|
Typing Monkeys unite! (DONE, thanks everybody!)
#393292 - 02/13/22 08:40 AM
|
|
|
Some months have passed since the last time a collective effort was needed, but now we need help to type the data of the internal ROM of a Pacman handheld which was recently trojaned out. Here are the instructions:
Quote:
There is an increasing decimal counter on the left (the address) To the right of this there is another value (the data) again decimal
for EVEN addresses the data will be displayed near the center of the display for ODD addresses the data will be displayed closer to the right of the display (this is done to avoid 'half and half values' appearing mid-update)
valid address/data combinations will persist for several frames, so valid frames should be obvious and transitional frames can be ignored.
the address counter updates slightly before the data updates, don't let this catch you out, as mentioned above, data for odd addresses will always be display to the right of the screen so if in doubt you can double check tht way.
data is needed in the format
address, data, eg
016384, 000000, 016385, 000000, 016386, 000000, 016387, 000000, 016388, 000000, 016389, 016703, 016390, 016692, 016391, 016441, 016392, 016470,
the first digit in each column will always be 0 (these are 16-bit values so highest possible value is 065535)
files should be named the same as the video name eg. 0_part1_4000_43ff.txt for 0_part1_4000_43ff.mp4
etc.
be sure to familiarize yourself with the different number graphics, they are all distinct but be careful not to confuse 1 with 7, or 3 with 8 or 0. the address display is useful for learning the character shapes.
videos starting with the final one here are a bit darker than the first batch, although still readable
most of these videos cover a 1024 word range (1kwords) although due to recording boundaries some are split smaller.
ideally 2 people should independently do each video, as it makes mistakes more likely to be found
https://og.kervella.org/pacman/0_part1_4000_43ff.mp4 (1024 words) DONE (Team Europe) https://og.kervella.org/pacman/0_part2_4400_47ff.mp4 (1024 words) DONE (Osso) https://og.kervella.org/pacman/0_part3_4800_4bff.mp4 (1024 words) DONE (Osso) https://og.kervella.org/pacman/0_part4_4c00_4fff.mp4 (1024 words) DONE (Osso) https://og.kervella.org/pacman/0_part5_5000_53ff.mp4 (1024 words) DONE (Team Europe) https://og.kervella.org/pacman/0_part6_5400_55ff.mp4 (512 words) DONE (Team Europe) (display is a bit darker on ones below) https://og.kervella.org/pacman/1_part1_5600_57ff.mp4 (512 words) DONE (Team Europe) https://og.kervella.org/pacman/1_part2_5800_5bff.mp4 (1024 words) DONE (jrideburg) https://og.kervella.org/pacman/1_part3_5c00_5fff.mp4 (1024 words) IN PROGRESS (jrideburg) https://og.kervella.org/pacman/1_part4_6000_63ff.mp4 (1024 words) https://og.kervella.org/pacman/1_part5_6400_67ff.mp4 (1024 words) DONE (Team Europe) https://og.kervella.org/pacman/1_part6_6800_6bff.mp4 (1024 words) https://og.kervella.org/pacman/2_part1_6c00_6fff.mp4 (1024 words) https://og.kervella.org/pacman/2_part2_7000_73ff.mp4 (1024 words) https://og.kervella.org/pacman/2_part3_7400_77ff.mp4 (1024 words) https://og.kervella.org/pacman/2_part4_7800_7bff.mp4 (1024 words) https://og.kervella.org/pacman/2_part5_7c00_7fff.mp4 (1024 words) DONE (f205v) https://og.kervella.org/pacman/2_part6_8000_81ff.mp4 (512 words) DONE (f205v) https://og.kervella.org/pacman/3_part1_8200_83ff.mp4 (512 words) DONE (jrideburg) https://og.kervella.org/pacman/3_part2_8400_87ff.mp4 (1024 words) DONE 2x (jrideburg and Osso) https://og.kervella.org/pacman/3_part3_8800_8bff.mp4 (1024 words) DONE (jrideburg) https://og.kervella.org/pacman/3_part4_8c00_8dff.mp4 (512 words) (mostly empty) DONE (Team Europe)
total 19.5kwords of data
useful technique: use VLC on a separate laptop, hit space to pause / unpause video for each value while entering data on PC
If you're interested, please post hereunder which video you are going to tackle, to avoid too many people working on the same video and noone working on others. Thanks to everyone who'll help to preserve another system in MAME!
|
.
|
|
f205v |
Dumper
|
|
|
Reged: 09/21/03
|
Posts: 193
|
Loc: Switzerland
|
|
Send PM
|
|
|
Re: Typing Monkeys unite!
