| Author |
Message |
|
JoostinOnline
Active Member
Joined: 22 Aug 2007
Posts: 559
|
 Posted: 02 Mar 2009 01:09:32 pm Post subject: |
|
|
Galandros wrote: Or simplify things using emu8x... hehe Downgrade my calculator. :biggrin: Not really a downgrade but quite near of that. xD
Emu8x does not run the TI-73, only the 82, 83, 85, and 86 calculators. |
|
| Back to top |
|
|
TheStorm
Calc Guru
Joined: 17 Apr 2007
Posts: 1233
|
| Posted: 02 Mar 2009 05:40:34 pm Post subject: |
|
|
| Well he could run PongOS, BEOS or one of the other third party os's available. Though none of them do much atm they are good POC's. |
|
| Back to top |
|
|
FloppusMaximus
Advanced Member
Joined: 22 Aug 2008
Posts: 472
|
| Posted: 02 Mar 2009 08:50:49 pm Post subject: |
|
|
I think you mean BAOS. BeOS on a Z80 would be a nice trick. :)
(It's been a while since I looked at BAOS... did the project ever go anywhere?) |
|
| Back to top |
|
|
critor
Member
Joined: 04 Feb 2009
Posts: 132
|
| Posted: 11 Mar 2009 04:59:14 pm Post subject: |
|
|
Another question.
The TI-83+SE (OS 1.13-1.19) and TI-84+SE have similar hardware (2048Kb ROM).
I know it is possible to install a TI-83+ OS (1.13-1.19) on a TI-84+SE.
I know Brandon has a tool to send the OS from a TI-83+ to a TI-84+.
My question is:
Would it be possible to patch a 83+ 8Xu upgrade file, so that it could be sent to a TI-84+ directly from the computer?
I've tried to modify a TI-83+ 8Xu file, like we did before (replacing the header, prevalidating).
It begins to transfer normally.
It says validating before 100% (something like 70% if I remember well - I knew it because a 83+ update is smaller than a 84+ update).
I immediately remove a battery, press and hold CLEAR, and put back the battery to force the booting, of the missigned OS.
But the system doesn't boot at all: I'm back at the "Install OS now" screen.
Just for information, I've tried other things, not directly related to my question:
* Send a TI-84+ OS to a TI-83+.
The transfer begins normally.
It reaches 100% and goes back to 0%, then 7%... (84+ update file is bigger than 83+ update file)
At something like 20% (which would mean 120%) it says Validating.
I immediately remove a battery, press and hold CLEAR, and put back the battery to force the booting, of the missigned OS.
The calculator seems to try to boot the system, beacause it doesn't go back to the "Install OS" screen.
But it doesn't work:
- Either the screen stays off.
- Either the screen turns on, but nothing is displayed.
It suppose there is a problem, with the missing USB and clock circuits.
* Send a 83+ OS to a basic 84+.
I've used Brandon's tool (as it doesn't work from a computer for now).
The 84+ reboots, and introduces itself as a TI-83+ Silver Edition.
It works for basic use, but accessing the archive memory hangs the system.
Of course, the TI-84+ has a 1024Kb ROM, and the 83+ OS is probably just compatible with 512Kb ROMs (TI-73, TI-83+) and 2048Kb ROMs (TI-83+SE, TI-84+SE).
I've noticed, the reported Archive size in the memory screen is bad (over 1000Kb which is impossible on a 1024Kb ROM... unless the system is strongly zipped) |
|
| Back to top |
|
|
brandonw
Advanced Member
Joined: 12 Jan 2007
Posts: 455
|
| Posted: 11 Mar 2009 06:56:23 pm Post subject: |
|
|
critor wrote: Another question.
The TI-83+SE (OS 1.13-1.19) and TI-84+SE have similar hardware (2048Kb ROM).
I know it is possible to install a TI-83+ OS (1.13-1.19) on a TI-84+SE.
I know Brandon has a tool to send the OS from a TI-83+ to a TI-84+.
My question is:
Would it be possible to patch a 83+ 8Xu upgrade file, so that it could be sent to a TI-84+ directly from the computer?
I've tried to modify a TI-83+ 8Xu file, like we did before (replacing the header, prevalidating).
It begins to transfer normally.
It says validating before 100% (something like 70% if I remember well - I knew it because a 83+ update is smaller than a 84+ update).
