@ GHz and faster system are now well in the cereal box prize category of bargain basement systems. Very little has been sold in the last year that wasn't that fast or faster. THat places the potential installed base of capable systems byt he time this actually is likely to be a retail product in the tens of millions. These are the sort of number most can only dream of selling into.
Completely wrong. The CPU requirements here are for GENERAL PURPOSE processors. This is a far cry from the efficiency of a dedicated hardware codec.
The first generation of x86 systems that could do good DVD playback in software were vastly more powerful in most ways than the chips in DVD players. Much more memory hungry, too, in that they had to run an application on a full feature OS rather than a tiny kernal pared down to just what was required for the intended task.
A dedicated codec for playback of WM9 or comparable codec is going to be vastly cheaper in volume than the general purpose CPU needed to achieve the same playback in software. As it only needs to perform a very limited set of functions compared to a CPU for a desktop the requirements for transistor count and speed are immemnsely lower. There is no reason an entry level WM9 capable video player should retail for more than $250, barring features for the high home theater options. That price would drop rapidly if the sales reached any great volume.
This guy took the words right out of my virtual mouth.
The one first generation SNES game that had a serious problem with maintaining sprite composure was Konami's Gradius III, and that was only in a single level featuring giant bubbles. This was an extremely overambitious design but the amount of attention it's gotten is ridiculous. You can't spend a fortune to master a machine without something to generate revenue in the near term. So of course your first game on a platform won't be a good as could have been at a later date with more experience on the system.
The SNES was more complex with a steeper learning curve, much like the PS2, but had advantages in color depth and palette along with chip functions that allowed to produce display that flat out were impossible to reproduce on the Genesis.
Sure, but keep in mind that the minimal recommended configurations for those games once represented the state of the art in technology for consumers. Those games exist with their level of quality and large sales because such equipment became common place, even obsolete compared to the bottom of the barrel $500 system in many stores. (Although video chips is one of the areas where low end machines are most likely to be deficient with many companies still selling Intel 810 based system with no AGP slot.)
AGP 8X may seem like too for anything mainstream to make good use of but in a few years 8X and NV30 will be in $500 systems. Developers will have had a few years to create better tools and learn to harness that power. Imagine then the games targeting what will then be considered a dull obsolete PC.
Then the people doing the purchasing for your university are drooling morons who've never worked out a proper procurement program to track such expenditures.
Even Microsoft would dislike this since a PC double billed on the OS is going to lead to less purchases of MS software than two computers. What may look like an accidental bonus can be a loss later on.
The problem with the piece is that John Lettice approaches the subject from an axe grinding position that blinds him to judging anything on its own merits. Microsoft? Ooh, it's just gotta be bad.
The Xbox Live approach is thus far the only fully formed end-to-end solution for adding online to the console world in a way that delivers a consistent consumer service, minimizes the infrastructure requirements to developers, and insures a revenue stream to make this a worthwhile thing to do for both the platform company (Microsoft) and small developers who don't want to run a back end operation and the related hassles for billing and customer support. If it doesn't grow the market in a big way online is simply betterleft to the PC realm which is better suited to small but profitable niche markets.
The companies like EA that claim MS want to control everything are blowing smoke up the public's collective hinder. What they're really saying is "WE want to control everything and most importantly not share so much as a penny of the revenue with anyone else." Considering that EA has managed to dump over $100 million down their online sinkhole I'm more than a little interested in seeing someone else take on the task.
THe Xbox setup doesn't prevent a developer from going their own way on online activity. It would just be very stupid on their part to waste resources duplicating the work and facilities already built by someone who can afford to shoulder the long term risk in pursuit of developing a new market.
At the other end of the spectrum it's easy to see why Nintendo is taking the approach of, "Here's the modem and Ethernet if you want to try something but don't expect any deeper involvement from us." Nintendo has been beating their head against the wall of online ventures for consoles since the mid-80's. Although I'm sure their management appreciates that it is an extremely different environment for such things today, especially the broadband aspect, it will be up to others to prove this is a worthwhile business. Even if they have to scramble to catch up later giving it a pass is a better thing to do right now while they have no shortage of opportunities to make massive sales intheir proven markets.
Different layer of the system. This just takes the stuff in ROM on an AirPort Base Station and let you run it on a Mac. Plenty of companies do the same for the PC side. This is a different issue from that of lowering the entry cost for the client.
