When millions of people get treated with a software based cancer treatment, a minor bug may kill thousands and no one will notice. In these cases there is no death due to the treatment, but because of worse than optimal treatment quality. Still, the latest optimization systems (IMRT) have already saved tens of thousands of people that could not be treated using manual techniques. Even if someone dies because of bad software, there is still enough quality software around to make it is an overall lifesaver.
I used to play xbattle around 1991. It was one of the first multi-player games for X11 and rather nice and strategic. You need to learn how to set it up, to get water obstacles, hills, city-building, digging and filling, etc. Also, it works only in 8-bit displays.
Even today, there are not too many multi-player games that are equally playable to xbattle.
It is not the format that is the problem. The format is rather well reverse-engineered already. The problem is the layout algorithm. People do care if their document looks different, a figure has jumped, there is one more page, etc.
Layout algorithms are very non-linear. Dramatic changes can happen in the layout due to differences in the rounding. Currently, there is absolutely no specification about the layout algorithms.
You don't need that much RAM do do a depth-first search of a play tree.
I disagree with this. Hashing is the key element for doing a fast chess search, and usually adding more RAM is more important to a chess program than improving on the CPU power.
Evaluation hashing and transposition hashing need a lot of memory. Also, just to setup the bitboard mechanism and all of its lookups require a megabyte or two. Without bitboards you lose possibly by a factor of 10 in speed.
I consider the depth of the opening books a much less important issue than having sufficient amounts of RAM available for the chess engine. The most important feature of the opening books is to bring variety into the game (for making the game more fun for human opponents), but that can be achieved with a relatively small opening library.
I prefer playing chess with my mobile phone over playing against a PDA. The modern phone software can beat most of the expert players in chess (elo 2500 is claimed for ChessGenius native version) and even the Java games like Deep Pocket Chess are strong enough for me. Well, to be honest, I have "archived" my PDA, since I have a limited pocket space.
What can I get for less than $100 without switching
Why are you introducing this completely artificial price limit on the phone, but do not have similar artificial price limit on the PDA? This is just plain stupid. I prefer to have a new phone over an old one. Also, I prefer to spend my $500 on the phone rather than having a lousy free phone phone and a PDA.
Let me turn your point completely over to show how silly it is: I actually got my PDA for free and I am not willing to to pay more than 100 euros for a new PDA. A PDA for 100 euros sucks. Thus a modern 500 euro phone is a better solution. Also, my PDA has a gray scale display so it is pretty much useless for gaming.
... or maybe I'm remembering what I read incorrectly.
Perhaps you should recheck before posting. I have not claimed that cell phones are more powerful than PDAs. It was another person replying to my original message who put that statement into my mouth. In my original post I claimed that the performance is about the same. Really, modern phones and PDAs are in the class of their own, while the GBA is quite different.
Further, you can play in dark with the phone, which is quite important. I really do not understand why they did not have a backlight for the GBA. It is just impossible to see anything, unless you have great lighting or a separate gadget for light.
The phones have 40% of the CPU of PDAs and 20x the CPU of GBA. It is sufficient to run most games, especially if compared to the GBA. The power consumption of modern phones for running the games is very, very low. Basically, the phones have a slightly more battery-saving architecture than the PDAs, but the difference in CPU performance is very small.
A Nokia series 40 phone that runs two weeks without reloading runs for 4 days with the CPU fully loaded, and a Symbian series 60 phone runs for 2-3 days with the CPU fully loaded. The batteris are not the problem, I have to admit that the form may be a problem for you. It is not a problem for me or my children.
When was the last time you used a good phone? Take a look at the Symbian series 60 phones, run a few benchmarks and come back. These are roughly 20 x faster than a GBA and about 40 % of the fastest PDAs, with 20 times the battery life. BTW, the fastest PDA's have 450 MHz, but are only around 100 MHz (Pentium equivalent) in computing power as the memory subsystem is so slow (minimal caches, narrow bandwidth) -- and even much less for floating point intensive computation.
