Did you read all my posts here? Do you know why Dunny get insulted?
So basically because I don't agree with you, therefore I must be too stupid to have understood the "arguments" that you have put forward.
It all boils down to a "my hobby is better than yours" mentality. You can't see the use for Sinclair BASIC (which is fair enough), but then you ridicule anyone who DOES see a use for it (whether that use is just programming for fun, or introducing kids to simple programming). I personally don't see a need for adding a hard-disk interface to my Spectrum - but I would never dream of insulting anyone who does, or dismissing you for all your work in this area.
Did you read all my posts here? Do you know why Dunny get insulted?
I think because youve been talking bollocks for so long, saying crap like "I agree. BASin or Sinclair Basic is suited for 5 year olds." and eventually winding many people up.
You dont listen to others, you just wind them up and dont seem to answer their questions.
So basically because I don't agree with you, therefore I must be too stupid to have understood the "arguments" that you have put forward...
No, here is problem that you said almost what I said. And then you talk about disagree. There is no problem who is smarter, who has better hobby or similar.
There is problem that you want it too easy - praising Speccy with childish posts and arguments, preaching nonsenses without reason - and all it after even not reading what others said. You should shame yourself.
I insulted nobody here. You insult yourself with your superficious attitude.
No, here is problem that you said almost what I said. And then you talk about disagree.
Well, let's see. I said that Sinclair BASIC has uses because people are programming in it as a hobby, and it can serve to introduce people, including children, to programming. On the other hand you say that Sinclair BASIC is entirely useless; it's only suited to introducing children to programming. How are these arguments "almost" the same?
I insulted nobody here.
It may well be that you INTENDED to insult nobody here. But the end result is that you at the very least insulted Dunny by saying that his work was a waste of time. Actions speak louder than lofty intentions.
How about taking the bickering to private messages since it has nothing to do with the topic?
Because it isn't a private discussion and because this is WoSF (Motto: Off Topic R Us). :P
Now then piters, the total sum of your previous post is: we all talk nonsense, put forward poor arguments and praise the speccy with "childish posts" while you, sir, are a shinining example of pure wit and eloquence. You sure have remarkable powers of persuasion.
.. On the other hand you say that Sinclair BASIC is entirely useless; it's only suited to introducing children to programming. How are these arguments "almost" the same?...
You again prove that not read my posts (or want not to read them, just look how to blacking me). I didn't say that it is useles, especially not entirely. Didn't said that it is 'only' for children.
Is it shame to making something for children? Or it is just not enough 'macho' ?
This thread is at least 50% total waste of time. Nothing is worth to say. If someone here not praises Speccy in all aspects, he is traitor. I didn't know that it is war state.
Now then piters, the total sum of your previous post is: we all talk nonsense, put forward poor arguments and praise the speccy with "childish posts" while you, sir, are a shinining example of pure wit and eloquence. You sure have remarkable powers of persuasion.
Why we must all time describing some people's characters? You abviously run out of arguments, and now wrote something what is already written here.
Someone said that I didn't answered to questions. I think that I was pretty fair, and did answered on most, maybe not on senseless ones (I mean who here cares about Kari's posts). Maybe people expected different answers, and therefore says that I didn't...
You can not satisfy everyone. Truth can hurt.
Why we must all time describing some people's characters? You abviously run out of arguments, and now wrote something what is already written here.
Note that the argument applies in reverse.
You can not satisfy everyone. Truth can hurt.
I suppose you speak the gospel truth for all of us?
Let's reconsider the arguments: everyone here accepts that the Sinclair BASIC is an old language which cannot meet the standards set by today's languages (OOPs or otherwise). We all already know that the Speccy doesn't have the graphical or processing powers of today's PC. So far we all agree? Good.
Now then to subjective arguments: does everyone think Sinclair BASIC is useless? No. Why? B'cos some of us find it enjoyable and are willing to find some time to program in it. Why's is this a problem for you? You find it useless so okay, no one is forcing you to code in it. Why then do you insist that we abandon it?
