I personally don'T like using fully assembler in GUI based systems like Windows. They tend to not very user friendly, lack of flexibility and serious cosmetic problems.
Assembly is fast, ok, but C is also compiles well enough today and easy to maintain.
It's better to get modern :) Eg. Python, even sid meier using it nowadays :P
I personally don'T like using fully assembler in GUI based systems like Windows. They tend to not very user friendly, lack of flexibility and serious cosmetic problems.
Assembly is fast, ok, but C is also compiles well enough today and easy to maintain.
It's better to get modern :) Eg. Python, even sid meier using it nowadays :P
Well, I can't use Python because it's interpreted and there won't be a python interpreter available for the next project. Assembly is just fine, because it won't run on Windows either :-)
eheh, linux. I still say python, easy to port, you can still keep woody's asm core in a dll.
I suspect you are gonna create unusual emulator, capable of strange abilities.
oh, wait.. "there won't be a python interpreter available for the next project." ? Spin for GP2x? Spin for SymbianOS? Spin for ZXspectrum 128k? Spin for Windows3.1? I'll get my coat.
I suspect you are gonna create unusual emulator, capable of strange abilities.
Nope.
oh, wait.. "there won't be a python interpreter available for the next project." ? Spin for GP2x? Spin for SymbianOS? Spin for ZXspectrum 128k? Spin for Windows3.1? I'll get my coat.
Dunny, I remember you as being a total delphi believer, what happened to you? but you are on the right track anyway. Even vb is better than delphi.
I've been using Delphi for years and I can't see how you can say VB is better. Do you want to backup your claim by giving a few examples of how VB is better ? I haven't seen many VB Spectrum emulators about.
I've been using Delphi for years and I can't see how you can say VB is better. Do you want to backup your claim by giving a few examples of how VB is better ? I haven't seen many VB Spectrum emulators about.
There's vbSpec, for one. And wasn't there a vb AY player too?
Not sure how that compares to the number of Delphi emulators, mind.
As for which language is best - it depends on the task and what the programmer feels comfortable with. Personally I'm a C++ man. But that's just me.
Oh, and as for Dunny saying the next version of Spin and BASin will be in asm... it's obvious he's going to be writing them for the ZX. Hell yeah!
dephi vs. vb is offtopic for this thread. I'm not saying VB is perfect. I said, "even vb is better" with a little sarcasm.
and it's a personal preference ofcourse, but I'm strongly against using pascal, if you want to use RAD, I prefer basic (maybe because I'm a specchum), if I need more power, I use ansi C. Pascal is in-between for me and it's pretty useles in my point of view. Yes, it's a good starter language and it's faster than basic. But --hey, I dont want to use "==" and ";"s in my RAD.
And for spin, I really like spin the way it is, this is the first project that uses delphi for good.
it's obvious he's going to be writing them for the ZX.
nope, few post ago, he said no to ZX.
But Dunny, I really don't care much for spin it's already complete and working. But basin needs maintennance and it must run on windows and/or linux, come on! if not, make it opensource then!
dephi vs. vb is offtopic for this thread. I'm not saying VB is perfect. I said, "even vb is better" with a little sarcasm.
VB is very much inferior to Delphi, in pretty much every way possible. They've even managed to corrupt BASIC so much that even Delphi's corrupted Pascal is better.
Basin needs maintennance and it must run on windows and/or linux, come on! if not, make it opensource then!
Um. Have you never actually downloaded BASin then?
... Pascal is in-between for me and it's pretty useles in my point of view.
I know this is way off topic but I just wanted to point out that my Delphi applications control the swipe card access to some of the biggest companies in the world. This involves database access, networking, XML parsing, NT services and user interfaces all written in Delphi (some of it still in 16bit !). I think you underestimate the power and RAD of Delphi.
Um. Have you never actually downloaded BASin then?