If you had read my comment, you should have notice that I did not claim that a modern phone has more resources than a PDA, but and about the same computational resources.
Read the comment and check your facts before posting!
I don't like your personal attack. I am the CEO of a game house that has created several globally marketed games, and I know what I am talking about. The phones will be the platform for mobile gaming, especially the multi-player games. N-gage is just showing the direction, but the real, normal, traditional phones will be the gaming platform of the future, no matter if you like it or not. Ten years and gaming without communication can only be found in museums.
Both PDA and GBA are silly. Modern people carry their phones around anyway - from young children to elderly. Play with the phone, it is likely to have much more computing resources than the GBA, and about the same as that bulky PDA. Several good games are available for phones already; check out http://www.midlet-review.com for details. N-cage is only one option, most Nokia games and a few other are really incredible platforms for running Java and native solutions. There is a lot of bad games, but that does not mean that all the mobile games are bad.
I have written several programs for several different phones. Nokia's phones run the program fastest, most reliably and with the least number of system-related bugs. P900 is also okish, but you should not compare P900 against 6310i, they belong to different classes.
You can change the operator, keep your phone number, and the operator change is free (as in free beer) and the zero cost of the operator change is enforced by the law... umm...but only in Finland. I hate to see you foreigners fooled by your mobile operators, and feel sorry for you.
It is quite nice being able photo your children when you are on summer picnic - even if you forgot the camera. I really like my Nokia 3650.
You seem to be igonrant about the browser, just use Nokias browser or download Opera.
On your last remark on batteries, the main difference between PocketPC and phones is that phones have better battery time, with an expense of some 30-50% less CPU power.
It seems that many people are whining about these new Nokia phones, and mostly because they are not sufficiently geekish. But they are. You can develop stuff - be it games, apps, graphics, or something else - to them with free toolkits, download it over the air (from your webserver) or infrared and use. These are truely wonderful machines that obsolete your old geeky-pdas. Ok, I would feel better if they run Linux, but still I am glad that it is not Windows CE!
I fail to understand why people would look stupid talking to an n-gage or 7700. I can talk to an n-gage or any phone and look completely normal.
It is no more or less stupid than talking to any other phone, or talking in general. People who think it is stupid would have had to gasp for more air if they would have seen the first person talking to the first phone in the world, finally bursting into manic laughter, "look, how stupid!" -- or the first singer singing to a microphone, "oh, how stupid can that be!"
Nokia has had quite good success with 9110 and 9210 in some markets, and this looks like a good successor for those phones.
The format is not the problem; it is the layout algorithm. It is no longer sufficient to be able to transfer the data, but the it needs to look the same on different systems. The layout algorithms need to be standardized to a bit-perfect level in order to have true compatibility between two word processing systems. Another (perhaps less intelligent possibility) is to add layout information to the file when it is saved, and this layout info is manipulated when and if the document is manipulated in different systems, removing the need to standardize layout algorithms.
Safety and ease-of-use are interconnected. I participated in a project where we did replace an old text-based menu system (designed in the 1980's) using a tcl/tk on top of a largish medical Fortran-based simulation and planning system. I am pretty sure that the gui actually made it safer to use. The selection of tcl was done at a time when python was still at 1.3, and leaked memory like hell. Today, I would choose Python for the gui-building tool, if I would have to add a GUI to a system like that. However, even the tcl/tk conversion was a success story. The company got millions by selling the system, the gui giving a significant boost to declining sales, and the project was ready earlier! than was anticipated.
Is it really a good idea to make the most efficient part of the world be less efficient in a rather significant way, just to be able to pay a measely $100 000 000 to the aids campaing of poor countries? Problems introduced by software monopolies are in ranges of $100 000 000 000 (add more 0s here), and the monopoly tax is harming everyone, especially the poor in under-developed countries. I hope you were looking for some Funny-moderation by the reasoning included in your responce.
Ouch. Terraflops for teraflops. I will be waiting to see terrafeet (for astronomical distance measures) in US corporate press releases and popular science articles.