Yes some of us view the speccy through nostalgia tinted glasses but that's why we are here isn't it? Why deny us the small pleasures in life?
:D Your post is example of huge inobjectiveness, but it actually deserves some harder words.
That's what I get for being polite I guess...
Spectrum's conception had one very big downside in compare to PC (it was mentioned here couple times): no future compability design. What I talk? Just remember QL, last Sinclair's machine (not counting handheld ones) - is it compatible with Speccy? There is of course more thing (about it later by economy)
Maybe the QL was incompatible because it used a different CPU? Because it was targeted at a different market? Because it was not designed from the start to be an extension of the Spectrum, but a totally different machine? Actually, the QL may well have been a brilliant concept which floundered due to factors that had little to do with the design itself. Here's an excerpt that explains why it failed to capture the market:
Due to its rushed development, the QL was plagued by a number of problems from release, particularly bugs in the QDOS operating system and SuperBASIC which lead to multiple releases of the firmware. In addition, early production QLs were shipped with an external 16 KB ROM cartridge (infamously known as the "kludge" or "dongle") containing part of the firmware until the QL was redesigned to accommodate the necessary 48 KB of ROM internally, instead of the 32 KB initially specified. The QL also suffered from reliability problems of its Microdrives. These problems were later rectified, by Sinclair engineers, especially on Samsung produced models, as well as by aftermarket firms such as Adman Services and TF Services ? to the point where several QL users report their Microdrives working perfectly even after almost 17 years of service (for Samsung QLs) ? but in any case much too late to redeem the negative image they had already created.
Actually, the QL seems to be still very much alive:
The QL lives on in the form of two self-contained mainboards named Q40 and Q60 (collectively referred to as Qx0) created by Peter Graf and marketed by D & D Systems. The Q40 and Q60, being based around the 68040 and 68060 CPUs respectively, are much more powerful than the original QL and have the ability among other things (such as multimedia, high resolution graphics, Ethernet networking etc.) to run the Linux operating system.
Hardware add-ons are still being produced for the original QL mainly by TF Services who supplies the RomDisq, the Minerva replacement Operating System, I2C bus based peripherals, superHermes (keyboard interface, fast RS232 etc) and MPlane buffered backplane systems.
It seems that the "lack of development potential" argument has been shot down from the water, has it not? Anyway, You don't have to build a Spectrum compatible to make a Spectrum-like machine. It's more of a design principle as I have already tried to explain.
Don't talk me about what people building - it is irrelevant here. People builds extensions for much less popular machines like Sam Coupe, Jupiter Ace etc.
Spectrum clones? Where are they from power of one PC? It is really poor argument. Same as with SW - it is only hobby, nothing really competent.
Actually, there is no space today to compete with industry in electronic, computer build. You can not build practically nothing even close to industrial price/power ratio.
Nothing really competent? Those poor chaps, maybe they should change hobby...
for all the reasons presented before, and that you choose to ignore, why on Earth would anyone want to emulate the PC's power output, when all it does is create HUGE overheating problems?
What do you think runs your dishwasher? The supermarket bar-counters? Lots of other appliances?Is it not a low-cost embedded CPU? It gets the job done, doesn't it? You already know this.
Economy: yes, Spectrum is very economically designed. Therefore was cheap. But not so cheap that Sinclair not earn money with it. It spared on everything possible - why sound chip when CPU can generate tone? Why keyboard decoding circuit - it can do CPU aided with few diodes. Why edge slot on backside - let all IF have it's slot, we just make here hole for PCB edge... Is it really economical for user, to have on each IF slot instead one on computer?
Such conception today has no place, with modern, fast circuits. And no need for it. Chips are cheap, printed circuits are cheap. Design is much faster.
Yes, modern chips are cheap, fast and OBSCENELY ENERGY-HUNGRY. Have you noticed that energy prices are sky-rocketing?
Yes, there is a need for economical designs, and that need will become a burgeoning necessity in the not-so-distant future.
