be Linux related jobs out there, but I wouldn't expect the numbers to be high enough to offset those of Microsoft.
Remember, the original post said
there might be more jobs on the Windows platform
so I wouldn't expect any evidence to back that up.
I think I might like Obj-C over C++, due solely to the really nice init/release/autorelease mech for memory allocation.
Sounds like someone needs to read up on the RAII design pattern (not to be confused with RIAA). Sensibly written C++ will automatically release memory when it is no longer used.
That's some rather screwed up maths there. They don't make $50 for every box sold. Some of that money goes to the retailer. Some of that money goes into packaging and shipping costs. I would hazard a guess that the people who invested the $200k would only see about $20 of the purchase cost (completely unfounded guess). That would bring the number of additional break even customers to 10,000.
That's what they said about Windows Mobile/CE. I can tell you right now that it isn't as simple as just recompiling under a different platform, a lot of the supposed same functions take different parameters or (worse) take the same parameters and return different types that the compiler doesn't warn about (int versus BOOL, where 0 means success as an int or fail as a BOOL - not to be confused with the built-in bool type).
Then there is the issue of input devices. The keyboard/mouse behaves differently to a console controller, you definitely need to do different things there.
Even if RAM were to be somehow "defragmented" how could it possibly make it any faster?
Cache locality. When you access data from memory, surrounding areas of memory is often copied to the cache on the CPU. This can often speed things up immensely, but it probably won't help speed up the described scenario.
Servers are plugged in at all times, and we still want minimal power draw to save money and heat output (and for people who care, the environment). It isn't just about battery life.
Anyway, Browsers are likely the most complex software to properly benchmark.
You have to be kidding. These browsers are userspace applications running on desktops and have a fairly consistent set of operations to complete the task. Try benchmarking various surveying data collection software applications on embedded devices and you will notice that they all do different things. It is near impossible to determine which is "faster" as that could mean testing sensible workflows, CPU efficiency and a variety of other factors. Or try benchmarking device firmware - it is extremely difficult to determine what is going on.
As a C++ programmer myself I generally agree with what you said. But when I say "isn't particularly well suited" I mean it isn't necessarily the best solution even though it may work. I was also referring to the typical inheritance and dynamic polymorphism style of OO that the "fanboys" tend to love.
PS: Thanks for the references, I'll look into them.
OO is very good for graphical interfaces, but it isn't particularly well suited for algorithms and other maths oriented stuff. Why should we care if OO fanboys are scared off? Decent developers know to use the right tool for the job, not try to shoehorn whatever their personal favourite is into every situation.
A single Radeon 4870x2 uses two chips. This Tesla thing uses 4 chips that are comparable to the Radeon ones. It should be obvious that they would be in a similar ballpark.
Seriously, why is this even news?
It isn't. Tesla was released a while ago, this is just a slashvertisement.
Slashdot Mirror
While Ubuntu was good at handling system upgrades with the Nvidia drivers, it was abysmal at handling the circumstance where I temporarily removed the Nvidia card and started using the onboard Intel graphics - I had to uninstall the Nvidia driver before X would start with the Intel driver. Gentoo gave the option to switch which library was active while both were currently installed. Granted, not a common situation but annoying nonetheless.
...by bundling the Nvidia package and its libraries to a package matching the Mesa components and install one or the other, but no one has yet done so.
Gentoo does an excellent job of managing the upgrades - the Mesa and Nvidia drivers are installed to different locations and symlinks are used to choose the right one, with a nice wrapper script to make it easy to choose what one you want with eselect.
Take your lame meme somewhere else, thank you.
I wonder what effect it would have on McDonalds to say to everyone in the US, "Is 2009 finally the year people stop stuffing themselves with fattening poison?"
I doubt it would have any effect unless you made it explicit what the fattening poison was.
They did that a while ago (XP or earlier) when transisioning the mainstream OS line from 9x to NT. They aren't bolting any legacy cruft on top of Vista to create 7.
it's adding all the legacy support that seems to slow things down (and add security holes). or have i got this wrong?
You have it wrong. They don't start from scratch and then add the legacy stuff back in. They start with the old code and incrementally improve it. Thus, the "legacy stuff slowing it down" is there from the beginning.
While all of them work well with SSDs as they write data more efficiently or run fewer applications in the background than XP
They also didn't really do a good job of comparing Vista to XP. Vista is by no means an OS that "runs fewer applications in the background". It also thrashes the hard drive a lot more than XP due to SuperFetch and the search indexer.