Future: Imagine a volunteer force of 100000000 people trying to find and collect the sensors of the previous generation, which is interfering with a newer generation of the global data collection system.
But we have been here before. Many goverments ruled that they need to have posix in their systems. Microsoft responded by building a posix subsystem that could not communicate with the rest of the system and started the posix processes with only about 80% reliability, so it could not be used. The 80% reliability worked well enough as a proof that posix was there to justify the purchase of the NT 3.1. Later it was removed, people were already buying NT 3.x, so it was not necessary.
Microsoft knows hot to set up a smokescreen of compatibility and remove it when the time comes. Most professionals will be fooled. If Microsoft is forced to set up a non-drm mode in their BIOS, it will be a 80 % quality solution. Just barely good enough to work as a proof of good will, but bad enough to stabilize Microsofts position in a monopoly of BIOS, life, computing, and everything.
Fast talk, crossfading two sentences, and not letting the consumer think, not even for a moment, are known manipulation methods used for example in tv commercials. Perhaps you are removing some important criticism processes by speed learning. If I need to keep a presentation, I try to place delays in suitable places to let the audience find things out by themselves.
One of my not-so-nerdish friend claimed that he is watching bold-and-beautiful (or some other all-american soap series, I am not sure which it was) by using 9x video forward.
We do not have dubbing here in Finland, we have subtitles. The subtitles are not only the reason for the high literacy rate in Finland, but also enable 9x fast forward for viewing this kind of quality entertainment.
My point with the xlib is that all software that has simple architecture should provide multiple levels of abstraction, known as layers. If you remove the layers, the result will be more complex.
In my opinion technology should not impose political decisions. Attaching the widget sets and enforcing similarity on the server side is doing just that, leading to situation where evolution (of widget sets, and windowing in general) is more difficult.
The true opportunities to improve on windowing systems lies in making the computer more like a news paper or a magazine, with proper columns, resource management for the columns, etc. Today, everyone is reading hugely wide text, leading to significant decrease in reading speed. Most of the work done with the computer is reading, and the UI should recognize this. Smaller screens should have less columns, while larger screens should allocate more. Minimization could make an article to become a headline, etc.
The paper is written around one idea, simplifying the internal structure of X. However, it is against the simplicity of X by claiming xlib's level of abstraction too low.
One should start from the requirements, like the internal simplicity, and modern features. One should also specify the scope of the windowing system; mobile systems need other type of solutions from zillion-pixel worksations.
Y lacks the obvious new features needed, such as subpixel cursor positioning, anti-aliasing, and 3d. IMHO, the paper is written by a confused state of mind - there is nothing shockingly good in it.
Many (if not most) scientists, who work with rats for several years, get allergic to rodents. Do not get casual with them if you have any dreams of working in a mice-lab later on in your life.
Slashdot Mirror
When millions of people get treated with a software based cancer treatment, a minor bug may kill thousands and no one will notice. In these cases there is no death due to the treatment, but because of worse than optimal treatment quality. Still, the latest optimization systems (IMRT) have already saved tens of thousands of people that could not be treated using manual techniques. Even if someone dies because of bad software, there is still enough quality software around to make it is an overall lifesaver.
Even today, there are not too many multi-player games that are equally playable to xbattle.
Layout algorithms are very non-linear. Dramatic changes can happen in the layout due to differences in the rounding. Currently, there is absolutely no specification about the layout algorithms.
Evaluation hashing and transposition hashing need a lot of memory. Also, just to setup the bitboard mechanism and all of its lookups require a megabyte or two. Without bitboards you lose possibly by a factor of 10 in speed.
I consider the depth of the opening books a much less important issue than having sufficient amounts of RAM available for the chess engine. The most important feature of the opening books is to bring variety into the game (for making the game more fun for human opponents), but that can be achieved with a relatively small opening library.
I prefer playing chess with my mobile phone over playing against a PDA. The modern phone software can beat most of the expert players in chess (elo 2500 is claimed for ChessGenius native version) and even the Java games like Deep Pocket Chess are strong enough for me. Well, to be honest, I have "archived" my PDA, since I have a limited pocket space.