Tossing: I don't know how it is on rusted West, but here on East people not throws out PCs, even not old ones. Actually, many people buys used PCs which are bringed from West, mostly from Germany. I have one secondhand monitor from Germany, 9 years old, but it works still fine. On other side I got couple Spectrum for practically free - people said that them just collect dust.
The East is rapidly becoming as consumistic as the West. Just wait a couple of years and you'll see PCs in Eastern bins trust me.
What you talk about overheating problems is pure incorrectness. I don't know what PCs and notebooks you refer - there are certainly some low quality models on market. I even fixed couple PCs due to everheat. Bit it was failure of staff which assembled them - they put 2.8 GHz CPU in too small case without fans.
Actually, my first Speccy (Issue 2) had problems because off too small heatsink . Now you will say that it is corrected in later versions...
I don't know, I have yet to see a laptop which doesn't overheat after a couple of hours, but maybe I'm wrong about this, and maybe you can point us one model which is devoid of such problems? I'm genuinely interested.
I have access to three different Spectrum models, and I've never seen them overheat.
For end: If some concludes from this post that I hate Speccy, he is lacking something. Right because I knew Speccies downsides I was able to improve it. First thing what I done was speedup of tape load/save. I did put heatsink on ULA chip, because I want it to live many years yet...
Not at all. You obviously like the Spectrum, as you would like an old toy. Most of us don't consider it a toy, though, mainly because it isn't .
Well, I won't address these side-issues anymore. Sorry about the bickering. The thread really is about Pascal versus Sinclair Basic.
And they get even hotter when the battery explodes randomly.
Yes, I saw that news item about the Dell laptop. I do have one fear about my own laptop, not dangerous as such, but co-incidentally also battery related. After buying the best laptop I could find for the cash I could spare (I HATE HATE HATE employment agencies :x:x:x ), an IBM Thinkpad T20, I read on the forums ( http://forum.thinkpads.com/ ) that there was an "issue" with some Thinkpads, including the T20, whereby if the power was interrupted at a critical time during boot up, then the laptop would never boot up again, instead it would stop with a CRC (checksum) error message, explained with the following (copied and pasted from the above forum)
"The problem is the Atmel EEPROM, which stores CRC and supervisory passwords. That chip is accessed a few thousand times during boot-up. If [you have no battery, or the battery is dead, and] you potentially step on the power cord during POST, your Thinkpad in 95% of cases is toast.
You also get the same fatal error in rare cases when you are offline and your battery runs down to about 2% and you forgot to turn off the notebook. When you turn on again, the Thinkpad might find the voltage to low and tell you a low battery error. OR it might boot and break down 2 seconds later -> CRC Error, fatal ! "
First of all, why does the laptop write to the EPROM when booting anyway? But far more importantly, why is there no fix for this, no cure after so many revisions and models of the laptops, and why, when the laptop checks it's CRC value and detects an error, doesn't it say something like:
"Press [F1] to restore factory defaults and reboot"
so you can just get on with life? Instead, the whole motherboard is now useless, as there is no easy easy way of fixing this (IBM recommend buying a new motherboard), which is ridiculous. And this is a genuine (and very stupid) problem, I checked and the problem is real (just search google.com or other laptop forums, it's a well known complaint). So I have to make sure that my battery is always charged enough to survive a boot at least, even if the laptop is plugged into the mains.
No Spectrum, or other 8 bit computer that I know of, had such a stupid hardware fault. Then again, what can else you expect from a PC? Boot one up without a keyboard attached and you get a "keyboard not found, press [F1] to continue" message.
Sort of on topic:
I think one of the best things Atari did was to put their BASIC on a cartridge.
Upgrading the language or changing it was easy. I think they offered Pilot and Logo but I don't think they were very popular.
Sadly, Atari never really upgraded it's BASIC themselves and eventually built it into the later machines with only minor bug fixes.
