Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - bsl

Pages: 1 ... 3 4 [5] 6 7 ... 11
61
Lua / Re: Will Lua erase the need for Ndless?
« on: June 11, 2011, 11:01:31 am »
Is there currently a version of Ndless 3.0 in the works?
Thats somehing completeley unpredictable.
Finding bugs and exploiting them is an art and a science.
Its a time consuming trial and error process.
How much of your personal time to do this is a factor.

62
But alas, I cannot see, how it should be possible to change the behaviour of the BASIC interpreter this way.
If you want to change/extend the BASIC-interpreter, you either have to write a completely new interpreter as G3A (sagarvaze did this some years ago at casiokingdoms) or you have to mod the OS.
MPoupe  outlines on the other link given above how to do this - there is interaction between the interpreter and the C program.
You establish an I/O protocol through Alpha variables to accomplish the custom call.
So you have to write the basic program differently. I think for the time being I will develop it this way .
The first two custom calls will be Asm and Syscall .

I would really like to make this transparent to the Basic coder as you are wondering about and find a syscall to the interpreter,
such as SetBasicErrorHandler to handle and process new calls if it exists ?

@bsl: are you one of the SDK-chums, which have been around at casiokingdoms some years ago? The old syscall-wrapper: long time no see!
It's in MPoupes code, I only started on Casio calcs January 2011. 

63
The zip is on the posting just above the first picture. Sorry - It is hard to see and easy to miss.

64
Looks interesting :) how exactly does this work?  Is it like an all new BASIC editor/interpreter, or does it just improve upon the current form of BASIC?  Does this have a documentation of commands and the like?
Nothing yet , but it shows promise - its very early at this point to use this - its just a demo

65
This is a port of MPoupe's TSR program for the fx-9860 to the Prizm. http://ourl.ca/10095
I had this idea earlier, but wasn't sure where to start:  http://ourl.ca/9185/173872
I added new syscalls to this program, not found in the mini-SDK, but proven to work here.
The program was developed using Simon's mini-SDK, on the cg10-20 Manager emulator and
finally tested on Prizm hardware. [easier to recover from emulator crashes]
The code is almost the same as MPoupe's , but I took out the scrolling screen and
put in the appearing rotating character sequence "/","-","\", ...
You get this on the lower right corner after the program appears to finish, and press and hold one of the navigation keys: UP,DOWN,LEFT,RIGHT
You exit the progran pressing EXIT.
To run again delete TSRDEMO, and activate another addin like Conversion , then back to BASICEXT


This is only a proof of concept, this works the same way n the Prizm as on the fx-9860
The wish list of what to add to this is long.  Adding custom Basic commmands is possible.


66
TI-Nspire / Re: Lua port on nspire
« on: May 09, 2011, 10:37:09 pm »
The math functins I have listed above will work if the endiannes of the doubles are reversed.
Is there a compiler option for this ?
Here is an example:
Code: [Select]
// test program for NONCAS 1.7
static const unsigned sqrt_addrs[] = {0X102A2490, 0x0};
#define sqrt SYSCALL_CUSTOM(sqrt_addrs, double ,double)
int main(void) {
        double rr;
        rr = 2.0;
        printf("square root of 2 is %lf\n",sqrt(rr) );
        return 0;
}
The double breaks up into the registers r0 and r1 like this:
The working Lua version of this has at entry r0=00000000,r1=40000000
This C version has at entry  r0=40000000,r1=00000000
So a macro that flips the double or a compiler option will fix this.

EDIT: The above program works if changed to:
Code: [Select]
static const unsigned sqrt_addrs[] = {0X102A2490, 0x0};
#define sqrt SYSCALL_CUSTOM(sqrt_addrs, double ,double)

double dswap(double v) 
{
    union {
        unsigned long long i;
        double  d;
    } conv;
    conv.d = v;
    conv.i = (conv.i << 32) | (conv.i >> 32);
    return conv.d;
}

int main(void) {
        double rr;
        rr = 2.0;
        printf("square root of 2 = %15.15f\n",sqrt(dswap(rr)));
        return 0;
}
However, a custom call has to be specially made for the remaing math functions.