Why are you introducing this completely artificial price limit on the phone, but do not have similar artificial price limit on the PDA? This is just plain stupid. I prefer to have a new phone over an old one. Also, I prefer to spend my $500 on the phone rather than having a lousy free phone phone and a PDA.
Let me turn your point completely over to show how silly it is: I actually got my PDA for free and I am not willing to to pay more than 100 euros for a new PDA. A PDA for 100 euros sucks. Thus a modern 500 euro phone is a better solution. Also, my PDA has a gray scale display so it is pretty much useless for gaming.
Perhaps you should recheck before posting. I have not claimed that cell phones are more powerful than PDAs. It was another person replying to my original message who put that statement into my mouth. In my original post I claimed that the performance is about the same. Really, modern phones and PDAs are in the class of their own, while the GBA is quite different.
Further, you can play in dark with the phone, which is quite important. I really do not understand why they did not have a backlight for the GBA. It is just impossible to see anything, unless you have great lighting or a separate gadget for light.
Wake up, USA. The PDA is dead. The GBA is dead.
A Nokia series 40 phone that runs two weeks without reloading runs for 4 days with the CPU fully loaded, and a Symbian series 60 phone runs for 2-3 days with the CPU fully loaded. The batteris are not the problem, I have to admit that the form may be a problem for you. It is not a problem for me or my children.
If you had read my comment, you should have notice that I did not claim that a modern phone has more resources than a PDA, but and about the same computational resources.
Read the comment and check your facts before posting!
I don't like your personal attack. I am the CEO of a game house that has created several globally marketed games, and I know what I am talking about. The phones will be the platform for mobile gaming, especially the multi-player games. N-gage is just showing the direction, but the real, normal, traditional phones will be the gaming platform of the future, no matter if you like it or not. Ten years and gaming without communication can only be found in museums.
Both PDA and GBA are silly. Modern people carry their phones around anyway - from young children to elderly. Play with the phone, it is likely to have much more computing resources than the GBA, and about the same as that bulky PDA. Several good games are available for phones already; check out http://www.midlet-review.com for details. N-cage is only one option, most Nokia games and a few other are really incredible platforms for running Java and native solutions. There is a lot of bad games, but that does not mean that all the mobile games are bad.
I have written several programs for several different phones. Nokia's phones run the program fastest, most reliably and with the least number of system-related bugs. P900 is also okish, but you should not compare P900 against 6310i, they belong to different classes.
You can change the operator, keep your phone number, and the operator change is free (as in free beer) and the zero cost of the operator change is enforced by the law... umm ...but only in Finland. I hate to see you foreigners fooled by your mobile operators, and feel sorry for you.
I enjoy playing mobile chess and reversi, and my children like Zoe's adventure. I like Morus, a in-progress multiplayer server-based medieval wargame.
You seem to be expecting wrong things about gaming. There is more gaming than just shoot-em-up.
You seem to be igonrant about the browser, just use Nokias browser or download Opera.
On your last remark on batteries, the main difference between PocketPC and phones is that phones have better battery time, with an expense of some 30-50% less CPU power.
It seems that many people are whining about these new Nokia phones, and mostly because they are not sufficiently geekish. But they are. You can develop stuff - be it games, apps, graphics, or something else - to them with free toolkits, download it over the air (from your webserver) or infrared and use. These are truely wonderful machines that obsolete your old geeky-pdas. Ok, I would feel better if they run Linux, but still I am glad that it is not Windows CE!
It is no more or less stupid than talking to any other phone, or talking in general. People who think it is stupid would have had to gasp for more air if they would have seen the first person talking to the first phone in the world, finally bursting into manic laughter, "look, how stupid!" -- or the first singer singing to a microphone, "oh, how stupid can that be!"
Nokia has had quite good success with 9110 and 9210 in some markets, and this looks like a good successor for those phones.