A third party did enhance Atari BASIC though. BASIC XL from Optimized Systems Software offered a significant speed boost for people that liked BASIC and it shows just how nice the changable BASIC cartridge was. It ran the same programs 2 to 4 times as fast and added additional commands. They later introduced BASIC XE that added even more commands and optimized math routines. Sadly, since it wasn't an Atari product you couldn't distribute programs based on these versions of BASIC because not everyone had it.
I'm pretty sure it was eventually published in a german magazine along with a compiler but the Wiki makes no mention of it.
I think a lot of machines would have benefitted from this scheme.
The Speccy's and it's slow BASIC would have been an excellent target for this sort of system.
Now, FLASH memory or MRAM can do the same thing without the expense of a cartridge slot and the cartridge. Too bad the technology didn't arrive in time for the 8 bit PC. Actually, it's too bad a lot of technologic advancments hadn't arrived yet.
Not so on topic:
I know someone talked about the Speccy having no migration path (or something to that effect) but you have to remember that the CPUs had no migration path to use as a basis for an advanced 8 bit machine. You could add graphics modes, more sound, more RAM or whatever but you were still stuck bank switching, single digit clock speeds and couldn't deal with larger numbers. It wasn't until the 65c816 or eZ80 that you could get away from many of the limitations of cpus from that time and by then the 32 bit CPUs had taken over.
BTW, I think most BASIC compilers still depend on the ROM routines so even if you compile it a slow ROM can result in slow code. Compiling a BASIC program does let it skip the parsing and searching for line numbers/variables which are very slow.
Someone said that I didn't answered to questions. I think that I was pretty fair, and did answered on most, maybe not on senseless ones (I mean who here cares about Kari's posts).
You are so full of your own self-importance. I think you'll find a lot more people care about my posts than yours...
Going back to the original question, if Pascal has been used instead of ZX Basic what sort of programs would have been written.
Would there have been enough memory left for programs like Football Manager and Valhalla?
Going back to the original question, if Pascal has been used instead of ZX Basic what sort of programs would have been written.
Would there have been enough memory left for programs like Football Manager and Valhalla?
There isn't a simple response to the first question since we don't know. However, a compiler included with the machine should mean faster programs.
I do think there would be enough memory for programs like you mentioned.
I think the more important question is, how practical would it be to have a built in Pascal compiler?
You could have a paged ROM with compiler, editor etc... on separate ROM pages or just use CP/M and load it from a high speed storage device. Either way, you end up with a higher retail price and the Speccy is suddenly aimed at a different market.
If you were talking another language that is commonly interpreted like Pilot, Comol or Forth then you'd have the same machine but it would just be harder to port existing Basic programs to. It's not like there were books full of games for other languages.
I think it's safe to say that BASIC was probably the best choice of languages but they should have at least optimized it and made the updated version available to existing users and standard on later machines.
I find it more than a little strange that Sinclair insisted on being so cheap with the system firmware and then blew money right and left on projects that drained the company.
Yes Michael Kölling and David J. Barnes are very influential figures in the world of introductory programming . They have published several books on the approach and are the authors of BlueJ (www.BlueJ.org). The OO first approach is now adopted by a large proportion of UK universities, London Metropolitan, University of Kent and South Bank University being three of them.
It seems I spoke too soon about what's happening here as I had a look around at the curriculums at schools I attended and half of them are doing objects in the first course as well, in Java but not using the same textbook. The book you linked to is very unique in that it seems to introduce a bunch of advanced concepts very early on. Does this not confuse students? And do you get some blowback from them? I can imagine many of them not buying into it and thinking "why the $%$# should I be making up a bunch of CRC cards or draw UML diagrams". There just isn't a plain reason for it until you know something about software engineering.