[Re: Osso1]
#393293 - 02/13/22 01:28 PM
|
|
|
|
|
Re: Typing Monkeys unite!
[Re: Osso1]
#393294 - 02/13/22 04:42 PM
|
|
|
About trying/typing some stuff...
Do I need to put a "space" at the end of each line? Like this?
"033877, 002624, "
...or... does it matter? Otherwise I´ll just type it like...
"033877, 002624," WITHOUT the "space"...
/jrideburg
|
|
|
f205v |
Dumper
|
|
|
Reged: 09/21/03
|
Posts: 193
|
Loc: Switzerland
|
|
Send PM
|
|
|
Re: Typing Monkeys unite!
[Re: jrideburg]
#393295 - 02/13/22 05:18 PM
|
|
|
It really doesn't matter. Files will be re-edited before being assembled into a .bin file. My suggestion is: save time by typing less!
|
Ciao
f205v
|
|
|
Re: Typing Monkeys unite!
[Re: f205v]
#393296 - 02/13/22 05:25 PM
|
|
|
Thanks!
By the way: I´ve started on a marathontyping, file no...
3_part2_8400_87ff.mp4
/jrideburg
|
|
|
f205v |
Dumper
|
|
|
Reged: 09/21/03
|
Posts: 193
|
Loc: Switzerland
|
|
Send PM
|
|
|
|
here is the file
|
Ciao
f205v
|
|
|
|
|
|
|
i finished 0_part1_4000_43ff
Edited by Team Europe (02/16/22 04:07 PM)
|
|
|
|
|
and here is 3_part4_8c00_8dff
Edited by Team Europe (02/16/22 04:08 PM)
|
|
|
|
Thank you so far
[Re: Osso1]
#393325 - 02/17/22 02:45 AM
|
|
|
results will be confirmed against an OCR version
both are expected to have errors, so having both a manual copy and OCR copy for each region will increase confidence.
the hope is that by doing this one 'the hard way' the hardware will be better understood so that in the future these things can be dumped more easily using the hardware and automated capture methods.
|
|
|
|
Re: Typing Monkeys unite!
[Re: Osso1]
#393334 - 02/17/22 05:07 PM
|
|
|
|
|
|
100% same as the one by jrideburg
|
.
|
|
|
|
|
|
|
|
|
Re: Typing Monkeys unite!