? I'm using basin regularly. But you said that basin will not be available in windows anymore and you know it's bugs more than I. it's the most needed program for a retro-hobbyist. it will be good to see basin still maintained for windows in the future.
he next "evolution" of the SPIN/BASin saga will be written 100% in assembler... ...assembly is just fine, because it won't run on Windows either
it's your program, we all are thankful to you. the sentence I said in my previous post about open source, was my wish actually. I don't want to see the source, I only want to see new version of basin :)
And Mr. Millside, I respect your thoughts, and they are true. it's just my preference, I know pascal > basic.
Of course, another possibility for the future of Spin (if it's not going to be a Windows thing) would be to make it the OS of your machine... kinda like ZX Emulator 3 years ago.
Of course, another possibility for the future of Spin (if it's not going to be a Windows thing) would be to make it the OS of your machine... kinda like ZX Emulator 3 years ago.
Crikey, that's close, but no cigar :-)
Okay, I'll come clean. Please bear in mind that this is not even started, let alone finalised in design yet. The groundwork has been laid, but it's still in planning. The idea is to write a new OS for x86 PCs. It's not a competing OS for windows or linux - it's not designed to be.
I'm going to port the Sinclair Spectrum ROM to 32bit x86 assembler. It's mostly so I can find a use for that old 486 laptop of mine.
Firstly, my main idea was drawn from all the talk of building a new spectrum. I decided that the way to tackle this was from a software point of view, rather than a hardware one - we already have hardware available, why not use it? There will be enhancements - a 640x480x8 screen with the adlib or soundblaster card available would make a nice little retro machine. Of course, with the interpreted BASIC being available from boot up you've got something that no current OS offers. As the BASIC will also serve as the DOS system, then you have quite a powerful little thing to play with right there.
New commands to handle sprites, much improved faster graphics commands, sound commands etc would mean that rewriting old Spectrum games in BASIC might actually be a reality. UDGs and fonts in 256 colours, PRINTing to any position on the screen (as there will be no attribute squares to speak of), an extended character set to handle new commands (up to 65535 of them with the current design)... the list goes on.
There won't be any multitasking, nor any memory protection - the whole lot (including the OS) will be available in one huge chunk like the Spectrum was. Machine code routines would be used the same way they are on a Speccy, LOADed as CODE and the executed with a USR call.
And of course, hopefully I can get the SPECTRUM command to do something useful by launching a Spectrum emulation for running tape images from hard disk.
I'm currently researching the things I need to know in order to write an OS - using the many open source linux repositories is a good start. First step is a bootloader that dumps the CPU into protected mode and exposes the RAM, with the VGA frame buffer at a sensible location, then loads the OS to virtual address zero.
We shall see how far I get. It's a massive project, but looks like fun. I've been discussing it with some people (Woody's a great help, as are the denizens of #speccy from the whole OSS pov) and it seems do-able.
I've also decided that it will be GPL from the word go, so if anyone wants to contribute then they can.
This is probably the next logical step for BASin, in my mind. Work begins mid-january, after my Uni course has finished and my time is my own again.
Yep. Go for it Dunny! Sounds like some amazing technicolour dream but it sure is the most exciting thing I've heard since... well.. ZX Shed issue 4 was announced! :grin:
Seriously, it's worth a shot and I'm sure the combined techno-wizardry of the speccy community will be right behind this one! Go! Go! Go Dunny!
Sounds like an interesting project - certainly it'll keep you off of the streets for a while :wink:
I've posted about it on raww.org, as I guess a lot of people would miss this thread.
By coincidence I had similar ideas - you've not being spying on me have you ;)
My plan was to take a uITX motherboard in a nice looking case then flash a spectrum emulator in place of the BIOS. When you turn the PC on the emulation is available instantly and exposes the hardware to any running program.
My plan was to take a uITX motherboard in a nice looking case then flash a spectrum emulator in place of the BIOS. When you turn the PC on the emulation is available instantly and exposes the hardware to any running program.