I suppose I should be more careful to avoid stodgy old man syndrome (SOMS) and not dismiss this approach out of hand. One thing I can understand from the idea of introducing OO concepts early is that it's a skill that takes time to master. It could very well be that most students who are taught it later are not as comfortable with it as they could be when they graduate. But I would be very, very disappointed if students were left with the impression that OO is the only way to solve the problem of complexity; there has to be balance by getting an overview of other language types and I would hope there would be significant experience with C, eg, a language without OO support. OO encourages code reuse and library dependence, particularly with Java, which is all good in a practical setting but it can mean the student never develops an intuitive understanding of what is going on in the hardware (as C or asm would do) and doesn't get much practice with algorithm development (Java and STL for C++ do it all for you).
I still believe this is just the wrong way to go. I think much better programmers would be made by making sure they understand what is going on at the hardware level, what it means to have a pointer (in Java no one cares as it's all garbage collected), and what the compiler is doing when it performs a virtual call or whatever. The software engineering problem of how to deal with complexity, changing requirements over time, maintainability, etc is much more naturally placed after the student has some programming experience. Beforehand understanding these problems has no concrete meaning. OO is there to deal with these problems so I still think its natural place is with software engineering. OO languages like Java are another step removed from the hardware and, if that is the primary tool of education, I believe students will have less opportunity to understand what goes on at a lower level.
I hope you've noticed that I mention OO here as that's not the same as teaching classes, virtual methods, polymorphism, etc. This post is going to come across as a less-than-articulate rambling. My vote is a conditional bad idea for objects first.
Yes, modern chips are cheap, fast and OBSCENELY ENERGY-HUNGRY. Have you noticed that energy prices are sky-rocketing?
Energy prices I doubt will ever affect how energy-efficient chips are made. We are talking about sums that are easily affordable by Western consumers. If your point-of-view is a green one, well then maybe, if consumers would care enough to make it a selling point.
Energy efficiency *is* a secondary concern in desktop cpus where increasing transistor density is making the problem of removing heat more acute. Efforts are made to reduce the amount of heat generated (and therefore the amount of energy consumed). However, speed and functionality are still the primary concern. Energy efficiency is a primary design criterion for battery operated devices (like laptops) and low power chips are designed to fill this niche but this will often mean lesser performance.
I don't know, I have yet to see a laptop which doesn't overheat after a couple of hours, but maybe I'm wrong about this, and maybe you can point us one model which is devoid of such problems? I'm genuinely interested.
The problem is that some laptop manufacturers try to make laptops like they're making desktops by using desktop parts. They do this because consumers have no sense. But you can find cooler running laptops with extended battery lives by buying appropriately. For example, get a Pentium M rather than a desktop Pentium. Use a lower spec graphics card. Forget about RAID in a laptop.
I went with a laptop whose only sin is having a fairly hefty graphics card in it. It gets warm to the touch after extensive use but as long as I keep the fans clean everything's comfortable. If I'm not gaming, the laptop is always cool. It's a Dell, BTW, but not one with a faulty battery. Too bad as I need a new one right about now :-)
Comments
It all boils down to a "my hobby is better than yours" mentality. You can't see the use for Sinclair BASIC (which is fair enough), but then you ridicule anyone who DOES see a use for it (whether that use is just programming for fun, or introducing kids to simple programming). I personally don't see a need for adding a hard-disk interface to my Spectrum - but I would never dream of insulting anyone who does, or dismissing you for all your work in this area.
If you can't play nicely, then don't play at all.
I think because youve been talking bollocks for so long, saying crap like "I agree. BASin or Sinclair Basic is suited for 5 year olds." and eventually winding many people up.
You dont listen to others, you just wind them up and dont seem to answer their questions.
No, here is problem that you said almost what I said. And then you talk about disagree. There is no problem who is smarter, who has better hobby or similar.
There is problem that you want it too easy - praising Speccy with childish posts and arguments, preaching nonsenses without reason - and all it after even not reading what others said. You should shame yourself.
I insulted nobody here. You insult yourself with your superficious attitude.
Well, let's see. I said that Sinclair BASIC has uses because people are programming in it as a hobby, and it can serve to introduce people, including children, to programming. On the other hand you say that Sinclair BASIC is entirely useless; it's only suited to introducing children to programming. How are these arguments "almost" the same?