Microsoft doesn't make WinModems or any part of them. Winmodem, in fact, is a 3COM brand. Why should Micosoft be expected to support someone else's product?
Although, XP did surprise me when it detected and supported a Lucent softmodem during an experiment to see how well XP handled some elderly hardware.
Get over yourself, lad. The PC market is vastly larger than that segment that interests you. Does the existence of tens of millions of Intel 810 -based business desktops somehow force you to use that chipset? I don't think so.
The make or break for this doesn't reside at MS or the comm chip companies. The people whose decison really counts are those at Dell, HP/Compaq, etc. If they get enthusiastic about offering WiFi in lower end portables this translates into revenue for those chip companies and the will to proceed. The primary market for this is as a built-in function for low end portables. Without interest from laptop vendor there will be no further investment in the effort.
Keep in mind that around the time this initiative is ready to ship in systems the average low end laptop will be based on a P4 or Athlon running at 1.5 GHz or greater. For the average user this is a box with a vast amount of spare CPU cycles to be tapped.
If you look at how Transmeta does this (Intel and AMD are moving in the same direction) you'll find that the throttling is based on load analysis. For instance, a Crusoe system will not throttling down as much for DVD playback as for something less demanding like word processing. It's already been demonstrated that CPU speed can be altered constantly and still receive benefit in extended battery life.
SoftWiFi would just be another factor in the load analysis. When no network transactions are occurring the CPU runs a wee bit slower and speed up that little bit when needed. SoftWiFi would be used primarily in low end system which are understood by most prospective purchasers to have lesser battery life. This would be consistent with the way the portable market has operated since the 80's.
No it doesn't. Keep in mind that the primary drivers behind softmodems were companies such as Motorola and Rockwell (now known as Conexent). THey knew exactly what they were doing, which was boosting the low end market for certain parts that remained in the soft device. The market for full implementations continued to do as well or better while the overall market grew a good deal.
Plus these same companies reaped the benefits when a portion of those systems were upgraded to better modems!
There was even earlier version of this. QMS (IIRC) released a QuickDraw-based laser printer for the Macintosh back in the 80's when the LaserWriter had the fastest CPU in the Apple product line! (16 MHz vs. 8 MHz in the early Macs.)
Really pushing the limits was the Amstrad PCW. This was Z80-based word processor sold in the UK that drove the printer off the system CPU.
Anything from Microsoft is a plot against the universe. Get over yourselves. This initiative is about one thing and and one thing only: selling more stuff. Just as it was with WinModems. Despite what one poster believes, a difference of a mere $25 to deliver a feature is hugely compelling to PC vendors. This significantly lowers the price point for a laptop with built-in WiFi.
To call this an undue burden on the CPU is ridiculous. The average user simply comes nowhere near making full use of what a GHz+ CPU is capable of providing. What they do know is that the new $1200 laptop they just bought has a hell of a lot more nifty toys built-in than the one they paid the same price for a few years back. More discriminating users will continue to have the option of buying systems using full hardwareimplementations for peripherals. They may pay more but then they're getting more, even with a CPU of identical performance. But they'll benefit from the existence of the performance draining soft-devices just the same. The expanded market thus created will lead to lower prices for items like access points and increased market for WiFi ISPs like BOINGO and Joltage leading to more access points for us frequent travelers.
Three years ago at IDF there was an Intel guy who proposed going the same route for DSL adapters and cable modems. The rise of home networking and inexpensive routers made this a non-starter. Te MS initiative is driven by the same intent to stimulate market growth by reducing the cost of entry.
Going the opposite direction, offloading functions from the CPU, will continue to make sense in the correct applications. Gaming is one of the few mainstream apps that truly uses every bit of power offered. Thus chips like the GeForce and Radeon series are extremely appropriate uses for supplemental processing in a PC. Likewise gamers will spend the extra money for full hardware implementations of peripherals to further free up the CPU for those things only it can do.
Sure, there'll be a lot of low end systems that suck for Quake X because the CPU is also doing the job of several hundred bucks worth of hardware. (Consider that Intel's horrible embedded graphics core is so common on corporate desktops that Intel technically is a major player in graphics as measured by market share.) Just smile and remember that their contribution to the mass market is lowering the overall cost for your own systems.