The format is not the problem; it is the layout algorithm. It is no longer sufficient to be able to transfer the data, but the it needs to look the same on different systems. The layout algorithms need to be standardized to a bit-perfect level in order to have true compatibility between two word processing systems. Another (perhaps less intelligent possibility) is to add layout information to the file when it is saved, and this layout info is manipulated when and if the document is manipulated in different systems, removing the need to standardize layout algorithms.
Safety and ease-of-use are interconnected. I participated in a project where we did replace an old text-based menu system (designed in the 1980's) using a tcl/tk on top of a largish medical Fortran-based simulation and planning system. I am pretty sure that the gui actually made it safer to use. The selection of tcl was done at a time when python was still at 1.3, and leaked memory like hell. Today, I would choose Python for the gui-building tool, if I would have to add a GUI to a system like that. However, even the tcl/tk conversion was a success story. The company got millions by selling the system, the gui giving a significant boost to declining sales, and the project was ready earlier! than was anticipated.
Is it really a good idea to make the most efficient part of the world be less efficient in a rather significant way, just to be able to pay a measely $100 000 000 to the aids campaing of poor countries? Problems introduced by software monopolies are in ranges of $100 000 000 000 (add more 0s here), and the monopoly tax is harming everyone, especially the poor in under-developed countries. I hope you were looking for some Funny-moderation by the reasoning included in your responce.
Ouch. Terraflops for teraflops. I will be waiting to see terrafeet (for astronomical distance measures) in US corporate press releases and popular science articles.
Future: Imagine a volunteer force of 100000000 people trying to find and collect the sensors of the previous generation, which is interfering with a newer generation of the global data collection system.
But we have been here before. Many goverments ruled that they need to have posix in their systems. Microsoft responded by building a posix subsystem that could not communicate with the rest of the system and started the posix processes with only about 80% reliability, so it could not be used. The 80% reliability worked well enough as a proof that posix was there to justify the purchase of the NT 3.1. Later it was removed, people were already buying NT 3.x, so it was not necessary.
Microsoft knows hot to set up a smokescreen of compatibility and remove it when the time comes. Most professionals will be fooled. If Microsoft is forced to set up a non-drm mode in their BIOS, it will be a 80 % quality solution. Just barely good enough to work as a proof of good will, but bad enough to stabilize Microsofts position in a monopoly of BIOS, life, computing, and everything.
Fast talk, crossfading two sentences, and not letting the consumer think, not even for a moment, are known manipulation methods used for example in tv commercials. Perhaps you are removing some important criticism processes by speed learning. If I need to keep a presentation, I try to place delays in suitable places to let the audience find things out by themselves.
We do not have dubbing here in Finland, we have subtitles. The subtitles are not only the reason for the high literacy rate in Finland, but also enable 9x fast forward for viewing this kind of quality entertainment.
In my opinion technology should not impose political decisions. Attaching the widget sets and enforcing similarity on the server side is doing just that, leading to situation where evolution (of widget sets, and windowing in general) is more difficult.
The true opportunities to improve on windowing systems lies in making the computer more like a news paper or a magazine, with proper columns, resource management for the columns, etc. Today, everyone is reading hugely wide text, leading to significant decrease in reading speed. Most of the work done with the computer is reading, and the UI should recognize this. Smaller screens should have less columns, while larger screens should allocate more. Minimization could make an article to become a headline, etc.
The paper is written around one idea, simplifying the internal structure of X. However, it is against the simplicity of X by claiming xlib's level of abstraction too low. One should start from the requirements, like the internal simplicity, and modern features. One should also specify the scope of the windowing system; mobile systems need other type of solutions from zillion-pixel worksations. Y lacks the obvious new features needed, such as subpixel cursor positioning, anti-aliasing, and 3d. IMHO, the paper is written by a confused state of mind - there is nothing shockingly good in it.
Many (if not most) scientists, who work with rats for several years, get allergic to rodents. Do not get casual with them if you have any dreams of working in a mice-lab later on in your life.