It may well be that you INTENDED to insult nobody here. But the end result is that you at the very least insulted Dunny by saying that his work was a waste of time. Actions speak louder than lofty intentions.
Because it isn't a private discussion and because this is WoSF (Motto: Off Topic R Us). :P
Now then piters, the total sum of your previous post is: we all talk nonsense, put forward poor arguments and praise the speccy with "childish posts" while you, sir, are a shinining example of pure wit and eloquence. You sure have remarkable powers of persuasion.
Bytes:Chuntey - Spectrum tech blog.
You again prove that not read my posts (or want not to read them, just look how to blacking me). I didn't say that it is useles, especially not entirely. Didn't said that it is 'only' for children.
Is it shame to making something for children? Or it is just not enough 'macho' ?
This thread is at least 50% total waste of time. Nothing is worth to say. If someone here not praises Speccy in all aspects, he is traitor. I didn't know that it is war state.
Why we must all time describing some people's characters? You abviously run out of arguments, and now wrote something what is already written here.
Someone said that I didn't answered to questions. I think that I was pretty fair, and did answered on most, maybe not on senseless ones (I mean who here cares about Kari's posts). Maybe people expected different answers, and therefore says that I didn't...
You can not satisfy everyone. Truth can hurt.
Note that the argument applies in reverse.
I suppose you speak the gospel truth for all of us?
Let's reconsider the arguments: everyone here accepts that the Sinclair BASIC is an old language which cannot meet the standards set by today's languages (OOPs or otherwise). We all already know that the Speccy doesn't have the graphical or processing powers of today's PC. So far we all agree? Good.
Now then to subjective arguments: does everyone think Sinclair BASIC is useless? No. Why? B'cos some of us find it enjoyable and are willing to find some time to program in it. Why's is this a problem for you? You find it useless so okay, no one is forcing you to code in it. Why then do you insist that we abandon it?
Yes some of us view the speccy through nostalgia tinted glasses but that's why we are here isn't it? Why deny us the small pleasures in life?
Each to his own mate. Each to his own. Amen.
Bytes:Chuntey - Spectrum tech blog.
Piters, you're a PITA.
That's what I get for being polite I guess...
Maybe the QL was incompatible because it used a different CPU? Because it was targeted at a different market? Because it was not designed from the start to be an extension of the Spectrum, but a totally different machine? Actually, the QL may well have been a brilliant concept which floundered due to factors that had little to do with the design itself. Here's an excerpt that explains why it failed to capture the market:
Actually, the QL seems to be still very much alive:
http://en.wikipedia.org/wiki/Sinclair_QL
It seems that the "lack of development potential" argument has been shot down from the water, has it not? Anyway, You don't have to build a Spectrum compatible to make a Spectrum-like machine. It's more of a design principle as I have already tried to explain.
Nothing really competent? Those poor chaps, maybe they should change hobby...
for all the reasons presented before, and that you choose to ignore, why on Earth would anyone want to emulate the PC's power output, when all it does is create HUGE overheating problems?
What do you think runs your dishwasher? The supermarket bar-counters? Lots of other appliances?Is it not a low-cost embedded CPU? It gets the job done, doesn't it? You already know this.
Yes, modern chips are cheap, fast and OBSCENELY ENERGY-HUNGRY. Have you noticed that energy prices are sky-rocketing?
Yes, there is a need for economical designs, and that need will become a burgeoning necessity in the not-so-distant future.
The East is rapidly becoming as consumistic as the West. Just wait a couple of years and you'll see PCs in Eastern bins trust me.
I don't know, I have yet to see a laptop which doesn't overheat after a couple of hours, but maybe I'm wrong about this, and maybe you can point us one model which is devoid of such problems? I'm genuinely interested.
I have access to three different Spectrum models, and I've never seen them overheat.
Not at all. You obviously like the Spectrum, as you would like an old toy. Most of us don't consider it a toy, though, mainly because it isn't .