[Re: Osso1]
#393367 - 02/21/22 05:46 PM
|
|
|
These are being used to help validate / correct an OCR method now, they've already revealed a number of errors with the OCR, along with manual comparison of code, hopefully between the efforts we'll get a good set of data :-)
> Some months have passed since the last time a collective effort was needed, but now > we need help to type the data of the internal ROM of a Pacman handheld which was > recently trojaned out. > Here are the instructions: > > There is an increasing decimal counter on the left (the address) > To the right of this there is another value (the data) again decimal > > for EVEN addresses the data will be displayed near the center of the display > for ODD addresses the data will be displayed closer to the right of the display > (this is done to avoid 'half and half values' appearing mid-update) > > valid address/data combinations will persist for several frames, so valid frames > should > be obvious and transitional frames can be ignored. > > the address counter updates slightly before the data updates, don't let this catch > you out, > as mentioned above, data for odd addresses will always be display to the right of the > screen > so if in doubt you can double check tht way. > > data is needed in the format > > address, data, > eg > > 016384, 000000, > 016385, 000000, > 016386, 000000, > 016387, 000000, > 016388, 000000, > 016389, 016703, > 016390, 016692, > 016391, 016441, > 016392, 016470, > > the first digit in each column will always be 0 (these are 16-bit values so highest > possible value is 065535) > > files should be named the same as the video name > eg. 0_part1_4000_43ff.txt for 0_part1_4000_43ff.mp4 > > etc. > > be sure to familiarize yourself with the different number graphics, they are all > distinct > but be careful not to confuse 1 with 7, or 3 with 8 or 0. the address display is > useful for > learning the character shapes. > > videos starting with the final one here are a bit darker than the first batch, > although still readable > > most of these videos cover a 1024 word range (1kwords) although due to recording > boundaries some are split smaller. > > ideally 2 people should independently do each video, as it makes mistakes more likely > to be found > > > https://og.kervella.org/pacman/0_part1_4000_43ff.mp4 (1024 words) > https://og.kervella.org/pacman/0_part2_4400_47ff.mp4 (1024 words) > https://og.kervella.org/pacman/0_part3_4800_4bff.mp4 (1024 words) > https://og.kervella.org/pacman/0_part4_4c00_4fff.mp4 (1024 words) > https://og.kervella.org/pacman/0_part5_5000_53ff.mp4 (1024 words) > https://og.kervella.org/pacman/0_part6_5400_55ff.mp4 (512 words) > (display is a bit darker on ones below) > https://og.kervella.org/pacman/1_part1_5600_57ff.mp4 (512 words) > https://og.kervella.org/pacman/1_part2_5800_5bff.mp4 (1024 words) > https://og.kervella.org/pacman/1_part3_5c00_5fff.mp4 (1024 words) > https://og.kervella.org/pacman/1_part4_6000_63ff.mp4 (1024 words) > https://og.kervella.org/pacman/1_part5_6400_67ff.mp4 (1024 words) > https://og.kervella.org/pacman/1_part6_6800_6bff.mp4 (1024 words) > https://og.kervella.org/pacman/2_part1_6c00_6fff.mp4 (1024 words) > https://og.kervella.org/pacman/2_part2_7000_73ff.mp4 (1024 words) > https://og.kervella.org/pacman/2_part3_7400_77ff.mp4 (1024 words) > https://og.kervella.org/pacman/2_part4_7800_7bff.mp4 (1024 words) > https://og.kervella.org/pacman/2_part5_7c00_7fff.mp4 (1024 words) > https://og.kervella.org/pacman/2_part6_8000_81ff.mp4 (512 words) > https://og.kervella.org/pacman/3_part1_8200_83ff.mp4 (512 words) > https://og.kervella.org/pacman/3_part2_8400_87ff.mp4 (1024 words) > https://og.kervella.org/pacman/3_part3_8800_8bff.mp4 (1024 words) > https://og.kervella.org/pacman/3_part4_8c00_8dff.mp4 (512 words) (mostly empty) > > total 19.5kwords of data > > useful technique: use VLC on a separate laptop, hit space to pause / unpause video > for each value while entering data on PC > > If you're interested, please post hereunder which video you are going to tackle, to > avoid too many people working on the same video and noone working on others. > Thanks to everyone who'll help to preserve another system in MAME!
|
|
|
|
|
|
|
|
|
|
|
Edited by Osso1 (02/23/22 09:19 AM)
|
|
|
|
Starting on: 1_part2_5800_5bff
[Re: Osso1]
#393374 - 02/22/22 04:45 PM
|
|
|
|
|
|
just finished 0_part6_5400_55ff
will do "1_part1_5600_57ff" tomorrow!
|
|
|
|
|
Another one done.
"1_part2_5800_5bff"
https://og.kervella.org/pacman/1_part2_5800_5bff.mp4
It seems that my brain needs a break now. I can tell you this: being an accountant doesn´t help (much). You might think "oh yes it does", but... no, not really...
Perserverance and stubbornness DOES help though.
