What's stopping you programming the Jaguar?

[+GroovyBee]

I use makefiles in my 7800 development. Saves quite a bit of hassle in my opinion.

[LinkoVitch]

I use makefiles in my 7800 development. Saves quite a bit of hassle in my opinion.

 

Wow, I didn't think of using them with Photoshop before! awesome ! :D :D :D :D

[+Gemintronic]

If people have this configured for Windows and Code::Blocks could they also virtualize it for mere mortals who don't know how to piece it together?

http://www.cameyo.com/

[LinkoVitch]

My environment isn't a single application, so I am not sure that would work.. I could possibly look into writing up how I have my build environment setup under windows and put that online somewhere so others can at least duplicate what I have, do people think that would be of use?

 

It would essentially be a list of links to software I use (notepad++, dosbox etc), along with simple instructions of where I put files and things.. and path settings..

[VladR]

You don`t have to put it online, just a forum post is enough.

 

Take a Removers lib, build some example and show the makefile, all tools (smac, gcc, whatever) and paths and describe the outputs of each substep (e.g. *.o files, ....).

 

 

You don`t have to do it for Hello World - that works, right out of the box

[VladR]

It`s so easy you can`t be bothered, eh ?

 

 

I had some time today and downloaded smac/sln toolchain. However, I don`t see how it can be beneficial to Removers, since it`s purely ASM processing tool that has been compiled for Windows.

Meaning, it is probably meant to be used under Dosbox to compile (assemble)/link *.s files into *.o and eventually into *.coff

 

However, the Removers lib is for C, not ASM, so the only way smac/sln could be beneficial [in the context of Removers] is that if one wanted to actually compile the library from its source code. Which is not really needed, since the Distribution provides for header files and libraries.

 

Unless, and that is a big if, it is possible to just use the sln linker to take the *.o files that have been generated by gcc (e.g. to compile the C code). Theoretically, it might work, but it surely is stretching it.

 

Anyone can shed more light into it ?

 

 

 

In the meantime, i`m going to check the cygwin installation (it finally finished installing 300 MBs of tools) and see if I can finally run gcc there, since Removers makefiles are totally based on linux.

 

EDIT : Mint gcc had to be reinstalled too.

 

I feel like I`m getting really close

Edited July 1, 2012 by VladR

[VladR]

Now, I`m getting somewhere. having fiddled with the makefile, I gave it location of the Vincent`s gcc so gcc has been finally executed....

 

 

.... only to report some jagdefs.h missing. Obviously, the Removers library doesn`t contain it, neither do other Jag C examples I downloaded from net, so I`m gonna try to find it elsewhere.

[VladR]

It`s getting hotter

 

I just configured the makefile to create the *.o file. So, I got the compilation part configured.

 

Now, I need to do that for linking.

[+CyranoJ]

It`s so easy you can`t be bothered, eh ?

 

No, I just don't see the point in creating a dev environment for you on my computer

 

As said above, if you'd spent as much time coding your stuff as you do telling us about you coding your stuff and all the stuff you've done before and how bad everything is and how much better .net is and blah blah blah you'd have finished it by now. Just get on with it fer gods sake, we don't need a running commentary

 

This is a TWENTY YEAR OLD MACHINE. It wasn't designed for C++, etc.

[VladR]

It`s so easy you can`t be bothered, eh ?

 

No, I just don't see the point in creating a dev environment for you on my computer

 

As said above, if you'd spent as much time coding your stuff as you do telling us about you coding your stuff and all the stuff you've done before and how bad everything is and how much better .net is and blah blah blah you'd have finished it by now. Just get on with it fer gods sake, we don't need a running commentary

 

But at least I`m doing something with it and trying to push it further.

 

You`re missing the point. I am commenting my progress here, since of all the Jag forums, no one ever bothered to try it. Or, to be more precise, finish it - since I`ve seen couple examples where people wanted to get the Removers lib up&running, but failed - since on Windows - it`s anything but drop&dead easy.

 

At least, anyone who`ll ever put Removers, Jaguar, Windows, C++ into google in future will find this post and at least see what needs to be done.