Well, I won't address these side-issues anymore. Sorry about the bickering. The thread really is about Pascal versus Sinclair Basic.
Yes, I saw that news item about the Dell laptop. I do have one fear about my own laptop, not dangerous as such, but co-incidentally also battery related. After buying the best laptop I could find for the cash I could spare (I HATE HATE HATE employment agencies :x :x :x ), an IBM Thinkpad T20, I read on the forums ( http://forum.thinkpads.com/ ) that there was an "issue" with some Thinkpads, including the T20, whereby if the power was interrupted at a critical time during boot up, then the laptop would never boot up again, instead it would stop with a CRC (checksum) error message, explained with the following (copied and pasted from the above forum)
"The problem is the Atmel EEPROM, which stores CRC and supervisory passwords. That chip is accessed a few thousand times during boot-up. If [you have no battery, or the battery is dead, and] you potentially step on the power cord during POST, your Thinkpad in 95% of cases is toast.
You also get the same fatal error in rare cases when you are offline and your battery runs down to about 2% and you forgot to turn off the notebook. When you turn on again, the Thinkpad might find the voltage to low and tell you a low battery error. OR it might boot and break down 2 seconds later -> CRC Error, fatal ! "
First of all, why does the laptop write to the EPROM when booting anyway? But far more importantly, why is there no fix for this, no cure after so many revisions and models of the laptops, and why, when the laptop checks it's CRC value and detects an error, doesn't it say something like:
"Press [F1] to restore factory defaults and reboot"
so you can just get on with life? Instead, the whole motherboard is now useless, as there is no easy easy way of fixing this (IBM recommend buying a new motherboard), which is ridiculous. And this is a genuine (and very stupid) problem, I checked and the problem is real (just search google.com or other laptop forums, it's a well known complaint). So I have to make sure that my battery is always charged enough to survive a boot at least, even if the laptop is plugged into the mains.
No Spectrum, or other 8 bit computer that I know of, had such a stupid hardware fault. Then again, what can else you expect from a PC? Boot one up without a keyboard attached and you get a "keyboard not found, press [F1] to continue" message.
I think one of the best things Atari did was to put their BASIC on a cartridge.
Upgrading the language or changing it was easy. I think they offered Pilot and Logo but I don't think they were very popular.
Sadly, Atari never really upgraded it's BASIC themselves and eventually built it into the later machines with only minor bug fixes.
A third party did enhance Atari BASIC though. BASIC XL from Optimized Systems Software offered a significant speed boost for people that liked BASIC and it shows just how nice the changable BASIC cartridge was. It ran the same programs 2 to 4 times as fast and added additional commands. They later introduced BASIC XE that added even more commands and optimized math routines. Sadly, since it wasn't an Atari product you couldn't distribute programs based on these versions of BASIC because not everyone had it.
I'm pretty sure it was eventually published in a german magazine along with a compiler but the Wiki makes no mention of it.
I think a lot of machines would have benefitted from this scheme.
The Speccy's and it's slow BASIC would have been an excellent target for this sort of system.
Now, FLASH memory or MRAM can do the same thing without the expense of a cartridge slot and the cartridge. Too bad the technology didn't arrive in time for the 8 bit PC. Actually, it's too bad a lot of technologic advancments hadn't arrived yet.
Not so on topic:
I know someone talked about the Speccy having no migration path (or something to that effect) but you have to remember that the CPUs had no migration path to use as a basis for an advanced 8 bit machine. You could add graphics modes, more sound, more RAM or whatever but you were still stuck bank switching, single digit clock speeds and couldn't deal with larger numbers. It wasn't until the 65c816 or eZ80 that you could get away from many of the limitations of cpus from that time and by then the 32 bit CPUs had taken over.
Would there have been enough memory left for programs like Football Manager and Valhalla?
I do think there would be enough memory for programs like you mentioned.
I think the more important question is, how practical would it be to have a built in Pascal compiler?