67
TI-Nspire / Re: Lua port on nspire
« on: May 05, 2011, 01:09:58 pm »
Here are some math library routines revealed by Lua,
 and cross referenced to OS noncas 1.7

0X102A12E0   acos
0X102A1438   atan
0X102A1688   ceil
0X102A1840   exp
0X102A19D8   fabs
0X102A1A28   floor
0X102A1B20   frexp
0X102A1C58   ldexp
0X102A1D40   fmod
0X102A1FB8   pow
0X102A2118   sin
0X102A22D0   cos
0X102A2490   sqrt
0X105DD938   ln
0X105DDC70   pow
0X102A5ACC   rand
0X102A5B6C   srand
0X102A2630   tan

These are untested and useful for games that use game physics.

68
News / Re: OSLauncher, LUA to TNS converter and TI document player
« on: April 15, 2011, 10:46:54 pm »
It does not. But at least when starting a Phoenix, the caches should be pretty devoid of the code that previously existed at the loading address, shouldn't they be ?
It should, since starting a Phoenix intilializes its own resources, same with boot1,boot2, and diags - they have there own
interrupt vector table and initialization routines.
So the DummyOS demo might encounter problems if ran at some lengthy time.

69
News / Re: Bypassing TI-Nspire RSA signatures now possible?
« on: April 09, 2011, 11:35:04 am »
:P
Now add another intermediate step - patch Ndless hooks into it before launching :)
May be able to port Ndless2.0 to Ndless3.0 this way ?

70
Other Calculators / Re: TI Nspire CAS+ ---- Is it worth it?
« on: April 03, 2011, 05:42:00 pm »
The standard trick in soldering microelectronics with a high wattage iron is to
wrap a thin copper wire on the tip of your soldering iron, to make a small
hot contact.
Make it long enough , then cut pieces of the wire shorter and shorter, till its hot enough
to melt the solder.
A solder sucker is the preferred way to eliminate the solder from the nand-board bond.

Update: Check your local electronics hobby store for desoldering tools also.(mention for microelectronics)
 There are inexpensive tools I found at my store .

71
Other Calculators / Re: TI-Nspire CAS prototype 1.1.6925
« on: April 01, 2011, 07:11:09 pm »
boot2,  hint:boot2launcher code
Start the emulator with a /d
then :
debug> d 1199a7f4

if you are booting with boot1:
k 11800000 +x
then continue, following a dump at that address

72
Other Calculators / Re: TI-Nspire CAS prototype 1.1.6925
« on: April 01, 2011, 07:01:59 pm »
This is looking more CAS+ like:
Code: [Select]
1199a7f4: /tmp/manifest_img
1199a80c: /tmp/TI-Nspire.tnc

73
Other Calculators / Re: TI-Nspire CAS prototype 1.1.6925
« on: April 01, 2011, 10:07:52 am »
Does the examples document mention "CAS+" also ?

Also when you get to the point of dumping- dump the diags too, even though
it appears not there.

74
News / Re: Bypassing TI-Nspire RSA signatures now possible?
« on: March 29, 2011, 08:40:23 pm »
Be careful with boot2launcher .
If you launch a developer boot2 on a production OS (or vice versa)it will delete that OS.

75
Other Calculators / Re: The 1st step into CAS+ flashing
« on: March 25, 2011, 10:17:27 pm »
Try:
type /phoenix/policy.dat

Maybe changing something in this file is all that is needed !!!!!

EDIT: re-naming this file to policy.back :
Code: [Select]
copy policy.dat policy.back
del policy.dat
, may enable USB, and other features.

Pages: 1 ... 3 4 [5] 6 7 ... 11