The first production run of the 4004 was in December 1970. Admittably the production run had to be tossed due to mask errors, but 2nd and 3rd production runs in Jan and Feb of 71 were more sucessful (the 2nd run still had errors). Sample calculator designs were shipped to Busicom in March 71 - comprising 4 4001s, 2 4002s, 2 4003s and 1 4001.
The only relevance of November 71 that I can find, was that the MCS-4 microcomputer based on the 400x series was released. But thats not the microprocessor itself.
One thing that stands out, is that Intel have had production problems and bugs since day 1:)
Should be easily synthesizable though. Opencores doesn't have one, but if someone can synthesize an ARM core, surely someone somewhere has synthesized the 4004.
I did a quick search, but it seems as if every VHDL engineer in the country has a phone number ending in 4004, so the search results weren't too easily navigatable:)
Re:Why the bloody hell does the release day matter
on
Gamecube Hits US Early
·
· Score: 2, Insightful
The ideas is almost certainly to catch X-box buyer's eyes as they go into the store on thursday. They'll get there with their $300 for X-box, and see the cube on sale, and think 'oh, the cube was released early, and its $100 less'. It won't catch many buyers that wouldn't have bought the cube anyway, but it might catch a few.
Also, you have the 'sold out of Xbox, but still have cubes left' syndrome, where people may go looking for an X-box, only to be told 'we don't have any left, but we have a few gamecubes', its unlikely to happen, since IIRC Xbox has a slightly larger initial unit allocation, but again, it might happen.
Also, its no big deal for nintendo to do this really, its not like they need to ship out the cubes on a different schedule, they're just telling retailers to ignore the release date when they receive them and sell them as soon as they want.
And last, but not least, it generates a little extra press attention, much like when Sega started selling the saturn 3 months early back in '95.
Re:1 week battery life and integrated fast etherne
on
Rolling Your Own Laptop?
·
· Score: 2, Insightful
Most power blocks for laptops ARE international. All you need is the right cable from power outlet to the block, usually its a tele-funken (looks like a figure of 8). Some MAY use IEC connectors (the same as on desktop PCs), which again is a univeral connector.
When I took my laptop to the UK, I had no problem at all running it off 240V/50Hz, I just had to buy a 3 prong UK->tele-funken cable (not a problem, they're fairly univeral).
Japan shouldn't be a problem either, since the connector is used on a variety of game consoles and other devices.
Alternatively, you can buy connector adapters (without step-down/up transformers) that are designed to convert one mains voltage connector to another, for devices that auto-sense.
Ability to use AC power in a different country is obviously something that the laptop manufacturers thought about, so I wouldn't say extremely long battery life is a real need just for international travel:).
I was looking at the new Dell laptops just the other day. The high end models (1600x1280 LCD and P3-1.13GHz mmmmm), have 3 drive bays, 1 CD or DVD, 1 HDD, and the 3rd can be used for: DVD-R Superdrive, CD-RW, 2nd HDD or a second battery. I don't see any reason why you couldn't run software RAID on 2 48GB drives on it.
Remember that most laptops have at least a VGA output as well, so while the advantage may not be so big on the LCD, carrying your laptop to someone that has a spare monitor for a LAN party or such, is still useful, and should let you choose 75Hz (or higher) refresh. Ditto with office presentations - often the laptop is run into a projector that accepts VGA connections, so if you have a need for accelerated 3d in presentations, you're not limited by the LCD's abilities.
Also as its been said, there is the advantage of having a sustainable performance level for those heavy scenes.
The other big problem is lack of upgrade path. The new nvidia mobile chipset is great, but other than buying a new laptop, there is no way for me to get it in my toshiba.
As such, your friend has the advantage now, but unless he buys a new laptop in the mean time, come Doom 3 (and other more intense titles) he's going to be left behind. (As I already am with the SavageIX in my Toshiba - it just can't run most modern games, despite the 650Mhz P3)
And honestly, I don't think it will happen anytime in the near future. The way layering works is naturally suited to a 'pressed write, focusable laser read', but not to the process of writing a disk at home.