 

This is a TWENTY YEAR OLD MACHINE. It wasn't designed for C++, etc.

And how would you know ? You`re doing everything in ASM. Can you give us an example of what type of C++ code you tried on Jag and it ran miserably slow ?

[rush6432]

Because if it WAS designed for C++ The system would have had more successfull library of games, and been a bit more popular at the time of release and possibily lived a bit longer...

 

Obviously it wasnt designed for it... the development manual even states this for the GPU and DSP processors i believe that they are designed to be programmed using assembly.

 

As said before, you can create a toolchain to compile c++ on your own and if you get it working, great!!! but there is no need to re-invent the wheel here. the machine WAS DESIGNED to be interacted with at a low level using ASM or a bit of C and nothing is going to change this unless you write it yourself.

 

Thats great that you want to document this, might try a blog however....

 

Send dr typo a pm. he has a successfully toolchain setup to code in C and it seems to work great for him.

 

However based on your posts and the fact that you refuse to use anything other than windows and dont SEEM to be able to code in anything BUT c++ or .net i think you're going to have a very hard time doing anything with the jaguar.

Edited July 1, 2012 by rush6432

[+CyranoJ]

And how would you know ? You`re doing everything in ASM. Can you give us an example of what type of C++ code you tried on Jag and it ran miserably slow ?

 

And you know what? So is everyone else. And there's a reason for that.

 

If you drop an apple, do you need to see it fall to know it hits the ground? I look forward to 2025, when the first frame of your C++ renderer has finished drawing.

[+Sauron]

And how would you know ? You`re doing everything in ASM. Can you give us an example of what type of C++ code you tried on Jag and it ran miserably slow ?

 

And you know what? So is everyone else. And there's a reason for that.

 

If you drop an apple, do you need to see it fall to know it hits the ground? I look forward to 2025, when the first frame of your C++ renderer has finished drawing.

 

There's no need for comments like this. Please be a little more tactful.

[rush6432]

It`s so easy you can`t be bothered, eh ?

 

 

I had some time today and downloaded smac/sln toolchain. However, I don`t see how it can be beneficial to Removers, since it`s purely ASM processing tool that has been compiled for Windows.

Meaning, it is probably meant to be used under Dosbox to compile (assemble)/link *.s files into *.o and eventually into *.coff

 

However, the Removers lib is for C, not ASM, so the only way smac/sln could be beneficial [in the context of Removers] is that if one wanted to actually compile the library from its source code. Which is not really needed, since the Distribution provides for header files and libraries.

 

 

sln/smac were designed to be used with updated windows/linux environments

 

No need to use dosbox unless you want to use aln.exe and mac.exe

 

Take a look at some of the source code around. there are some examples of c code in the club drive beta source code, dr typo's c coded games, atari 3d renderer example from (95ish?!?) is done in C also.... Doom was written in C as well... you can view the source from that and makefiles also..

doom, 3d renderer and club drive beta src are all using the old atari toolchain however...

Edited July 1, 2012 by rush6432

[VladR]

And how would you know ? You`re doing everything in ASM. Can you give us an example of what type of C++ code you tried on Jag and it ran miserably slow ?

 

And you know what? So is everyone else. And there's a reason for that.

 

If you drop an apple, do you need to see it fall to know it hits the ground? I look forward to 2025, when the first frame of your C++ renderer has finished drawing.

I never said I would be doing the rendering of actual pixels through C++. I said numerous times, that the time-critical routines should be in ASM. Which is why I`m planning to use Removers lib, since it has everything I need to start.

 

But the remaining 95% of the code (input,logic,AI,tools) doesn`t have to be in ASM. Perhaps logic that gets executed every frame.

 

 

Still, it`s pretty hilarious, that if one wants to start creating games for this platform, that it pisses so many people

 

Abrasive community at its best

[VladR]

And you know what ?

 

It`s better to try and fail, than not be bothered to try at all.

 

 

 

However, I fully understand that not many people will understand that. In fact, I don`t think I`ve met more than 3 people in my life that share this school of thought...

[+CyranoJ]

Not abrasive at all, just think your barking up the wrong tree.