Slashdot Mirror
@ GHz and faster system are now well in the cereal box prize category of bargain basement systems. Very little has been sold in the last year that wasn't that fast or faster. THat places the potential installed base of capable systems byt he time this actually is likely to be a retail product in the tens of millions. These are the sort of number most can only dream of selling into.
Completely wrong. The CPU requirements here are for GENERAL PURPOSE processors. This is a far cry from the efficiency of a dedicated hardware codec. The first generation of x86 systems that could do good DVD playback in software were vastly more powerful in most ways than the chips in DVD players. Much more memory hungry, too, in that they had to run an application on a full feature OS rather than a tiny kernal pared down to just what was required for the intended task. A dedicated codec for playback of WM9 or comparable codec is going to be vastly cheaper in volume than the general purpose CPU needed to achieve the same playback in software. As it only needs to perform a very limited set of functions compared to a CPU for a desktop the requirements for transistor count and speed are immemnsely lower. There is no reason an entry level WM9 capable video player should retail for more than $250, barring features for the high home theater options. That price would drop rapidly if the sales reached any great volume.
This guy took the words right out of my virtual mouth.
The one first generation SNES game that had a serious problem with maintaining sprite composure was Konami's Gradius III, and that was only in a single level featuring giant bubbles. This was an extremely overambitious design but the amount of attention it's gotten is ridiculous. You can't spend a fortune to master a machine without something to generate revenue in the near term. So of course your first game on a platform won't be a good as could have been at a later date with more experience on the system.
The SNES was more complex with a steeper learning curve, much like the PS2, but had advantages in color depth and palette along with chip functions that allowed to produce display that flat out were impossible to reproduce on the Genesis.
Sure, but keep in mind that the minimal recommended configurations for those games once represented the state of the art in technology for consumers. Those games exist with their level of quality and large sales because such equipment became common place, even obsolete compared to the bottom of the barrel $500 system in many stores. (Although video chips is one of the areas where low end machines are most likely to be deficient with many companies still selling Intel 810 based system with no AGP slot.)
AGP 8X may seem like too for anything mainstream to make good use of but in a few years 8X and NV30 will be in $500 systems. Developers will have had a few years to create better tools and learn to harness that power. Imagine then the games targeting what will then be considered a dull obsolete PC.
Then the people doing the purchasing for your university are drooling morons who've never worked out a proper procurement program to track such expenditures.
Even Microsoft would dislike this since a PC double billed on the OS is going to lead to less purchases of MS software than two computers. What may look like an accidental bonus can be a loss later on.
Aargh! One big paragraph! I forgot to switch to plain text for a long post. My apologies.
The problem with the piece is that John Lettice approaches the subject from an axe grinding position that blinds him to judging anything on its own merits. Microsoft? Ooh, it's just gotta be bad. The Xbox Live approach is thus far the only fully formed end-to-end solution for adding online to the console world in a way that delivers a consistent consumer service, minimizes the infrastructure requirements to developers, and insures a revenue stream to make this a worthwhile thing to do for both the platform company (Microsoft) and small developers who don't want to run a back end operation and the related hassles for billing and customer support. If it doesn't grow the market in a big way online is simply betterleft to the PC realm which is better suited to small but profitable niche markets. The companies like EA that claim MS want to control everything are blowing smoke up the public's collective hinder. What they're really saying is "WE want to control everything and most importantly not share so much as a penny of the revenue with anyone else." Considering that EA has managed to dump over $100 million down their online sinkhole I'm more than a little interested in seeing someone else take on the task. THe Xbox setup doesn't prevent a developer from going their own way on online activity. It would just be very stupid on their part to waste resources duplicating the work and facilities already built by someone who can afford to shoulder the long term risk in pursuit of developing a new market. At the other end of the spectrum it's easy to see why Nintendo is taking the approach of, "Here's the modem and Ethernet if you want to try something but don't expect any deeper involvement from us." Nintendo has been beating their head against the wall of online ventures for consoles since the mid-80's. Although I'm sure their management appreciates that it is an extremely different environment for such things today, especially the broadband aspect, it will be up to others to prove this is a worthwhile business. Even if they have to scramble to catch up later giving it a pass is a better thing to do right now while they have no shortage of opportunities to make massive sales intheir proven markets.
Different layer of the system. This just takes the stuff in ROM on an AirPort Base Station and let you run it on a Mac. Plenty of companies do the same for the PC side. This is a different issue from that of lowering the entry cost for the client.