Pressed DVD's are basically produced as 2 disk surfaces, which are then fixed together.
Reading dual-layer disks is fundamentally different to writing them. An analogy would be to take 2 sheets of tracing paper, if you punch holes in them with a pencil, then put them together, looking through the sheets you can see both layers. Now try to put them together first, and punch holes in the 'bottom' layer using a pencil, it can't be done without punching a hole in the top layer, no matter how fancy a pencil you use. Writing dual-layer DVD's will likely prove to be a similar problem.
If there IS ever a solution, it will likely involve a different way of approaching the problem than using a plain DVD compatible disk. Something akin to how MO disks work - ie the writing process is a combination of a magnetic field AND a laser. Even then, its going to be a tough problem to solve, IMO.
For the average modem user, V92 doesn't offer much. The only real benefit is the fast-connect, and that doesn't work all that often from what I've seen of the technical details.
The 'faster uploads' is a misdirection if you're an average user, you have to sacrifice download speed to get the upload speed, so while you can have 50-56k uploads, it means going back to 33.6k downloads. There MAY be a reasonable compromise somewhere in between (there are several settings for Down:Up ratio), but it would take trial and error, and you'd probably spend a lot of time worrying about whether you could be getting a faster download.
Also, as another poster mentioned, many ISPs don't support it yet. Back in July/August I flashed a V90 3com to V92, it was the worst move I made, the new firmware dropped my download rate from 50.6k every time, to
Of course, any external V92 modem will work fine with linux, the differences are only a handful of +P prefix AT commands.
You know, when I wrote the post, I actually sat and thought 'but, how do SimCity and Civ fit into what I'm saying, they mimic life - to a degree - but they are certainly works of arts', in the end, I thought to myself, well, there are exceptions to every rule.
After your reply, I kind of wish I'd clarified that, but thats life. You're certainly right that Civ is art, possibly even the mona lisa of the game world - I wouldn't be surprised to see Civ in a museaum as an example of the brilliance of 'those early days', anyway:)
I wasn't in any way trying to imply that you need a story to be 'art'. My point was simply that a story-based game such as an RPG or RTS is inherently art because of the story, but other game styles can still be art through other means.
As I said further down in the post, I feel that FPS and other titles where the environment and character/monster design are well thought out also qualify as 'art'. The only game styles that I'd NOT consider art are those that are pure simulations of real-world environments and characters, such as most sports titles. They can still be great games, but they are entertainment instead of art. Although there may be artistic elements to the UI of the game, much of the actual game is centered around something that didn't really require any creativity to implement (real sports personalities, team data, etc).
Well, I said 'and little room for direct innovation', I don't deny that there is innovation in apps every now and then, but for games its *almost* a nessecity these days. Especially with everyone expecting new levels in eye candy with each new title.
Of course, the target hardware is often a faster moving target too, so in a sense that drives a demand for innovation too - look how few 2D games are sold these days. The majority of consumers expect games to make full use of the latest, greatest, fastest video card, regardless if it actually involves gameplay being improved.
As someone that has programmed in both areas, I can say that Games programming definately requires a different approach to problems. With App programming you occasionally run into a need to use a similar method of lateral thinking to solving a design reguirement, but with games the requirement for it is much more intense.
But programming is more of a craft. It's about making something well.
Bear in mind that programming apps and programming games can be very different tasks.
Games programming CAN be an art, it depends on the programmer. In the past, performance was always such an issue that artistic approaches to solving a specific problem were a nessecity. These days that isn't so much of an issue, but there are still areas where an 'artistic' attitude is beneficial. Still, even with 3D acceleration and fast CPUs, there are many areas where you have to start thinking 'outside the box'.
As an example, shadows aren't automaticly generated, so when people first started with shadows, they went with the tried and tested approach of a translucent circle rendered flat on the ground. People started trying to come up with more innovative methods of achieving the effect, including pre-rendered outlines, lower-detail rendering, and the standard trick thats used these days (in OpenGL based games at least) which is to use the stencil buffer.