Similar, but I'm not interested in emulation other than as a compatibility layer this time. As you're all aware, I'm pretty much all about BASIC. The only way to get a usable, fast BASIC is to port the ROM to x86 and go from there.
I imagine that with a reasonable processor (486 up) and decent sprite/graphics commands, games like a Cybernoid remake could be written in BASIC without having to resort to m/c. Of course, time will tell, but with a lot more room to get a decent interpreter going and other luxuries like an FPU and sane display layout, I'm confident that impressive speeds can be had.
my main idea was drawn from all the talk of building a new spectrum. I decided that the way to tackle this was from a software point of view, rather than a hardware one
I started to imagine the project as complete, before me, installed on my laptop. it will be a brand new thing -- not a regular speccy, let's say a ZX Spectrum NewGen.
I'm not sure if it's too early to ask about this, but to clear my thoughts I have a question:
what if somebody produce an ZX NG emulator for windows? What will be the difference? what will keep people using ZX NG as their base system rather than running it on a virtual pc or as an NG emulator?
After completing the software part, are you going to announce some strict specs, in case somebody wants to build the real hardware for it?
*Maybe you should create a new topic for that matter, because your project will be the best thing for community, and I'm sure everybody wants to say something about it. And find a temp name for your project, that makes it easier to refer.
what if somebody produce an ZX NG emulator for windows? What will be the difference? what will keep people using ZX NG as their base system rather than running it on a virtual pc or as an NG emulator?
As there will be people who do not own the spare hardware to run this, I would expect them to run it in vPC or Bochs or similar. After all, I'll be using vPC to develop it. People will *not* run it on their main PC. It will have no internet capability. No networking. No multitasking. No 3D acceleration. It will not run DOS programs, Linux programs or Windows programs. There is no reason at all to run it on your 2.8ghz dual-core wonder machine. Although it would be unbelievably fast on something like that...
After completing the software part, are you going to announce some strict specs, in case somebody wants to build the real hardware for it?
The hardware specs are simple: a PC. Specifically, a 486 or better. With a graphics card capable of basic VGA output of at least 640x480x8bpp. Probably a Soundblaster card for sound, but that's not finalised yet. Certainly the internal beeper will be utilised at first.
*Maybe you should create a new topic for that matter, because your project will be the best thing for community, and I'm sure everybody wants to say something about it. And find a temp name for your project, that makes it easier to refer.
The temporary name is SpecOS. Or maybe not. I have no idea yet.
For now, just let me get going on it. Once I have the ROM ported over and have achieved as much BASIC compatibility as the hardware differences allow, then we can start to think about how to improve and modify things. It may take up to a year to get that far.
This is one of the best ideas I've heard about in ages! For people like me who don't have an old PC to spare, I'm hoping it'll be bootable from a USB key (or at least a separate partition on my HD) so that we can have the best of both worlds. I had some very basic ideas for writing some sort of a Spectrum simulator program written at the very low level (as I've been playing around with x86 assembly lately). But this would be very primitive compared to the type of stuff Dunny is capable of.
No doubt I'll have some ideas about graphics modes and stuff that I'd like to contribute at some point, once I've decided what those ideas are. I wouldn't want to step on ayone's toes, though.
Comments
Assembly is fast, ok, but C is also compiles well enough today and easy to maintain.
It's better to get modern :) Eg. Python, even sid meier using it nowadays :P
Well, I can't use Python because it's interpreted and there won't be a python interpreter available for the next project. Assembly is just fine, because it won't run on Windows either :-)
D.
Oh no, are you taking Spin to command line? :P
eheh, linux. I still say python, easy to port, you can still keep woody's asm core in a dll.
I suspect you are gonna create unusual emulator, capable of strange abilities.
oh, wait.. "there won't be a python interpreter available for the next project." ? Spin for GP2x? Spin for SymbianOS? Spin for ZXspectrum 128k? Spin for Windows3.1? I'll get my coat.
No.
Nope.
No, no, no and finally not that either.
D.
