The article doesn't make sense when you try to put the numbers together.
The xbox360 has a tripple core PPC processor running at 3.2 ghz. Anand claims that this only twice as fast as a 733mhz Celeron. That would mean that each core is 33% slower than a 733mhz Celeron.
Granted, this PPC chip doesn't have all of optimization feature's you'd expect to see in a current generation Intel or AMD chip, but neither does a Celeron. To make matters worse, the Celeron in the xbox has a meager 128kb of L2 cache (compared to the 1mb L2 availble to the xbox360's cores) and stuck with slower memory and a smaller pipe to access it with.
If the processor were truely only 2x faster than a 733mhz Celeron, then there isn't any chance that backwards compatibility could be achieved through emulation. Yet, that is the route Microsoft has chosen to take.
...And some (like me) feel that anything from Hotmail most likely counts as spam anyway, and have the entire domain in my filter list.
Except the spam you're getting from hotmail isn't really coming from hotmail. Which is the kind of problem SenderID is supposed to solve in the first place.
Yes, all of them. Alternatively, you tell your employees which ISPs they can send email from on behalf of the company. This can be managed through whatever process you use to keep track of employee home phone numbers if you so desire.
The problem being solved is "I have no way of knowing if the email I am receiving was sent on behalf of someone from that domain."
Your complaint is "I don't want to have to tell anyone if the email was really sent from my domain." Guess what, we tried that -- it doesn't work. Get over it.
So, how does this work for companies with large numbers of home-workers who are happily sending main aout throught their home ISP's with "spoofed" headers claiming, quite correctly, that their email comes from the company?
Aside from the mirade of other alternatives listed here, the employer can also add the ISP's email servers to the list of "valid email from us can come from here" list.
For them to cancel it when they did, they probably ran across some VERY bad numbers...and I bet showing the episodes out of order, frequently pre-empting the show with sports, and placing it in the worst time slot on tv had nothing to do with it.../me wants to shoot Fox execs
At times I wonder why they even bothered producing the series in the first place.
Think about it for a minute. What component in the system is going to be hitting memory the most? (hint: it renders textures to the screen) Look at what kind of bus that most PC video cards have between the GPU and on-card memory -- they're built to reduce latency as much as possible.
Moving the memory controller onto the same die as the GPU yields a non-trivial performance benefit both in available bandwidth and latency (look at what kind of gains AMD got by moving the NB on-chip).
If the controller were on neither the CPU or the GPU, you end up with performance of both the CPU and GPU decreasing and higher production costs (more chips to produce/install and more traces on the board).
If the controller had been moved onto the CPU, CPU performance would be better but GPU performance would suffer. CPU performance constraints can be reduced significantly by proper management of the L2 cache. However, caching for a GPU is not an effective stratedgy given that most of the bandwidth needed is either pushing rendered data out to memory (never to be read again) or pulling in data for textures. In other words, placing the memory controller on the CPU is not an optimal performance gain.
Sticking that puppy on the GPU was definately the right decision.
The original xbox design was meh. It was a brute force attempt at designing a console in a short period of time. It did bring some unique things to the console party -- an integrated hd and xbox live.
Of course, I wouldn't expect someone with the mental capacity of a fish to be able to objectivly evaluate the situation, so feel free to continue trolling.
The GPU also functions as the main memory controller; it has full access to all of the memory on the machine.
MEMEXPORT is exceptionally cool because this means that the CPU can stream data to the cache and let the GPU access it directory. This reduces latency and yields more useable bandwidth for main memory for both the CPU and the GPU.
I don't. The difference between the two is the "character" of the graphics.
GT4 has a "brighter" more pleasant look to some of its tracks, but isn't very realistic; for larger tracks GT4 is lacking detail/draw distance. The overall graphic quality hasn't changed much since GT3 IMO.
Forza tracks are "duller", but is a more accurate reproduction of the environments they're trying to re-create (with the exception being the ring; it "looks" like the ring, but the layout isn't much like the real thing). Forza has proper color reflections and shadow occlusion.
Forza's graphics are AA at a solid 30fps; GT4 runs at 60fps with no AA.
Car models are a tossup; some are better in one game, some in the other. The cars in GT4 always seem overly shiny/reflective, which isn't the case in Forza.
I personally like the graphics in Forza better than in GT4, but it really depends on your tastes and what kind of "candy" you prefer.
The real advantage forza has is not the graphics, but rather the physics -- Forza is a much better sim than GT4 IMO.
I don't. The difference between the two is the "character" of the graphics.
Tracks: GT4 is "brighter" and more pleasant to look at, but isn't very realistic; for larger tracks GT4 is lacking detail/draw distance. Forza is "duller", but is a more accurate reproduction of the environments they're trying to re-create.
Environment: Forza wins this one hands down; color reflections and accurate shadows are the big differences.
Cars: A bit of a tossup; GT4 is better with some models, Forza is better with other models.
Document all the COM interfaces provided by Media Player in sufficent detail that anybody could create objects which could implement those interfaces (NOTE: I say NOTHING about the ability to implement the actual FUNCTIONALITY behind those interfaces - just because you have a void foo(HWND *bar,FILE *narf) call does not mean that you actually foo correctly.)