General programming (Apps, etc) generally starts with a goal, and there are distinct steps to take to achieve that goal, usually there are clearly documented methods to get to that goal, and little room for direct innovation. Games programming can start with a storyboard and a fairly basic design document that usually outlines the overall issues. Graphic and gameplay features are often introduced as ideas along the way, ideas that may not immediately have any established steps to achieve it. The programmers then either say that it can't be done, or they sit down and try and find some way to achieve the seemingly impossible. THAT is an art.
First, let me say that I'm a games developer, so perhaps I'm a little biased here..
Now, the way I see it is this, SOME games are art, some are simply entertainment. Ok, 'thats obvious' you might be thinking, however, WHERE the line is may not be.
The way I see it, a baseball/other-sport simulations is simply entertainment. It might have artistic elements (presentation issues usually), but overall it is entertainment.
Games such as RPGs, RTS, etc are obviously art, since a storyline is often involved, and I'm pretty sure that most people will agree that story telling is an artform.
As for games such as Quake, UT, etc. Yes, they are art, IMO, and heres why... These games are not based around existing concepts from real life, oh sure combat exists, but the environments, the character/monster design and other such issues all require a form of story telling. The story may not be a linear form of plotline, but there is still some story telling involved in designing 'scary monsters', or an alien landscape.
These elements require someone to put thought into telling the story, whether its via a plotline, or through the environment and creatures inhabiting it.
Anyway, thats the way I see things on this topic:)
"Mission Time" and "time to get there" aren't nessecarily the same things.
The original plan of the Pluto-Kuiper Express involved mission time beyond the fly-by with pluto. So the 10-12 years is likely including time AFTER its pluto encounter. Hence it would likely be 8-9 years to pluto, then 2-3 years beyond that.
You could use mkisofs via linux, it will accept a boot disk IMAGE for el-torito bootable CD's.
If you're not a linux user, well, there may be a port of mkisofs for windows somewhere, in fact I'd be surprised if there isn't. Of course, running a tiny linux partition for just these tasks would probably be better - you could use the loopback driver to mount disk images as 'virtual floppies' to let you build the disk images with the right autoexec.bat/config.sys files and add your firmware updaters. You wouldn't need a HUGE linux partition, just a few MB, enough for a fairly basic kernel + cdrtools, no need for any fancy GUI stuff.
There may be some way to manipulate FAT disk images under windows, though I've never come across such a tool, though I've never had a need to be honest.
As I said, the problem isn't GPL. Its trolltech's license.
To illustrate, the following is from their FAQ:
Q: Can we use the Free Edition while developing our non-free application and then purchase commercial licenses when we start to sell it?
A: No. The Free Edition license applies to the development phase - anything developed without Professional or Enterprise Edition licenses must be released as free/open source software.
As you can see, its their license that forbids anyone from using QT/Free on a project that MAY be commercial at some point.
This not only restricts small dev houses that can't afford to buy $2000 license's 'on a whim' to test, but it also affects hobbyists. There have been several projects that started out as small freeware/open-source projects, and later gained a level of functionality that made the authors think 'Maybe I should charge (a shareware fee) for this'
Qt/Free (or KDE) programmers in that situation will find themselves staring at a nightmare, they either start rewriting fresh code, without using any of the original source - if that is even a legal defense against such a license (I'm not a lawyer, so I don't know), or they take the chance and risk Trolltech suing them.
Now, they DO have a 'Qt/Windows Evaluation version' which doesn't appear to lock you into the open-source license model, and is in fact a fair product - you get a 30 day trial of the software, a little short for evaluating an API, but fair enough. There is no QT/X11 evaluation though, only QT/Free and the Professional/Enterprise versions.
Because Trolltech's license expressly forbids that. Its in the FAQ *twice*. The license you choose to DEVELOP against determines the license your product can be distributed as. Ie, use the QT/Free license ANYWHERE in the development of a product, and it HAS to be GPL when finally released.
This isn't the 'GPL is viral', personally, I respect the GPL and feel that it has a very valid and important place in the market, as does LGPL. The problem here is Trolltech's thought that whatever you use on your development sources 'infects' the final distribution license.