I didn't write Steve's TZX Reporting tool in assembly ;)
Yeah, that pretty much sums up Delphi.
D.
but you thought to do that in asm at first, didn't you?
Dunny, I remember you as being a total delphi believer, what happened to you? but you are on the right track anyway. Even vb is better than delphi.
I've been using Delphi for years and I can't see how you can say VB is better. Do you want to backup your claim by giving a few examples of how VB is better ? I haven't seen many VB Spectrum emulators about.
Not sure how that compares to the number of Delphi emulators, mind.
As for which language is best - it depends on the task and what the programmer feels comfortable with. Personally I'm a C++ man. But that's just me.
Oh, and as for Dunny saying the next version of Spin and BASin will be in asm... it's obvious he's going to be writing them for the ZX. Hell yeah!
and it's a personal preference ofcourse, but I'm strongly against using pascal, if you want to use RAD, I prefer basic (maybe because I'm a specchum), if I need more power, I use ansi C. Pascal is in-between for me and it's pretty useles in my point of view. Yes, it's a good starter language and it's faster than basic. But --hey, I dont want to use "==" and ";"s in my RAD.
And for spin, I really like spin the way it is, this is the first project that uses delphi for good.
nope, few post ago, he said no to ZX.
But Dunny, I really don't care much for spin it's already complete and working. But basin needs maintennance and it must run on windows and/or linux, come on! if not, make it opensource then!
OK!
Installed it a while back now and it runs fine although I have no drivers for my M-Audio sound card.
VB is very much inferior to Delphi, in pretty much every way possible. They've even managed to corrupt BASIC so much that even Delphi's corrupted Pascal is better.
Um. Have you never actually downloaded BASin then?
D.
No, not at all. It's purely a myth that I write everything in assembly!
I know this is way off topic but I just wanted to point out that my Delphi applications control the swipe card access to some of the biggest companies in the world. This involves database access, networking, XML parsing, NT services and user interfaces all written in Delphi (some of it still in 16bit !). I think you underestimate the power and RAD of Delphi.
? I'm using basin regularly. But you said that basin will not be available in windows anymore and you know it's bugs more than I. it's the most needed program for a retro-hobbyist. it will be good to see basin still maintained for windows in the future.
it's your program, we all are thankful to you. the sentence I said in my previous post about open source, was my wish actually. I don't want to see the source, I only want to see new version of basin :)
And Mr. Millside, I respect your thoughts, and they are true. it's just my preference, I know pascal > basic.
I feel like the troll of the month :)
Crikey, that's close, but no cigar :-)
Okay, I'll come clean. Please bear in mind that this is not even started, let alone finalised in design yet. The groundwork has been laid, but it's still in planning. The idea is to write a new OS for x86 PCs. It's not a competing OS for windows or linux - it's not designed to be.
I'm going to port the Sinclair Spectrum ROM to 32bit x86 assembler. It's mostly so I can find a use for that old 486 laptop of mine.
Firstly, my main idea was drawn from all the talk of building a new spectrum. I decided that the way to tackle this was from a software point of view, rather than a hardware one - we already have hardware available, why not use it? There will be enhancements - a 640x480x8 screen with the adlib or soundblaster card available would make a nice little retro machine. Of course, with the interpreted BASIC being available from boot up you've got something that no current OS offers. As the BASIC will also serve as the DOS system, then you have quite a powerful little thing to play with right there.
New commands to handle sprites, much improved faster graphics commands, sound commands etc would mean that rewriting old Spectrum games in BASIC might actually be a reality. UDGs and fonts in 256 colours, PRINTing to any position on the screen (as there will be no attribute squares to speak of), an extended character set to handle new commands (up to 65535 of them with the current design)... the list goes on.
There won't be any multitasking, nor any memory protection - the whole lot (including the OS) will be available in one huge chunk like the Spectrum was. Machine code routines would be used the same way they are on a Speccy, LOADed as CODE and the executed with a USR call.