 

I don't really care what games are written in as long as they are fun. Bring on the Jaguar games, I say. Just don't spent endless years shovelling false promises at the Jaguar community. It doesn't need more of that.

[rush6432]

Couldnt have said it better myself...

Edited July 1, 2012 by rush6432

[VladR]

_{It`s getting hotter}

 

_{I just configured the makefile to create the *.o file. So, I got the compilation part configured.}

 

_{Now, I need to do that for linking.}

_{Getting back to original problem - I just realized that I only have the compiler part under CygWin. For obvious reasons (it`s not a virtual machine, it just appears linux-like), cygwin won`t run regular linux executables.}

 

_{Which means, it won`t run the original ALN linker. I could try to create one with the smac/sln linker, since that`s a Win32 executable (but I`m afraid that can`t work - trying surely doesn`t hurt).}

 

_{Of course, if I ran Virtual Machine, and had a full linux distro installed there, it should work.}

[VladR]

Not abrasive at all, just think your barking up the wrong tree.

It may not look like that to you after all those years, but it definitely is "sharp" to an untrained eye.

 

Where else am I supposed to ask technical questions about Jag if not on Jag Dev forum ? XBOX section perhaps ? Or Off-Topic, perhaps ?

 

Seriously ?!?

 

 

I have been warned beforehead, so it`s not like I didn`t expect this

 

Come on, do yourself a favour. Kick once more

[+Sauron]

VladR, the Jag community is a bit abrasive, yes, and CyranoJ especially. Don't let it put you off, but at the same time, you should give some weight to what he says. He's done quite a bit for the Jag these last few years, and really knows his way around the system.

 

Personally, I think you should give more thought to developing in ASM for the Jag. The requisite C tools that you seem to be looking for simply do not exist, and it's not like the Jag has a large enough community to bring in enough developers to create them.

[+Gemintronic]

_{It`s getting hotter}

 

_{I just configured the makefile to create the *.o file. So, I got the compilation part configured.}

 

_{Now, I need to do that for linking.}

_{Getting back to original problem - I just realized that I only have the compiler part under CygWin. For obvious reasons (it`s not a virtual machine, it just appears linux-like), cygwin won`t run regular linux executables.}

 

_{Which means, it won`t run the original ALN linker. I could try to create one with the smac/sln linker, since that`s a Win32 executable (but I`m afraid that can`t work - trying surely doesn`t hurt).}

 

_{Of course, if I ran Virtual Machine, and had a full linux distro installed there, it should work.}

 

I know 1.5 gb is a bit much, but, if there's no Win32 ALR linker I still think Portable Ubuntu may give it a chance. We're talking native Linux running under Windows.

 

The point is to get people dev'ing on the Jag. Most people came from a higher level language background. If you want people to go ASM then give them a stepping stone.

Edited July 1, 2012 by theloon

[+GroovyBee]

If people spent more time researching rather than complaining about the lack of tools they might have found this :-

 

http://www.hillsoftware.com/file-downloads-2/

 

Just download the first link "Windows 2K/XP/VISTA comp Jag dev tools v1.02 (5MB installer)". Then go in the hello folder after you've installed it and take a looksee in there.

[+Gemintronic]

If people spent more time researching rather than complaining about the lack of tools they might have found this :-

 

http://www.hillsoftw...le-downloads-2/

 

Just download the first link "Windows 2K/XP/VISTA comp Jag dev tools v1.02 (5MB installer)". Then go in the hello folder after you've installed it and take a looksee in there.

 

The installer doesn't put an IDE or even some sort of console into the Programs folder. When I take a peek at the hello directory there seems to be some sort of script called makefile. How does one compile the first example?

[+GroovyBee]

The installer doesn't put an IDE or even some sort of console into the Programs folder. When I take a peek at the hello directory there seems to be some sort of script called makefile. How does one compile the first example?

 

Yep! Many compiler systems don't have an IDE. I use my favourite text editor and makefiles so that doesn't really bother me. To build the example just open a DOS box, make sure a version of NMAKE is in your path some place, navigate to the directory and type NMAKE. Watch it build and fix any errors that occur.

 

If you Google around you can get an old version of Microsoft NMAKE (v1.5?) from their website.