Nothing prevents this from happening today. The COM interfaces are already documented on MSDN; hell, if you use the API you generally must know what it takes in as input and spits back out.
You've got it backwards. Reduced Media Edition was the name Microsoft suggested. The EU demanded that it be changed to something else. Everything they suggested that communicated that Media Center was not installed with the OS was rejected by the EU.
You just implied moronity, and given that the article is about MS the reference to MS is implicit.
You've also proceeded to miss the entire point, being that you CAN check for similar problems in the code, but not catch them. If the many eyes theory of open source development doesn't catch it, manual examination of code isn't an effective tool to combat all problems of this nature.
They aren't a CA based company, but they have a physical presence in California, which means purchases from the company to residents of California are subject to the CA state tax.
Borders was trying to argue that their online business wasn't connected to their physical presence in California, excempting them from CA state sales tax.
The court rule that the online store was indeed connected to their physical store, and thus wasn't exempt. This, to me, would seem to be the proper interpretation.
Draw distance on the current crop of racing games is already to the horizon, as best as I can tell.
Digital broadcasts may not be in HD, but I can't imagine that people will be getting SD sets in droves (for the same reason why people who just need a work processor get a 3ghz P4).
OSA is typically used to "start" the shortcut bar (it's a commandline option).
There is no such thing as an ActiveX "object library". You're thinking of ActiveX controls, which are by their very definition a "user-interface element." I'll assume that you meant "COM" instead of "ActiveX" (ActiveX controls are glorified com objects defined in a COM library).
So where does the code for these things come from? The tooth fairy? They reside in executables and DLLs. Code shared between the Office apps will reside in a DLL (you can't load an exe into a different exe).
I challenge you to find Office DLLs resident in memory. Getting a few system libraries into memory and showing a toolbar is hardly "keeping Office in ram".
There has never been a utility to keep Office in ram. To prove this to yourself, find a utility that scrapes for loaded/cached dlls and see if you can find any part of word loaded on the machine. You won't.
Their load times are the result of pure binary optimization. They order the binary bits and dll loads in the app to optimize the startup time. For an application of its size, it is quite impressive.
Slashdot Mirror
The article doesn't make sense when you try to put the numbers together.
The xbox360 has a tripple core PPC processor running at 3.2 ghz. Anand claims that this only twice as fast as a 733mhz Celeron. That would mean that each core is 33% slower than a 733mhz Celeron.
Granted, this PPC chip doesn't have all of optimization feature's you'd expect to see in a current generation Intel or AMD chip, but neither does a Celeron. To make matters worse, the Celeron in the xbox has a meager 128kb of L2 cache (compared to the 1mb L2 availble to the xbox360's cores) and stuck with slower memory and a smaller pipe to access it with.
If the processor were truely only 2x faster than a 733mhz Celeron, then there isn't any chance that backwards compatibility could be achieved through emulation. Yet, that is the route Microsoft has chosen to take.
Anand's numbers just don't add up.
Whenever you want to start an application as Administrator, right click on the item and select "Run As" from the context menu that appears.
This doesn't require you to open a command prompt, terminal session, or create a shortcut.
This is NOT an option available for the control panel, but in reality -- how often do you muck about in the control panel anyway?
You've got to take the injured person SOMEWHERE...
There would be a clear public benefit for providing emergency services based at a location which minimizes response time.
...And some (like me) feel that anything from Hotmail most likely counts as spam anyway, and have the entire domain in my filter list.
Except the spam you're getting from hotmail isn't really coming from hotmail. Which is the kind of problem SenderID is supposed to solve in the first place.
Yes, all of them. Alternatively, you tell your employees which ISPs they can send email from on behalf of the company. This can be managed through whatever process you use to keep track of employee home phone numbers if you so desire.
The problem being solved is "I have no way of knowing if the email I am receiving was sent on behalf of someone from that domain."
Your complaint is "I don't want to have to tell anyone if the email was really sent from my domain." Guess what, we tried that -- it doesn't work. Get over it.
So, how does this work for companies with large numbers of home-workers who are happily sending main aout throught their home ISP's with "spoofed" headers claiming, quite correctly, that their email comes from the company?
Aside from the mirade of other alternatives listed here, the employer can also add the ISP's email servers to the list of "valid email from us can come from here" list.
No, they said that the potential market of gamers they could tap into was a billion.
For them to cancel it when they did, they probably ran across some VERY bad numbers ...and I bet showing the episodes out of order, frequently pre-empting the show with sports, and placing it in the worst time slot on tv had nothing to do with it... /me wants to shoot Fox execs
At times I wonder why they even bothered producing the series in the first place.
Think about it for a minute. What component in the system is going to be hitting memory the most? (hint: it renders textures to the screen) Look at what kind of bus that most PC video cards have between the GPU and on-card memory -- they're built to reduce latency as much as possible.
.
Moving the memory controller onto the same die as the GPU yields a non-trivial performance benefit both in available bandwidth and latency (look at what kind of gains AMD got by moving the NB on-chip)
If the controller were on neither the CPU or the GPU, you end up with performance of both the CPU and GPU decreasing and higher production costs (more chips to produce/install and more traces on the board).