If they allowed you to develop with QT/Free, then say 'ok, here's my $1550', get QT/Pro and relink. it would be ideal. But as it stands, you cannot do that. I suppose the chances that they would sue you over it, or be able to prove in court that you used QT/Free on internal company builds would be difficult, but its a risk I don't want to take.
This isn't a bashing of GPL (its not even a GPL issue, as far as I can tell), its a bashing of Trolltech's shortsightedness - not being able to evaluate QT for a commercial/non-OSS project without outlaying cash before-hand HAS to be costing them customers.
Most likely they just don't care, and the political statement is more important than marketshare for their product. After all, whether a potential customer pays up front, or before release, makes no real difference to them, they'd still have a customer either way.
Well, I'm not an 'admin' anymore, other than my personal systems. However, it all started for me while studying EE at uni. The ultix systems were the easiest to develop on, although the VMS systems were nice for 'real work'.
So I got used to developing/using unix back in the late 80s, then I switched to a CS degree, grabbed linux for my PC in the early days (around the time of 0.13/0.96) and got used to using unix systems all the time.
Of course, leaving university with a mix of admin ability and development ability meant that finding jobs that needed a mix of both - many companies with smaller (20-30 workstation) unix nets don't need a fulltime admin, but someone that can act as a developer 4/5 of the time and admin a couple of hours a day is ideal.
The biggest problem for me, and I suspect others, is that you can't even evaluate whether Qt is suitable for your app, without paying $1550 or $2000 (the professional license is $1550).
The reason being, that the moment any of your project touches Qt/free, your app is committed to being open-source/free. This might be ok for the bigger app vendors, where laying out $1550 to evaluate the prospect of porting your app to Qt is ok. But for a small software company, its rather crippling.
I'm not quite sure why Trolltech act this way, what is so wrong about allowing Qt/Free to evaluate a product before purchasing a license for the pro/enterprise products ? Surely making life easier for smaller dev. houses to use their product would be in their interests ?
Oh well, I'll stick with gnome/gtk/gtkmm for my framework. At least until RMS insists that they are all GPL instead of LGPL and hence no longer usable in a non-oss model.
The worst is those Flash ad's that appear over the content.
Yesterday I was at IGN's site, when one appeared that dropped $X.XX prices down the screen for about 2 minutes, gradually covering the page I was trying to read until the window was full, then the 'close ad' button appeared. Most annoying, and eerily reminisant of several DOS viruses that appeared around 1989/1990.
That was the last straw for me with IGN, after the 'you have to become an Insider to read today's headlines' routine over the past week or two, now the ads are getting worse (they've run the flash ads before, but never as bad as that one), so I made a note that I'll not be using IGN as my source of gaming info/news from now on.
The 32bit Addressing limit isn't 'really' there though.
True, you have to use CPU extensions in code, but all processors since the PPro have had > 32 bit address buses. IIRC, the P3's support something like 64GB of physical ram, the Xeon processors support more (1TB iirc - 40bit).
The problem, of course, is the chipsets/motherboard manufacturers that fail to make use of the wider address bus.
Slashdot Mirror
According to The Chronology of Personal Computers (1969-1971):
The first production run of the 4004 was in December 1970. Admittably the production run had to be tossed due to mask errors, but 2nd and 3rd production runs in Jan and Feb of 71 were more sucessful (the 2nd run still had errors). Sample calculator designs were shipped to Busicom in March 71 - comprising 4 4001s, 2 4002s, 2 4003s and 1 4001.
The only relevance of November 71 that I can find, was that the MCS-4 microcomputer based on the 400x series was released. But thats not the microprocessor itself.
One thing that stands out, is that Intel have had production problems and bugs since day 1 :)
I did a quick search, but it seems as if every VHDL engineer in the country has a phone number ending in 4004, so the search results weren't too easily navigatable :)
Also, you have the 'sold out of Xbox, but still have cubes left' syndrome, where people may go looking for an X-box, only to be told 'we don't have any left, but we have a few gamecubes', its unlikely to happen, since IIRC Xbox has a slightly larger initial unit allocation, but again, it might happen.