Microsoft doesn't make WinModems or any part of them. Winmodem, in fact, is a 3COM brand. Why should Micosoft be expected to support someone else's product?
Although, XP did surprise me when it detected and supported a Lucent softmodem during an experiment to see how well XP handled some elderly hardware.
Get over yourself, lad. The PC market is vastly larger than that segment that interests you. Does the existence of tens of millions of Intel 810 -based business desktops somehow force you to use that chipset? I don't think so.
The make or break for this doesn't reside at MS or the comm chip companies. The people whose decison really counts are those at Dell, HP/Compaq, etc. If they get enthusiastic about offering WiFi in lower end portables this translates into revenue for those chip companies and the will to proceed. The primary market for this is as a built-in function for low end portables. Without interest from laptop vendor there will be no further investment in the effort.
Keep in mind that around the time this initiative is ready to ship in systems the average low end laptop will be based on a P4 or Athlon running at 1.5 GHz or greater. For the average user this is a box with a vast amount of spare CPU cycles to be tapped.
If you look at how Transmeta does this (Intel and AMD are moving in the same direction) you'll find that the throttling is based on load analysis. For instance, a Crusoe system will not throttling down as much for DVD playback as for something less demanding like word processing. It's already been demonstrated that CPU speed can be altered constantly and still receive benefit in extended battery life.
SoftWiFi would just be another factor in the load analysis. When no network transactions are occurring the CPU runs a wee bit slower and speed up that little bit when needed. SoftWiFi would be used primarily in low end system which are understood by most prospective purchasers to have lesser battery life. This would be consistent with the way the portable market has operated since the 80's.
No it doesn't. Keep in mind that the primary drivers behind softmodems were companies such as Motorola and Rockwell (now known as Conexent). THey knew exactly what they were doing, which was boosting the low end market for certain parts that remained in the soft device. The market for full implementations continued to do as well or better while the overall market grew a good deal.
Plus these same companies reaped the benefits when a portion of those systems were upgraded to better modems!
There was even earlier version of this. QMS (IIRC) released a QuickDraw-based laser printer for the Macintosh back in the 80's when the LaserWriter had the fastest CPU in the Apple product line! (16 MHz vs. 8 MHz in the early Macs.)
Really pushing the limits was the Amstrad PCW. This was Z80-based word processor sold in the UK that drove the printer off the system CPU.
Anything from Microsoft is a plot against the universe. Get over yourselves. This initiative is about one thing and and one thing only: selling more stuff. Just as it was with WinModems. Despite what one poster believes, a difference of a mere $25 to deliver a feature is hugely compelling to PC vendors. This significantly lowers the price point for a laptop with built-in WiFi. To call this an undue burden on the CPU is ridiculous. The average user simply comes nowhere near making full use of what a GHz+ CPU is capable of providing. What they do know is that the new $1200 laptop they just bought has a hell of a lot more nifty toys built-in than the one they paid the same price for a few years back. More discriminating users will continue to have the option of buying systems using full hardwareimplementations for peripherals. They may pay more but then they're getting more, even with a CPU of identical performance. But they'll benefit from the existence of the performance draining soft-devices just the same. The expanded market thus created will lead to lower prices for items like access points and increased market for WiFi ISPs like BOINGO and Joltage leading to more access points for us frequent travelers. Three years ago at IDF there was an Intel guy who proposed going the same route for DSL adapters and cable modems. The rise of home networking and inexpensive routers made this a non-starter. Te MS initiative is driven by the same intent to stimulate market growth by reducing the cost of entry. Going the opposite direction, offloading functions from the CPU, will continue to make sense in the correct applications. Gaming is one of the few mainstream apps that truly uses every bit of power offered. Thus chips like the GeForce and Radeon series are extremely appropriate uses for supplemental processing in a PC. Likewise gamers will spend the extra money for full hardware implementations of peripherals to further free up the CPU for those things only it can do. Sure, there'll be a lot of low end systems that suck for Quake X because the CPU is also doing the job of several hundred bucks worth of hardware. (Consider that Intel's horrible embedded graphics core is so common on corporate desktops that Intel technically is a major player in graphics as measured by market share.) Just smile and remember that their contribution to the mass market is lowering the overall cost for your own systems.