If the controller had been moved onto the CPU, CPU performance would be better but GPU performance would suffer. CPU performance constraints can be reduced significantly by proper management of the L2 cache. However, caching for a GPU is not an effective stratedgy given that most of the bandwidth needed is either pushing rendered data out to memory (never to be read again) or pulling in data for textures. In other words, placing the memory controller on the CPU is not an optimal performance gain.
Sticking that puppy on the GPU was definately the right decision.
The original xbox design was meh. It was a brute force attempt at designing a console in a short period of time. It did bring some unique things to the console party -- an integrated hd and xbox live.
Of course, I wouldn't expect someone with the mental capacity of a fish to be able to objectivly evaluate the situation, so feel free to continue trolling.
I look at it and I don't see how anyone with real world hardware experience can have your reaction.
The design is brilliant.
Indeed you are correct. My brain left out some words as I read it for some reason ...
The GPU also functions as the main memory controller; it has full access to all of the memory on the machine.
MEMEXPORT is exceptionally cool because this means that the CPU can stream data to the cache and let the GPU access it directory. This reduces latency and yields more useable bandwidth for main memory for both the CPU and the GPU.
I don't. The difference between the two is the "character" of the graphics.
GT4 has a "brighter" more pleasant look to some of its tracks, but isn't very realistic; for larger tracks GT4 is lacking detail/draw distance. The overall graphic quality hasn't changed much since GT3 IMO.
Forza tracks are "duller", but is a more accurate reproduction of the environments they're trying to re-create (with the exception being the ring; it "looks" like the ring, but the layout isn't much like the real thing). Forza has proper color reflections and shadow occlusion.
Forza's graphics are AA at a solid 30fps; GT4 runs at 60fps with no AA.
Car models are a tossup; some are better in one game, some in the other. The cars in GT4 always seem overly shiny/reflective, which isn't the case in Forza.
I personally like the graphics in Forza better than in GT4, but it really depends on your tastes and what kind of "candy" you prefer.
The real advantage forza has is not the graphics, but rather the physics -- Forza is a much better sim than GT4 IMO.
I don't. The difference between the two is the "character" of the graphics.
Tracks: GT4 is "brighter" and more pleasant to look at, but isn't very realistic; for larger tracks GT4 is lacking detail/draw distance. Forza is "duller", but is a more accurate reproduction of the environments they're trying to re-create.
Environment: Forza wins this one hands down; color reflections and accurate shadows are the big differences.
Cars: A bit of a tossup; GT4 is better with some models, Forza is better with other models.
Document all the COM interfaces provided by Media Player in sufficent detail that anybody could create objects which could implement those interfaces (NOTE: I say NOTHING about the ability to implement the actual FUNCTIONALITY behind those interfaces - just because you have a
void foo(HWND *bar,FILE *narf)
call does not mean that you actually foo correctly.)
Nothing prevents this from happening today. The COM interfaces are already documented on MSDN; hell, if you use the API you generally must know what it takes in as input and spits back out.
Except they didn't. The monopoly stuff, by the virtue of being a monopoly, can't occur until a monopoly exists.
You've got it backwards. Reduced Media Edition was the name Microsoft suggested. The EU demanded that it be changed to something else. Everything they suggested that communicated that Media Center was not installed with the OS was rejected by the EU.
You just implied moronity, and given that the article is about MS the reference to MS is implicit.
You've also proceeded to miss the entire point, being that you CAN check for similar problems in the code, but not catch them. If the many eyes theory of open source development doesn't catch it, manual examination of code isn't an effective tool to combat all problems of this nature.
They aren't a CA based company, but they have a physical presence in California, which means purchases from the company to residents of California are subject to the CA state tax.
Borders was trying to argue that their online business wasn't connected to their physical presence in California, excempting them from CA state sales tax.
The court rule that the online store was indeed connected to their physical store, and thus wasn't exempt. This, to me, would seem to be the proper interpretation.
Draw distance on the current crop of racing games is already to the horizon, as best as I can tell.
Digital broadcasts may not be in HD, but I can't imagine that people will be getting SD sets in droves (for the same reason why people who just need a work processor get a 3ghz P4).
OSA is typically used to "start" the shortcut bar (it's a commandline option).
There is no such thing as an ActiveX "object library". You're thinking of ActiveX controls, which are by their very definition a "user-interface element." I'll assume that you meant "COM" instead of "ActiveX" (ActiveX controls are glorified com objects defined in a COM library).
So where does the code for these things come from? The tooth fairy? They reside in executables and DLLs. Code shared between the Office apps will reside in a DLL (you can't load an exe into a different exe).
I challenge you to find Office DLLs resident in memory. Getting a few system libraries into memory and showing a toolbar is hardly "keeping Office in ram".
There has never been a utility to keep Office in ram. To prove this to yourself, find a utility that scrapes for loaded/cached dlls and see if you can find any part of word loaded on the machine. You won't.
Their load times are the result of pure binary optimization. They order the binary bits and dll loads in the app to optimize the startup time. For an application of its size, it is quite impressive.