Also, its no big deal for nintendo to do this really, its not like they need to ship out the cubes on a different schedule, they're just telling retailers to ignore the release date when they receive them and sell them as soon as they want.
And last, but not least, it generates a little extra press attention, much like when Sega started selling the saturn 3 months early back in '95.
When I took my laptop to the UK, I had no problem at all running it off 240V/50Hz, I just had to buy a 3 prong UK->tele-funken cable (not a problem, they're fairly univeral).
Japan shouldn't be a problem either, since the connector is used on a variety of game consoles and other devices.
Alternatively, you can buy connector adapters (without step-down/up transformers) that are designed to convert one mains voltage connector to another, for devices that auto-sense.
Ability to use AC power in a different country is obviously something that the laptop manufacturers thought about, so I wouldn't say extremely long battery life is a real need just for international travel:).
I was looking at the new Dell laptops just the other day. The high end models (1600x1280 LCD and P3-1.13GHz mmmmm), have 3 drive bays, 1 CD or DVD, 1 HDD, and the 3rd can be used for: DVD-R Superdrive, CD-RW, 2nd HDD or a second battery. I don't see any reason why you couldn't run software RAID on 2 48GB drives on it.
Also as its been said, there is the advantage of having a sustainable performance level for those heavy scenes.
As such, your friend has the advantage now, but unless he buys a new laptop in the mean time, come Doom 3 (and other more intense titles) he's going to be left behind. (As I already am with the SavageIX in my Toshiba - it just can't run most modern games, despite the 650Mhz P3)
And honestly, I don't think it will happen anytime in the near future. The way layering works is naturally suited to a 'pressed write, focusable laser read', but not to the process of writing a disk at home.
Pressed DVD's are basically produced as 2 disk surfaces, which are then fixed together.
Reading dual-layer disks is fundamentally different to writing them. An analogy would be to take 2 sheets of tracing paper, if you punch holes in them with a pencil, then put them together, looking through the sheets you can see both layers. Now try to put them together first, and punch holes in the 'bottom' layer using a pencil, it can't be done without punching a hole in the top layer, no matter how fancy a pencil you use. Writing dual-layer DVD's will likely prove to be a similar problem.
If there IS ever a solution, it will likely involve a different way of approaching the problem than using a plain DVD compatible disk. Something akin to how MO disks work - ie the writing process is a combination of a magnetic field AND a laser. Even then, its going to be a tough problem to solve, IMO.
The 'faster uploads' is a misdirection if you're an average user, you have to sacrifice download speed to get the upload speed, so while you can have 50-56k uploads, it means going back to 33.6k downloads. There MAY be a reasonable compromise somewhere in between (there are several settings for Down:Up ratio), but it would take trial and error, and you'd probably spend a lot of time worrying about whether you could be getting a faster download.
Also, as another poster mentioned, many ISPs don't support it yet. Back in July/August I flashed a V90 3com to V92, it was the worst move I made, the new firmware dropped my download rate from 50.6k every time, to
Of course, any external V92 modem will work fine with linux, the differences are only a handful of +P prefix AT commands.
After your reply, I kind of wish I'd clarified that, but thats life. You're certainly right that Civ is art, possibly even the mona lisa of the game world - I wouldn't be surprised to see Civ in a museaum as an example of the brilliance of 'those early days', anyway :)
I wasn't in any way trying to imply that you need a story to be 'art'. My point was simply that a story-based game such as an RPG or RTS is inherently art because of the story, but other game styles can still be art through other means.
As I said further down in the post, I feel that FPS and other titles where the environment and character/monster design are well thought out also qualify as 'art'. The only game styles that I'd NOT consider art are those that are pure simulations of real-world environments and characters, such as most sports titles. They can still be great games, but they are entertainment instead of art. Although there may be artistic elements to the UI of the game, much of the actual game is centered around something that didn't really require any creativity to implement (real sports personalities, team data, etc).
Of course, the target hardware is often a faster moving target too, so in a sense that drives a demand for innovation too - look how few 2D games are sold these days. The majority of consumers expect games to make full use of the latest, greatest, fastest video card, regardless if it actually involves gameplay being improved.
