| Author |
Message |
|
joeyadams
Newbie
Joined: 06 Aug 2009
Posts: 4
|
 Posted: 27 Aug 2009 05:43:16 pm Post subject: |
|
|
adriweb wrote: squalyl wrote: (well... you mean OSes and flashapps for all ti calcs)
yes, right, I said 84+ because I got on this topic from a website dealing only with 84+ :P
and btw, it's sad my Mac can't help with Boinc.... I'll see if my PC laptop can help....
What kind of Mac is it? PowerPC or Intel? If there's plenty of interest, I'll try to get ggnfs working on PowerPC.
By the way, anybody here have a PS3? |
|
| Back to top |
|
|
limx
Newbie
Joined: 18 Aug 2009
Posts: 15
|
| Posted: 28 Aug 2009 02:25:33 am Post subject: |
|
|
Joey Adams wrote: By the way, anybody here have a PS3?
I have a PS3, but for me this game console is too loud for 24h/7d computing. And sometimes I want to play with it or watch movies, so there can't be any computations beside to it.  |
|
| Back to top |
|
|
adriweb
Newbie
Joined: 28 Aug 2007
Posts: 22
|
| Posted: 28 Aug 2009 05:24:48 am Post subject: |
|
|
Joey Adams wrote: adriweb wrote: squalyl wrote: (well... you mean OSes and flashapps for all ti calcs)
yes, right, I said 84+ because I got on this topic from a website dealing only with 84+ :P
and btw, it's sad my Mac can't help with Boinc.... I'll see if my PC laptop can help....
What kind of Mac is it? PowerPC or Intel? If there's plenty of interest, I'll try to get ggnfs working on PowerPC.
By the way, anybody here have a PS3?
I have a MacBook and a MacBook Pro : both Intel.
I've had a PowerPC, but this era is almost over haha |
|
| Back to top |
|
|
fullmetalcoder
Member
Joined: 01 Aug 2009
Posts: 139
|
| Posted: 31 Aug 2009 12:45:01 pm Post subject: |
|
|
Congrats to everyone that got involved and especially to FloppusMaximus who started this and squalyl that allowed the massive increase of computing power by settin up the BOINC project.
edit : just found out about the DMCAs... I'm amazed TI did this so late. It feels a bit like the "Send(9 backdoor" : as if TI actually wants to see how far the community can go but don't want to admit it publicly.
Last edited by Guest on 31 Aug 2009 01:19:31 pm; edited 1 time in total |
|
| Back to top |
|
|
Michael
Newbie
Joined: 21 Dec 2007
Posts: 39
|
| Posted: 31 Aug 2009 07:21:31 pm Post subject: |
|
|
| To the best of my knowledge, TI employees had internal bets on whether the TI-85 would be hacked to run ASM. It wouldn't surprise me. |
|
| Back to top |
|
|
fullmetalcoder
Member
Joined: 01 Aug 2009
Posts: 139
|
| Posted: 02 Sep 2009 07:18:26 am Post subject: |
|
|
critor wrote: Is the following key file correct for signing 84+ OS with RabbitSign ?
000A.key:
Code: 40B11C71D4EA2C13C9AB2E501C6085FEC87FF3B88BFD783EAC43351E1B10F65AD31C79C1268F7505
1DC8FC008EBF593AE5912E8B653975C13127E2B60A0BEF5FEF
21F5E2DFBCA3D5C86F4815DC69452E4FC7D122D650374E880B62294279F21A439D01
204DD5B4E544CACBC4EF869CDA30A6908FF7D1BDDB92B1ADE1E1C93CC614904894
OS signed with this key are not considered as valid by my 84+ (the 05 key works though). Could this come from a typo? FYI my calc is "antique" (I bought it in fall 2004 and 84+ series arrived on the market that year if I remember well). |
|
| Back to top |
|
|
brandonw
Advanced Member
Joined: 12 Jan 2007
Posts: 455
|
| Posted: 02 Sep 2009 11:48:15 am Post subject: |
|
|
They're considered valid by mine. Try changing or adding a byte and then re-assemble and re-sign (I've seen it fail in rare situations, similar to app signing).
What are you using to sign it? Perhaps you're not doing it right. |
|
| Back to top |
|
|
fullmetalcoder
Member
Joined: 01 Aug 2009
Posts: 139
|
| Posted: 02 Sep 2009 12:02:31 pm Post subject: |
|
|
brandonw wrote: They're considered valid by mine. Try changing or adding a byte and then re-assemble and re-sign (I've seen it fail in rare situations, similar to app signing).
Well, I've tried more than that. I tried swapping the p and q lines just in case, I tried re-assembling and re-signing periodically after having done modifications and it still does not work.
brandonw wrote: What are you using to sign it? Perhaps you're not doing it right.
packxxu and rabbitsign. I use the very same procedure with key 05 and it works like a charm so it can only be a key file or calc issue. |
|
| Back to top |
|
|
Mapar007
Advanced Member
Joined: 04 Oct 2008
Posts: 365
|
| Posted: 02 Sep 2009 12:10:57 pm Post subject: |
|
|
| Header issue? |
|
| Back to top |
|
|
brandonw
Advanced Member
Joined: 12 Jan 2007
Posts: 455
|
| Posted: 02 Sep 2009 01:01:41 pm Post subject: |
|
|
I'm not sure. I have no idea how to use RabbitSign to sign OSes (I'm guessing you're giving it an 8XU and telling it to re-sign it). Your header does have 80110A in it, right? Does the calculator say "Validating..." and then "ERROR!" or does it just refuse to transfer?