I didn't say they were king of servers. What I was suggesting is that the lead that Linux might take away in the server world in terms of job prospects probably isn't sufficient to offset the lead that Microsoft has in the desktop world. I could easily be wrong, but it would surprise me.
That's true. A fifth of our company are programmers (~20/100) and we do nothing but Linux and web-based work.
And your company must be the only place that employs programmers. Where I work, a rough guestimate would put our programmer count to around 1000 (out of perhaps 3000) and we almost exclusively deal with Microsoft platforms.
I'm willing to agree with the statement that there are more job opportunities on Microsoft platforms even without the conclusive evidence because it just seems to make more sense. Windows has a much larger market share of desktop computing, and much of the Linux software is open source software written by skilled enthusiasts in their free time. There are a lot of Linux servers out there and clearly there must be Linux related jobs out there, but I wouldn't expect the numbers to be high enough to offset those of Microsoft.
Remember, the original post said
there might be more jobs on the Windows platform
so I wouldn't expect any evidence to back that up.
I think I might like Obj-C over C++, due solely to the really nice init/release/autorelease mech for memory allocation.
Sounds like someone needs to read up on the RAII design pattern (not to be confused with RIAA). Sensibly written C++ will automatically release memory when it is no longer used.
That's some rather screwed up maths there. They don't make $50 for every box sold. Some of that money goes to the retailer. Some of that money goes into packaging and shipping costs. I would hazard a guess that the people who invested the $200k would only see about $20 of the purchase cost (completely unfounded guess). That would bring the number of additional break even customers to 10,000.
Doesnt anyone do testing? Is it really that hard to find 200k to find 8 testers?
Wow, you really undervalue the testers...
That's what they said about Windows Mobile/CE. I can tell you right now that it isn't as simple as just recompiling under a different platform, a lot of the supposed same functions take different parameters or (worse) take the same parameters and return different types that the compiler doesn't warn about (int versus BOOL, where 0 means success as an int or fail as a BOOL - not to be confused with the built-in bool type).
Then there is the issue of input devices. The keyboard/mouse behaves differently to a console controller, you definitely need to do different things there.
Even if RAM were to be somehow "defragmented" how could it possibly make it any faster?
Cache locality. When you access data from memory, surrounding areas of memory is often copied to the cache on the CPU. This can often speed things up immensely, but it probably won't help speed up the described scenario.
That is true, but that has nothing to do with being plugged in to the wall. That is being more efficient.
Servers are plugged in at all times, and we still want minimal power draw to save money and heat output (and for people who care, the environment). It isn't just about battery life.
Bonus points if eyeball movement can be detected and the screen be moved in time with the wobble.
That might make it difficult if you actually want to look at a different part of the screen...
No, I just hate memes.
Anyway, Browsers are likely the most complex software to properly benchmark.
You have to be kidding. These browsers are userspace applications running on desktops and have a fairly consistent set of operations to complete the task. Try benchmarking various surveying data collection software applications on embedded devices and you will notice that they all do different things. It is near impossible to determine which is "faster" as that could mean testing sensible workflows, CPU efficiency and a variety of other factors. Or try benchmarking device firmware - it is extremely difficult to determine what is going on.
As a rather skinny virgin nerd user of Chrome who has my own place, I take offence to that.
Try using a RAM disk.
I suppose I'm one of those guys now. Hook, line and sinker.
As a C++ programmer myself I generally agree with what you said. But when I say "isn't particularly well suited" I mean it isn't necessarily the best solution even though it may work. I was also referring to the typical inheritance and dynamic polymorphism style of OO that the "fanboys" tend to love.
PS: Thanks for the references, I'll look into them.
OO is very good for graphical interfaces, but it isn't particularly well suited for algorithms and other maths oriented stuff. Why should we care if OO fanboys are scared off? Decent developers know to use the right tool for the job, not try to shoehorn whatever their personal favourite is into every situation.
A single Radeon 4870x2 is 2.4 TFLOPS.
A single Radeon 4870x2 uses two chips. This Tesla thing uses 4 chips that are comparable to the Radeon ones. It should be obvious that they would be in a similar ballpark.
Seriously, why is this even news?
It isn't. Tesla was released a while ago, this is just a slashvertisement.