I immediately remove a battery, press and hold CLEAR, and put back the battery to force the booting, of the missigned OS.
But the system doesn't boot at all: I'm back at the "Install OS now" screen.
Just for information, I've tried other things, not directly related to my question:
* Send a TI-84+ OS to a TI-83+.
The transfer begins normally.
It reaches 100% and goes back to 0%, then 7%... (84+ update file is bigger than 83+ update file)
At something like 20% (which would mean 120%) it says Validating.
I immediately remove a battery, press and hold CLEAR, and put back the battery to force the booting, of the missigned OS.
The calculator seems to try to boot the system, beacause it doesn't go back to the "Install OS" screen.
But it doesn't work:
- Either the screen stays off.
- Either the screen turns on, but nothing is displayed.
It suppose there is a problem, with the missing USB and clock circuits.
* Send a 83+ OS to a basic 84+.
I've used Brandon's tool (as it doesn't work from a computer for now).
The 84+ reboots, and introduces itself as a TI-83+ Silver Edition.
It works for basic use, but accessing the archive memory hangs the system.
Of course, the TI-84+ has a 1024Kb ROM, and the 83+ OS is probably just compatible with 512Kb ROMs (TI-73, TI-83+) and 2048Kb ROMs (TI-83+SE, TI-84+SE).
I've noticed, the reported Archive size in the memory screen is bad (over 1000Kb which is impossible on a 1024Kb ROM... unless the system is strongly zipped)
Okay...one thing at a time.
You cannot put the 83+/SE OS on an 84+. There is no support whatsoever for the regular 84+ Flash chip, as you've discovered. You cannot avoid those "beyond basic use" issues, and it's not worth messing with.
You can't just pull a battery and try sending an invalid OS to the 84+/SE because the boot code "bug"/"feature" which made that work was fixed in the 84+/SE boot code. To send an invalid OS to it, you must use another exploit. I wrote a small program that does this (I think it's brandonw.net/calcstuff/sendOS.zip), which is probably the tool you're referring to.
A PC counterpart was written by FloppusMaximus here called Ponginstaller, but to my knowledge he never released it anywhere for fear of people modifying TI's copyrighted OS and then sending it to an 84+/SE.
If you'd like to write your own, the exploit which he likely uses is found in sendOS.zip. |
|
| Back to top |
|
|
critor
Member
Joined: 04 Feb 2009
Posts: 132
|
| Posted: 08 Aug 2009 04:01:01 pm Post subject: |
|
|
Using RabbitSign, and the TI-83+ private key found by FloppusMaximus, I was able to sign TI-73 OSes in order to install them on a TI-83+.
I've got OSes 1.3004, 1.40, 1.60, 1.80, 1.90 and 1.91.
(I'm missing 1.3007 and 1.50 to my knowledge)
Months ago, I had allready installed TI-73 OSes on an old black TI-83+ (not SE) with boot code 1.00.
There was a display bug with 8pixels-fonts, when displayed on a 12- or 14-pixels line, but no major issue.
Now, I don't have that old TI-83+ any more.
I've got a new blue TI-83+ (not SE) with boot code 1.01 and hardware revision L (manufactured 06/2008).
With OSes 1.90 and 1.91, everything seems to be like before.
With OSes 1.80 and lower, I've got a big problem...
When I press ON for the 1st time after the OS has been validated, the screen turns on for a second (nothing is displayed) and then immediatly turns off...
Although I can't see anything, the OS seems to be working: I was able to transfer data to the computer.
By dismantling the calculator, I noticed that the display driver is not a Toshiba T6A04 / T6K04 chip any more, but a Novatek NT7564H chip.
Does anybody have any idea about what's going on?