And of course, hopefully I can get the SPECTRUM command to do something useful by launching a Spectrum emulation for running tape images from hard disk.
I'm currently researching the things I need to know in order to write an OS - using the many open source linux repositories is a good start. First step is a bootloader that dumps the CPU into protected mode and exposes the RAM, with the VGA frame buffer at a sensible location, then loads the OS to virtual address zero.
We shall see how far I get. It's a massive project, but looks like fun. I've been discussing it with some people (Woody's a great help, as are the denizens of #speccy from the whole OSS pov) and it seems do-able.
I've also decided that it will be GPL from the word go, so if anyone wants to contribute then they can.
This is probably the next logical step for BASin, in my mind. Work begins mid-january, after my Uni course has finished and my time is my own again.
Thoughts, anyone?
Cheers,
D.
Yep. Go for it Dunny! Sounds like some amazing technicolour dream but it sure is the most exciting thing I've heard since... well.. ZX Shed issue 4 was announced! :grin:
Seriously, it's worth a shot and I'm sure the combined techno-wizardry of the speccy community will be right behind this one! Go! Go! Go Dunny!
Bytes:Chuntey - Spectrum tech blog.
We'll need a serious thinking about it. it's not a joke.
I've posted about it on raww.org, as I guess a lot of people would miss this thread.
A P-133 would certainly be quite powerful for this. The BASIC would certainly move rather nicely.
No, it's certainly not a joke.
By coincidence I had similar ideas - you've not being spying on me have you ;)
My plan was to take a uITX motherboard in a nice looking case then flash a spectrum emulator in place of the BIOS. When you turn the PC on the emulation is available instantly and exposes the hardware to any running program.
MikeW
but nonetheless fascinating and intriguing...
Similar, but I'm not interested in emulation other than as a compatibility layer this time. As you're all aware, I'm pretty much all about BASIC. The only way to get a usable, fast BASIC is to port the ROM to x86 and go from there.
I imagine that with a reasonable processor (486 up) and decent sprite/graphics commands, games like a Cybernoid remake could be written in BASIC without having to resort to m/c. Of course, time will tell, but with a lot more room to get a decent interpreter going and other luxuries like an FPU and sane display layout, I'm confident that impressive speeds can be had.
D.
I started to imagine the project as complete, before me, installed on my laptop. it will be a brand new thing -- not a regular speccy, let's say a ZX Spectrum NewGen.
I'm not sure if it's too early to ask about this, but to clear my thoughts I have a question:
what if somebody produce an ZX NG emulator for windows? What will be the difference? what will keep people using ZX NG as their base system rather than running it on a virtual pc or as an NG emulator?
After completing the software part, are you going to announce some strict specs, in case somebody wants to build the real hardware for it?
*Maybe you should create a new topic for that matter, because your project will be the best thing for community, and I'm sure everybody wants to say something about it. And find a temp name for your project, that makes it easier to refer.
As there will be people who do not own the spare hardware to run this, I would expect them to run it in vPC or Bochs or similar. After all, I'll be using vPC to develop it. People will *not* run it on their main PC. It will have no internet capability. No networking. No multitasking. No 3D acceleration. It will not run DOS programs, Linux programs or Windows programs. There is no reason at all to run it on your 2.8ghz dual-core wonder machine. Although it would be unbelievably fast on something like that...
The hardware specs are simple: a PC. Specifically, a 486 or better. With a graphics card capable of basic VGA output of at least 640x480x8bpp. Probably a Soundblaster card for sound, but that's not finalised yet. Certainly the internal beeper will be utilised at first.
The temporary name is SpecOS. Or maybe not. I have no idea yet.
For now, just let me get going on it. Once I have the ROM ported over and have achieved as much BASIC compatibility as the hardware differences allow, then we can start to think about how to improve and modify things. It may take up to a year to get that far.
D.
No doubt I'll have some ideas about graphics modes and stuff that I'd like to contribute at some point, once I've decided what those ideas are. I wouldn't want to step on ayone's toes, though.