As someone that has programmed in both areas, I can say that Games programming definately requires a different approach to problems. With App programming you occasionally run into a need to use a similar method of lateral thinking to solving a design reguirement, but with games the requirement for it is much more intense.
Bear in mind that programming apps and programming games can be very different tasks.
Games programming CAN be an art, it depends on the programmer. In the past, performance was always such an issue that artistic approaches to solving a specific problem were a nessecity. These days that isn't so much of an issue, but there are still areas where an 'artistic' attitude is beneficial. Still, even with 3D acceleration and fast CPUs, there are many areas where you have to start thinking 'outside the box'.
As an example, shadows aren't automaticly generated, so when people first started with shadows, they went with the tried and tested approach of a translucent circle rendered flat on the ground. People started trying to come up with more innovative methods of achieving the effect, including pre-rendered outlines, lower-detail rendering, and the standard trick thats used these days (in OpenGL based games at least) which is to use the stencil buffer.
General programming (Apps, etc) generally starts with a goal, and there are distinct steps to take to achieve that goal, usually there are clearly documented methods to get to that goal, and little room for direct innovation. Games programming can start with a storyboard and a fairly basic design document that usually outlines the overall issues. Graphic and gameplay features are often introduced as ideas along the way, ideas that may not immediately have any established steps to achieve it. The programmers then either say that it can't be done, or they sit down and try and find some way to achieve the seemingly impossible. THAT is an art.
First, let me say that I'm a games developer, so perhaps I'm a little biased here..
:)
Now, the way I see it is this, SOME games are art, some are simply entertainment. Ok, 'thats obvious' you might be thinking, however, WHERE the line is may not be.
The way I see it, a baseball/other-sport simulations is simply entertainment. It might have artistic elements (presentation issues usually), but overall it is entertainment.
Games such as RPGs, RTS, etc are obviously art, since a storyline is often involved, and I'm pretty sure that most people will agree that story telling is an artform.
As for games such as Quake, UT, etc. Yes, they are art, IMO, and heres why... These games are not based around existing concepts from real life, oh sure combat exists, but the environments, the character/monster design and other such issues all require a form of story telling. The story may not be a linear form of plotline, but there is still some story telling involved in designing 'scary monsters', or an alien landscape.
These elements require someone to put thought into telling the story, whether its via a plotline, or through the environment and creatures inhabiting it.
Anyway, thats the way I see things on this topic
"Mission Time" and "time to get there" aren't nessecarily the same things.
The original plan of the Pluto-Kuiper Express involved mission time beyond the fly-by with pluto. So the 10-12 years is likely including time AFTER its pluto encounter. Hence it would likely be 8-9 years to pluto, then 2-3 years beyond that.
You could use mkisofs via linux, it will accept a boot disk IMAGE for el-torito bootable CD's.
If you're not a linux user, well, there may be a port of mkisofs for windows somewhere, in fact I'd be surprised if there isn't. Of course, running a tiny linux partition for just these tasks would probably be better - you could use the loopback driver to mount disk images as 'virtual floppies' to let you build the disk images with the right autoexec.bat/config.sys files and add your firmware updaters. You wouldn't need a HUGE linux partition, just a few MB, enough for a fairly basic kernel + cdrtools, no need for any fancy GUI stuff.
There may be some way to manipulate FAT disk images under windows, though I've never come across such a tool, though I've never had a need to be honest.
To illustrate, the following is from their FAQ:
Q: Can we use the Free Edition while developing our non-free application and then purchase commercial licenses when we start to sell it?
A: No. The Free Edition license applies to the development phase - anything developed without Professional or Enterprise Edition licenses must be released as free/open source software.
As you can see, its their license that forbids anyone from using QT/Free on a project that MAY be commercial at some point.
This not only restricts small dev houses that can't afford to buy $2000 license's 'on a whim' to test, but it also affects hobbyists. There have been several projects that started out as small freeware/open-source projects, and later gained a level of functionality that made the authors think 'Maybe I should charge (a shareware fee) for this'
Qt/Free (or KDE) programmers in that situation will find themselves staring at a nightmare, they either start rewriting fresh code, without using any of the original source - if that is even a legal defense against such a license (I'm not a lawyer, so I don't know), or they take the chance and risk Trolltech suing them.