Thanks. |
|
| Back to top |
|
|
calcdude84se
Member
Joined: 09 Aug 2009
Posts: 207
|
| Posted: 15 Aug 2009 02:48:46 pm Post subject: |
|
|
TI obviously replaced the display driver (not that was any help, you figured that out already). If the new driver uses different commands than the old one, that would explain the problem, but it also means that the older TI-79 OS can't communicate with it properly. You would have to patch the 79 OS so you could see what you were doing (or just keep taking screenshots...  ) |
|
| Back to top |
|
|
fullmetalcoder
Member
Joined: 01 Aug 2009
Posts: 139
|
| Posted: 15 Aug 2009 03:17:49 pm Post subject: |
|
|
calcdude84se wrote: TI obviously replaced the display driver (not that was any help, you figured that out already). If the new driver uses different commands than the old one, that would explain the problem, but it also means that the older TI-79 OS can't communicate with it properly. You would have to patch the 79 OS so you could see what you were doing (or just keep taking screenshots... )
my guess is that the LCD commands did not change (or a new OS would have to be released and many apps would be broken) but that the required delays may have... There is an assembly program floating around to adjust LCD delays but it works only on SE hardware because it uses harwdare ports not available on 83+BE. |
|
| Back to top |
|
|
FloppusMaximus
Advanced Member
Joined: 22 Aug 2008
Posts: 472
|
| Posted: 15 Aug 2009 05:11:50 pm Post subject: |
|
|
That would make sense, but I don't see anything obviously different about how the new OS accesses the LCD.
critor: Try running this program, if you can figure out how to type it and execute it 
Code:
prgm00000005
3E03D310C9
|
|
| Back to top |
|
|
critor
Member
Joined: 04 Feb 2009
Posts: 132
|
| Posted: 15 Aug 2009 06:24:52 pm Post subject: |
|
|
Thank you for your explanations.
FloppusMaximus -> I've typed your program for the Mallard TI-73 shell.
I've launched it on the calculator. After execution, it goes back to the Mallard menu.
But... it didn't turn on the screen.
I've checked all that by taking screenshots.
Additional information:
I've tested with the 1.85 TI-73 OS on my recent TI-83+, and the screen is working
So with 1.3004, 1.40, 1.50, 1.60 and 1.80, the screen of my "recent" TI-83+ stays off after "flashing" for less than a second at boot time just after the installation... (it then never flashes again)
With 1.85, 1.90 and 1.91, the screen turns on correctly.
Do you think the problem comes from the 1.01 boot code, or from the Novatek display chip?
Unfortunately, I don't have a TI-83+ with boot code 1.01 and the Toshiba chip for testing... |
|
| Back to top |
|
|
calc84maniac
Elite
Joined: 22 Jan 2007
Posts: 770
|
| Posted: 16 Aug 2009 03:30:50 pm Post subject: |
|
|
| Wait... I thought there was no known boot code 1.01? |
|
| Back to top |
|
|
critor
Member
Joined: 04 Feb 2009
Posts: 132
|
| Posted: 16 Aug 2009 03:48:49 pm Post subject: |
|
|
calc84maniac ->
As you are a TI-84+ maniac, your question is understandable
The TI-84+BE has boot codes 1.00 and 1.02.
The TI-84+SE has boot codes 1.00 and 1.02.
The TI-84+SE emulated by the TI-nSpire has boot code 1.02.
The TI-83+BE has boot codes 1.00 and 1.01.
The TI-83+SE has boot code 1.00.
The TI-73 has boot codes 1.3004, 1.3005 and 1.3007.
We're talking about installing a TI-73 operating system on a TI-83+BE, not on a TI-84+
Last edited by Guest on 16 Aug 2009 03:49:52 pm; edited 1 time in total |
|
| Back to top |
|
|
calc84maniac
Elite
Joined: 22 Jan 2007
Posts: 770
|
| Posted: 16 Aug 2009 03:56:23 pm Post subject: |
|
|
| I'm just going by what I read on WikiTI... |
|
| Back to top |
|
|
FloppusMaximus
Advanced Member
Joined: 22 Aug 2008
Posts: 472
|
| Posted: 17 Aug 2009 06:14:40 pm Post subject: |
|
|
| I didn't know the Nspire's boot version was 1.02. That's delightfully annoying. And I suppose it reports hardware version 3, as well? |
|
| Back to top |
|
|
Taricorp
Member
Joined: 09 Mar 2006
Posts: 188
|
| Posted: 17 Aug 2009 11:42:05 pm Post subject: |
|
|
calc84maniac wrote: Wait... I thought there was no known boot code 1.01?