/jrideburg
|
|
|
|
|
and here is 1_part1_5600_57ff
|
|
|
f205v |
Dumper
|
|
|
Reged: 09/21/03
|
Posts: 193
|
Loc: Switzerland
|
|
Send PM
|
|
|
|
2_part5_7c00_7fff.mp4 done see attachment
|
|
|
|
Re: Typing Monkeys unite! (updated with progress up to 23/02/22)
[Re: Osso1]
#393381 - 02/23/22 03:54 PM
|
|
|
i will do next:
0_part5 1_part5
|
|
|
|
Starting on "1_part3_5c00_5fff"
[Re: Osso1]
#393383 - 02/23/22 05:50 PM
|
|
|
|
|
|
here is 0_part5_5000_53ff
|
|
|
|
|
here is 1_part5_6400_67ff
|
|
|
|
I THINK WE'RE GOOD!
[Re: Osso1]
#393400 - 02/24/22 07:55 PM
|
|
|
With the help of these ones done so far to find errors in the OCR data I have a ROM where the checksum of the data matches the internal checksum stored in the program. That's a good sign.
The 4000-47ff range isn't covered by the checksum, but given the high confidence at this point I think we're good there too.
> Some months have passed since the last time a collective effort was needed, but now > we need help to type the data of the internal ROM of a Pacman handheld which was > recently trojaned out. > Here are the instructions: > > There is an increasing decimal counter on the left (the address) > To the right of this there is another value (the data) again decimal > > for EVEN addresses the data will be displayed near the center of the display > for ODD addresses the data will be displayed closer to the right of the display > (this is done to avoid 'half and half values' appearing mid-update) > > valid address/data combinations will persist for several frames, so valid frames > should > be obvious and transitional frames can be ignored. > > the address counter updates slightly before the data updates, don't let this catch > you out, > as mentioned above, data for odd addresses will always be display to the right of the > screen > so if in doubt you can double check tht way. > > data is needed in the format > > address, data, > eg > > 016384, 000000, > 016385, 000000, > 016386, 000000, > 016387, 000000, > 016388, 000000, > 016389, 016703, > 016390, 016692, > 016391, 016441, > 016392, 016470, > > the first digit in each column will always be 0 (these are 16-bit values so highest > possible value is 065535) > > files should be named the same as the video name > eg. 0_part1_4000_43ff.txt for 0_part1_4000_43ff.mp4 > > etc. > > be sure to familiarize yourself with the different number graphics, they are all > distinct > but be careful not to confuse 1 with 7, or 3 with 8 or 0. the address display is > useful for > learning the character shapes. > > videos starting with the final one here are a bit darker than the first batch, > although still readable > > most of these videos cover a 1024 word range (1kwords) although due to recording > boundaries some are split smaller. > > ideally 2 people should independently do each video, as it makes mistakes more likely > to be found > > > https://og.kervella.org/pacman/0_part1_4000_43ff.mp4 (1024 words) DONE (Team Europe) > https://og.kervella.org/pacman/0_part2_4400_47ff.mp4 (1024 words) DONE (Osso) > https://og.kervella.org/pacman/0_part3_4800_4bff.mp4 (1024 words) DONE (Osso) > https://og.kervella.org/pacman/0_part4_4c00_4fff.mp4 (1024 words) DONE (Osso) > https://og.kervella.org/pacman/0_part5_5000_53ff.mp4 (1024 words) DONE (Team Europe) > https://og.kervella.org/pacman/0_part6_5400_55ff.mp4 (512 words) DONE (Team Europe) > (display is a bit darker on ones below) > https://og.kervella.org/pacman/1_part1_5600_57ff.mp4 (512 words) DONE (Team Europe) > https://og.kervella.org/pacman/1_part2_5800_5bff.mp4 (1024 words) DONE (jrideburg) > https://og.kervella.org/pacman/1_part3_5c00_5fff.mp4 (1024 words) IN PROGRESS > (jrideburg) > https://og.kervella.org/pacman/1_part4_6000_63ff.mp4 (1024 words) > https://og.kervella.org/pacman/1_part5_6400_67ff.mp4 (1024 words) IN PROGRESS (Team > Europe) > https://og.kervella.org/pacman/1_part6_6800_6bff.mp4 (1024 words) > https://og.kervella.org/pacman/2_part1_6c00_6fff.mp4 (1024 words) > https://og.kervella.org/pacman/2_part2_7000_73ff.mp4 (1024 words) > https://og.kervella.org/pacman/2_part3_7400_77ff.mp4 (1024 words) > https://og.kervella.org/pacman/2_part4_7800_7bff.mp4 (1024 words) > https://og.kervella.org/pacman/2_part5_7c00_7fff.mp4 (1024 words) DONE (f205v) > https://og.kervella.org/pacman/2_part6_8000_81ff.mp4 (512 words) DONE (f205v) > https://og.kervella.org/pacman/3_part1_8200_83ff.mp4 (512 words) DONE (jrideburg) > https://og.kervella.org/pacman/3_part2_8400_87ff.mp4 (1024 words) DONE 2x (jrideburg > and Osso) > https://og.kervella.org/pacman/3_part3_8800_8bff.mp4 (1024 words) DONE (jrideburg) > https://og.kervella.org/pacman/3_part4_8c00_8dff.mp4 (512 words) (mostly empty) DONE > (Team Europe) > > total 19.5kwords of data > > useful technique: use VLC on a separate laptop, hit space to pause / unpause video > for each value while entering data on PC > > If you're interested, please post hereunder which video you are going to tackle, to > avoid too many people working on the same video and noone working on others. > Thanks to everyone who'll help to preserve another system in MAME!
|
|
|
f205v |
Dumper
|
|
|
Reged: 09/21/03
|
Posts: 193
|
Loc: Switzerland
|
|
Send PM
|
|
|
Re: I THINK WE'RE GOOD!
[Re: Haze]
#393409 - 02/25/22 09:39 AM
|
|
|
I'm a proud monkey!
|
|
|
gregf |
Ramtek's Trivia promoter
|
|
|
Reged: 09/21/03
|
Posts: 8608
|
Loc: southern CA, US
|
|
Send PM
|
|
|
from ‘blurst to worst’
[Re: f205v]
#393411 - 02/25/22 11:56 AM
|
|
|
|
|
It's Alive! (sort of)
[Re: Haze]
#393416 - 02/25/22 01:15 PM
|
|
|
https://www.youtube.com/watch?v=w-cfbSnwURs
This boots and run an attract loop, obviously still a lot of unknowns with this, but hopefully having the code running will allow better techniques to be developed for extracting the internal ROMs (every single game, and revision of each game has a different internal ROM)
> With the help of these ones done so far to find errors in the OCR data I have a ROM > where the checksum of the data matches the internal checksum stored in the program. > That's a good sign. > > The 4000-47ff range isn't covered by the checksum, but given the high confidence at > this point I think we're good there too. > > > > Some months have passed since the last time a collective effort was needed, but now > > we need help to type the data of the internal ROM of a Pacman handheld which was > > recently trojaned out. > > Here are the instructions: > > > > There is an increasing decimal counter on the left (the address) > > To the right of this there is another value (the data) again decimal > > > > for EVEN addresses the data will be displayed near the center of the display > > for ODD addresses the data will be displayed closer to the right of the display > > (this is done to avoid 'half and half values' appearing mid-update) > > > > valid address/data combinations will persist for several frames, so valid frames > > should > > be obvious and transitional frames can be ignored. > > > > the address counter updates slightly before the data updates, don't let this catch > > you out, > > as mentioned above, data for odd addresses will always be display to the right of > the > > screen > > so if in doubt you can double check tht way. > > > > data is needed in the format > > > > address, data, > > eg > > > > 016384, 000000, > > 016385, 000000, > > 016386, 000000, > > 