Now, they DO have a 'Qt/Windows Evaluation version' which doesn't appear to lock you into the open-source license model, and is in fact a fair product - you get a 30 day trial of the software, a little short for evaluating an API, but fair enough. There is no QT/X11 evaluation though, only QT/Free and the Professional/Enterprise versions.
Because Trolltech's license expressly forbids that. Its in the FAQ *twice*. The license you choose to DEVELOP against determines the license your product can be distributed as. Ie, use the QT/Free license ANYWHERE in the development of a product, and it HAS to be GPL when finally released.
This isn't the 'GPL is viral', personally, I respect the GPL and feel that it has a very valid and important place in the market, as does LGPL. The problem here is Trolltech's thought that whatever you use on your development sources 'infects' the final distribution license.
If they allowed you to develop with QT/Free, then say 'ok, here's my $1550', get QT/Pro and relink. it would be ideal. But as it stands, you cannot do that. I suppose the chances that they would sue you over it, or be able to prove in court that you used QT/Free on internal company builds would be difficult, but its a risk I don't want to take.
This isn't a bashing of GPL (its not even a GPL issue, as far as I can tell), its a bashing of Trolltech's shortsightedness - not being able to evaluate QT for a commercial/non-OSS project without outlaying cash before-hand HAS to be costing them customers.
Most likely they just don't care, and the political statement is more important than marketshare for their product. After all, whether a potential customer pays up front, or before release, makes no real difference to them, they'd still have a customer either way.
Well, I'm not an 'admin' anymore, other than my personal systems. However, it all started for me while studying EE at uni. The ultix systems were the easiest to develop on, although the VMS systems were nice for 'real work'.
So I got used to developing/using unix back in the late 80s, then I switched to a CS degree, grabbed linux for my PC in the early days (around the time of 0.13/0.96) and got used to using unix systems all the time.
Of course, leaving university with a mix of admin ability and development ability meant that finding jobs that needed a mix of both - many companies with smaller (20-30 workstation) unix nets don't need a fulltime admin, but someone that can act as a developer 4/5 of the time and admin a couple of hours a day is ideal.
The biggest problem for me, and I suspect others, is that you can't even evaluate whether Qt is suitable for your app, without paying $1550 or $2000 (the professional license is $1550). The reason being, that the moment any of your project touches Qt/free, your app is committed to being open-source/free. This might be ok for the bigger app vendors, where laying out $1550 to evaluate the prospect of porting your app to Qt is ok. But for a small software company, its rather crippling. I'm not quite sure why Trolltech act this way, what is so wrong about allowing Qt/Free to evaluate a product before purchasing a license for the pro/enterprise products ? Surely making life easier for smaller dev. houses to use their product would be in their interests ? Oh well, I'll stick with gnome/gtk/gtkmm for my framework. At least until RMS insists that they are all GPL instead of LGPL and hence no longer usable in a non-oss model.
Actually, we CAN see it over here :)
:)
www.adcritic.com hosts it, needs quicktime but oh well, it IS funny
The worst is those Flash ad's that appear over the content.
Yesterday I was at IGN's site, when one appeared that dropped $X.XX prices down the screen for about 2 minutes, gradually covering the page I was trying to read until the window was full, then the 'close ad' button appeared. Most annoying, and eerily reminisant of several DOS viruses that appeared around 1989/1990.
That was the last straw for me with IGN, after the 'you have to become an Insider to read today's headlines' routine over the past week or two, now the ads are getting worse (they've run the flash ads before, but never as bad as that one), so I made a note that I'll not be using IGN as my source of gaming info/news from now on.
The 32bit Addressing limit isn't 'really' there though.
True, you have to use CPU extensions in code, but all processors since the PPro have had > 32 bit address buses. IIRC, the P3's support something like 64GB of physical ram, the Xeon processors support more (1TB iirc - 40bit).
The problem, of course, is the chipsets/motherboard manufacturers that fail to make use of the wider address bus.