I can confirm it exists. I've seen it on a couple 83+ BEs. If anybody cares to play 'spot the changes' with it, I don't have a dump of it, but I gave a copy to BrandonW at some point- he might still have it (do you?). |
|
| Back to top |
|
|
FloppusMaximus
Advanced Member
Joined: 22 Aug 2008
Posts: 472
|
| Posted: 18 Aug 2009 12:13:16 am Post subject: |
|
|
Here are all the changes from boot 1.00 to 1.01:
Code:
--- stdin
+++ ./sdasm ti83plus_boot.sd -a2 -d TI83Pv115.rom -p 1F
@@ -7,7 +7,7 @@
rst $xxxx
rst $xxxx
rst $xxxx
- .DB "1.00", 0
+ .DB "1.01", 0
.DB $FF, $FF, $FF, $FF
.DPP MD5Final, $1F
.DPP RSAValidate, $1F
@@ -247,7 +247,7 @@
ld bc, $xxxx
ld (hl), $00
ldir
- ld a, $1F
+ ld a, $1C
ld (contrast), a
ret
ld sp, $xxxx
@@ -845,12 +845,13 @@
;==== ROM Call 80B7 ====
GetBootVer:
ld a, $01
- ld b, $00
+ ld b, $01
ret
;==== ROM Call 80BA ====
GetHWVer:
ld a, $00
+ ld b, $01
ret
;==== ROM Call 80CF ====
@@ -6640,7 +6641,13 @@
ld a, $03
call $xxxx
out ($10), a
- ld a, $1F
+ ld a, $17
+ call $xxxx
+ out ($10), a
+ ld a, $0B
+ call $xxxx
+ out ($10), a
+ ld a, $1C
ld (contrast), a
ld a, (contrast)
add a, $18
@@ -8496,8 +8503,6 @@
.DB $FF, $FF, $FF, $FF, $FF, $FF, $FF, $FF
.DB $FF, $FF, $FF, $FF, $FF, $FF, $FF, $FF
.DB $FF, $FF, $FF, $FF, $FF, $FF, $FF, $FF
- .DB $FF, $FF, $FF, $FF, $FF, $FF, $FF, $FF
- .DB $FF, $FF, $FF, $FF, $FF, $FF, $FF, $FF
rst $xxxx
rst $xxxx
rst $xxxx
...Ignore the last part, that's just my disassembler being silly.
The first change is obvious, as is the change to GetBootVer. The change to GetHWVer is interesting, I never noticed that before. As for the LCD-related changes, I suppose it's possible that those could affect your problem, critor, but it doesn't seem likely, as those changes are in internal boot-code routines that (as far as I know) are only used during OS upgrading, not while an OS is installed.
By the way, critor, you have tried adjusting the contrast after the OS is installed, right? |
|
| Back to top |
|
|
critor
Member
Joined: 04 Feb 2009
Posts: 132
|
| Posted: 18 Aug 2009 03:59:13 am Post subject: |
|
|
FloppusMaximus wrote: By the way, critor, you have tried adjusting the contrast after the OS is installed, right?
Of course. :)
The problem is more complex than that...
The screen is really off (no electricity), while the rest of the calculator is on... |
|
| Back to top |
|
|
ztrumpet
Active Member
Joined: 06 May 2009
Posts: 555
|
| Posted: 18 Aug 2009 08:19:08 am Post subject: |
|
|
FloppusMaximus wrote: I didn't know the Nspire's boot version was 1.02.
Also, based on this http://tibasicdev.wikidot.com/forum/t-3681/thecalcs (click and read the second to last post. Click the link in that post if you want to.  ) the Nspire runs os 2.42... |
|
| Back to top |
|
|
critor
Member
Joined: 04 Feb 2009
Posts: 132
|
| Posted: 18 Aug 2009 08:57:17 am Post subject: |
|
|
ztrumpet wrote: FloppusMaximus wrote: I didn't know the Nspire's boot version was 1.02.
Also, based on this http://tibasicdev.wikidot.com/forum/t-3681/thecalcs (click and read the second to last post. Click the link in that post if you want to. ) the Nspire runs os 2.42...
Yes the nSpire runs OSes 2.42, 2.44, 2.46 and 2.48.
But the boot code reported version is 1.02. |
|
| Back to top |
|
|
calcdude84se
Member
Joined: 09 Aug 2009
Posts: 207
|
| Posted: 18 Aug 2009 03:02:19 pm Post subject: |
|
|
Speaking of running a different OS, would it be possible to run say, os 2.48, on a normal ti-84+(SE)?
I'm surprised (and annoyed) that TI hasn't released any updates past 2.43 for the real (non-emulated) ti-84+(SE) |
|
| Back to top |
|
|
|
Calculator Modifications => Calculator Hardware, Electronics, Robotics