016387, 000000, > > 016388, 000000, > > 016389, 016703, > > 016390, 016692, > > 016391, 016441, > > 016392, 016470, > > > > the first digit in each column will always be 0 (these are 16-bit values so highest > > possible value is 065535) > > > > files should be named the same as the video name > > eg. 0_part1_4000_43ff.txt for 0_part1_4000_43ff.mp4 > > > > etc. > > > > be sure to familiarize yourself with the different number graphics, they are all > > distinct > > but be careful not to confuse 1 with 7, or 3 with 8 or 0. the address display is > > useful for > > learning the character shapes. > > > > videos starting with the final one here are a bit darker than the first batch, > > although still readable > > > > most of these videos cover a 1024 word range (1kwords) although due to recording > > boundaries some are split smaller. > > > > ideally 2 people should independently do each video, as it makes mistakes more > likely > > to be found > > > > > > https://og.kervella.org/pacman/0_part1_4000_43ff.mp4 (1024 words) DONE (Team > Europe) > > https://og.kervella.org/pacman/0_part2_4400_47ff.mp4 (1024 words) DONE (Osso) > > https://og.kervella.org/pacman/0_part3_4800_4bff.mp4 (1024 words) DONE (Osso) > > https://og.kervella.org/pacman/0_part4_4c00_4fff.mp4 (1024 words) DONE (Osso) > > https://og.kervella.org/pacman/0_part5_5000_53ff.mp4 (1024 words) DONE (Team > Europe) > > https://og.kervella.org/pacman/0_part6_5400_55ff.mp4 (512 words) DONE (Team Europe) > > (display is a bit darker on ones below) > > https://og.kervella.org/pacman/1_part1_5600_57ff.mp4 (512 words) DONE (Team Europe) > > https://og.kervella.org/pacman/1_part2_5800_5bff.mp4 (1024 words) DONE (jrideburg) > > https://og.kervella.org/pacman/1_part3_5c00_5fff.mp4 (1024 words) IN PROGRESS > > (jrideburg) > > https://og.kervella.org/pacman/1_part4_6000_63ff.mp4 (1024 words) > > https://og.kervella.org/pacman/1_part5_6400_67ff.mp4 (1024 words) IN PROGRESS (Team > > Europe) > > https://og.kervella.org/pacman/1_part6_6800_6bff.mp4 (1024 words) > > https://og.kervella.org/pacman/2_part1_6c00_6fff.mp4 (1024 words) > > https://og.kervella.org/pacman/2_part2_7000_73ff.mp4 (1024 words) > > https://og.kervella.org/pacman/2_part3_7400_77ff.mp4 (1024 words) > > https://og.kervella.org/pacman/2_part4_7800_7bff.mp4 (1024 words) > > https://og.kervella.org/pacman/2_part5_7c00_7fff.mp4 (1024 words) DONE (f205v) > > https://og.kervella.org/pacman/2_part6_8000_81ff.mp4 (512 words) DONE (f205v) > > https://og.kervella.org/pacman/3_part1_8200_83ff.mp4 (512 words) DONE (jrideburg) > > https://og.kervella.org/pacman/3_part2_8400_87ff.mp4 (1024 words) DONE 2x > (jrideburg > > and Osso) > > https://og.kervella.org/pacman/3_part3_8800_8bff.mp4 (1024 words) DONE (jrideburg) > > https://og.kervella.org/pacman/3_part4_8c00_8dff.mp4 (512 words) (mostly empty) > DONE > > (Team Europe) > > > > total 19.5kwords of data > > > > useful technique: use VLC on a separate laptop, hit space to pause / unpause video > > for each value while entering data on PC > > > > If you're interested, please post hereunder which video you are going to tackle, to > > avoid too many people working on the same video and noone working on others. > > Thanks to everyone who'll help to preserve another system in MAME!
|
|
|
|
Re: I THINK WE'RE GOOD!
[Re: f205v]
#393417 - 02/25/22 05:02 PM
|
|
|
Me too. Although... I´m glad that I can take a (hopefully) loooong rest from these numbers.
When you think you´re seeing strange things in the shape of NUMBERS you really start to question yourself... haha
Happy to be of service!
/jrideburg
|
|
|
|
Re: It's Alive! (sort of) (update)
[Re: Haze]
#393418 - 02/25/22 09:08 PM
|
|
|
|