You could have a paged ROM with compiler, editor etc... on separate ROM pages or just use CP/M and load it from a high speed storage device. Either way, you end up with a higher retail price and the Speccy is suddenly aimed at a different market.
If you were talking another language that is commonly interpreted like Pilot, Comol or Forth then you'd have the same machine but it would just be harder to port existing Basic programs to. It's not like there were books full of games for other languages.
I think it's safe to say that BASIC was probably the best choice of languages but they should have at least optimized it and made the updated version available to existing users and standard on later machines.
I find it more than a little strange that Sinclair insisted on being so cheap with the system firmware and then blew money right and left on projects that drained the company.
It seems I spoke too soon about what's happening here as I had a look around at the curriculums at schools I attended and half of them are doing objects in the first course as well, in Java but not using the same textbook. The book you linked to is very unique in that it seems to introduce a bunch of advanced concepts very early on. Does this not confuse students? And do you get some blowback from them? I can imagine many of them not buying into it and thinking "why the $%$# should I be making up a bunch of CRC cards or draw UML diagrams". There just isn't a plain reason for it until you know something about software engineering.
I suppose I should be more careful to avoid stodgy old man syndrome (SOMS) and not dismiss this approach out of hand. One thing I can understand from the idea of introducing OO concepts early is that it's a skill that takes time to master. It could very well be that most students who are taught it later are not as comfortable with it as they could be when they graduate. But I would be very, very disappointed if students were left with the impression that OO is the only way to solve the problem of complexity; there has to be balance by getting an overview of other language types and I would hope there would be significant experience with C, eg, a language without OO support. OO encourages code reuse and library dependence, particularly with Java, which is all good in a practical setting but it can mean the student never develops an intuitive understanding of what is going on in the hardware (as C or asm would do) and doesn't get much practice with algorithm development (Java and STL for C++ do it all for you).
I still believe this is just the wrong way to go. I think much better programmers would be made by making sure they understand what is going on at the hardware level, what it means to have a pointer (in Java no one cares as it's all garbage collected), and what the compiler is doing when it performs a virtual call or whatever. The software engineering problem of how to deal with complexity, changing requirements over time, maintainability, etc is much more naturally placed after the student has some programming experience. Beforehand understanding these problems has no concrete meaning. OO is there to deal with these problems so I still think its natural place is with software engineering. OO languages like Java are another step removed from the hardware and, if that is the primary tool of education, I believe students will have less opportunity to understand what goes on at a lower level.
I hope you've noticed that I mention OO here as that's not the same as teaching classes, virtual methods, polymorphism, etc. This post is going to come across as a less-than-articulate rambling. My vote is a conditional bad idea for objects first.
Write games in C using Z88DK and SP1
Energy prices I doubt will ever affect how energy-efficient chips are made. We are talking about sums that are easily affordable by Western consumers. If your point-of-view is a green one, well then maybe, if consumers would care enough to make it a selling point.
Energy efficiency *is* a secondary concern in desktop cpus where increasing transistor density is making the problem of removing heat more acute. Efforts are made to reduce the amount of heat generated (and therefore the amount of energy consumed). However, speed and functionality are still the primary concern. Energy efficiency is a primary design criterion for battery operated devices (like laptops) and low power chips are designed to fill this niche but this will often mean lesser performance.
The problem is that some laptop manufacturers try to make laptops like they're making desktops by using desktop parts. They do this because consumers have no sense. But you can find cooler running laptops with extended battery lives by buying appropriately. For example, get a Pentium M rather than a desktop Pentium. Use a lower spec graphics card. Forget about RAID in a laptop.
I went with a laptop whose only sin is having a fairly hefty graphics card in it. It gets warm to the touch after extensive use but as long as I keep the fans clean everything's comfortable. If I'm not gaming, the laptop is always cool. It's a Dell, BTW, but not one with a faulty battery. Too bad as I need a new one right about now :-)
Write games in C using Z88DK and SP1