I'll re-sign and attach an OS to prove that the key works when I get a chance. |
|
| Back to top |
|
|
fullmetalcoder
Member
Joined: 01 Aug 2009
Posts: 139
|
| Posted: 02 Sep 2009 01:18:38 pm Post subject: |
|
|
brandonw wrote: I'm not sure. I have no idea how to use RabbitSign to sign OSes (I'm guessing you're giving it an 8XU and telling it to re-sign it).
That's it. Here is the full sequence I use (with key 05 as it is the only one I got to work with my calc) :
Code:
tools/multihex 00 build/page00.hex 1C build/page1C.hex | tools/encdos build/xos.hex
tools/packxxu -v 0.1 -h 255 build/xos.hex -t 83p -q 05 -o build/xos.8xu
rabbitsign -t 8xu -k tools/05.key build/xos.8xu -o xos.8xu
brandonw wrote: Does the calculator say "Validating..." and then "ERROR!" or does it just refuse to transfer?
It tries to validate and fail...
brandonw wrote: I'll re-sign and attach an OS to prove that the key works when I get a chance.
If you do that please post the key file you use. |
|
| Back to top |
|
|
brandonw
Advanced Member
Joined: 12 Jan 2007
Posts: 455
|
| Posted: 02 Sep 2009 07:32:18 pm Post subject: |
|
|
I know this is a stupid question, but you're trying it on an 84+ or 84+SE, right? Because that's all this key is good for. I only ask because of the "83p" in that command (but that's probably just how RabbitSign works; I don't know).
Last edited by Guest on 02 Sep 2009 07:33:00 pm; edited 1 time in total |
|
| Back to top |
|
|
fougere
Advanced Newbie
Joined: 07 Aug 2009
Posts: 56
|
| Posted: 02 Sep 2009 07:59:23 pm Post subject: |
|
|
packxxu requires either type '73' or '83p'. Using 83p will work for both 83's and 84's. (with the 05 key at least)
And i've had the same problem as fullmetalcoder: the 05 key works on my 84+SE, but not the new 0A key |
|
| Back to top |
|
|
brandonw
Advanced Member
Joined: 12 Jan 2007
Posts: 455
|
| Posted: 02 Sep 2009 11:17:07 pm Post subject: |
|
|
Whoops, I meant to say packxxu, not RabbitSign.
The OS at http://brandonw.net/crap/OS2.8xu was signed with the 0A key from above (I have p and q reversed, actually, but it makes no difference) and it validates just fine on a real 84+ and 84+SE with unmodified certificates.
I use Build8XU from OS2Tools (hacked up to use key 0A instead of key 05), not packxxu/RabbitSign. I really don't know where the problem is because you said the 05 key works, so RabbitSign must be working correctly. |
|
| Back to top |
|
|
fullmetalcoder
Member
Joined: 01 Aug 2009
Posts: 139
|
| Posted: 03 Sep 2009 03:52:29 am Post subject: |
|
|
| maybe free83p sort of breaks the certificate in regard to official keys? I'll try restoring the certificate and see if it helps... |
|
| Back to top |
|
|
brandonw
Advanced Member
Joined: 12 Jan 2007
Posts: 455
|
| Posted: 03 Sep 2009 10:00:21 am Post subject: |
|
|
When you install a certification block in the certificate (like Free83P does), every public key is overriden (including the 83+ and 84+ OS AND application keys), so all four (in addition to 05) are included in Free83P. It's theoretically possible that there's a typo which prevents 0A-signed OSes from being loaded, but I'm pretty sure I tested all the keys to make sure I didn't break any of them (because that would be a very bad thing, and you wouldn't be able to load the TI-OS again).
Scary stuff like this is why it's better to just have the factors. :)
EDIT: could you try sending that OS and see if you get "ERROR!" during validation?
Last edited by Guest on 03 Sep 2009 10:01:36 am; edited 1 time in total |
|
| Back to top |
|
|
fullmetalcoder
Member
Joined: 01 Aug 2009
Posts: 139
|
| Posted: 03 Sep 2009 11:08:36 am Post subject: |
|
|
brandonw wrote: When you install a certification block in the certificate (like Free83P does), every public key is overriden (including the 83+ and 84+ OS AND application keys), so all four (in addition to 05) are included in Free83P. It's theoretically possible that there's a typo which prevents 0A-signed OSes from being loaded, but I'm pretty sure I tested all the keys to make sure I didn't break any of them (because that would be a very bad thing, and you wouldn't be able to load the TI-OS again).
Scary stuff like this is why it's better to just have the factors. :)
EDIT: could you try sending that OS and see if you get "ERROR!" during validation?
Well I have switched back and forth between TIOS and third party OS signed with the 05 key a number of times so I guess the certificate is OK actually.
At which instant must I expect the error message to appear and what should I expect it to look like to be sure that it is a validation error? |
|
| Back to top |
|
|
brandonw
Advanced Member
Joined: 12 Jan 2007
Posts: 455
|
| Posted: 03 Sep 2009 01:04:29 pm Post subject: |
|
|
If you don't sign something correctly, typically it will say "Validating..." for a little while and then instead of immediately booting, you see either "Validation Error" or "ERROR!", I can't remember. (There is a "Validation Error" but I'm not sure if that's what you see.)
Did you already try it? I double-checked that OS by erasing the certificate on an 84+ and transferring it, and it was fine.
Last edited by Guest on 03 Sep 2009 01:05:24 pm; edited 1 time in total |
|
| Back to top |
|
|
|
Open Topic & United-TI Talk => General Open Topic
