Tool for exporting and importing high score times from Hydro Thunder Arcade as CSV files. This program works with a full drive image, not CHD or just files. The high score data is not stored as a normal file and only exists as raw data at the end of the drive.
WARNING - THIS MAY CAUSE A FACTORY RESET OF YOUR MACHINE WIPING SCORES, SETTINGS, AND AUDIT DATA
First, image your HDD. If you are using linux a simple dd image will do:
dd if=/dev/sd[letter here] of=./HT-HDD.imgMake a backup of this file, this is not a perfect process and the game will reset all data when things go wrong.
To read high score times run the program with the drive image as a parameter:
httt ./HT-HDD.imgIt will create a HT-HDD.img.csv file with all the high scores extracted.
To write high score times back to the drive image run the program with the drive image and CSV as a parameters: !!NOT YET USABLE FOR REAL!!
httt ./HT-HDD.img ./HT-HDD.img.csvNOTE: Currently this version can write times back to the game but they fail a validity check and the machine resets to factory defaults.
You can then image the drive back but it would be better to only write the
score data. A dd like with seek and skip can do this.
TODO - provide example
The data area that contains the high score table is towards the end of the HDD in an unformatted area, is internally referred to as "cmos" in HYDRO.EXE (1.00d). This CMOS area also contains other data that has not yet been decoded.
This data (or a checksum of it) may also be stored in the PC motherboard battery backed CMOS memory and flushed periodically to the non-volatile HDD, further reverse engineering required to determine this.
| Offset | Internal Game Name | Decoded | CSV export | Write back |
|---|---|---|---|---|
| 62B300h | _OperatorSettings | No | No | No |
| 622640h | _Hi_Score_Table | Yes | Yes | Partial |
| 622A50h | _Hud_SplitTimes | Yes | No | No |
| 6221FCh | _Diagnos_DiagnosticInfo | No | No | No |
| 620C18h | _Audits_Data | No | No | No |
| 62091Ch | _AiRabbit_WinHistoryData | No | No | No |
| Offset (1.00d) | Description |
|---|---|
| 0000h | Start of CMOS area (01 00 00 00 98 BA DC FE 01) |
| 0000h - 0063h | Unknown, some changes between 1.00d & 1.01b, suspect settings |
| 0064h - 0097h | Track difficultly |
| 0098h - 00CBh | AI difficultly |
| 00CCh - 016Fh | Unknown, significant change between versions (extra 11 bytes in 1.01b offseting all later data!) |
| 0170h - 057Fh | Hi_Score_Table |
| 0580h - 0583h | Unknown DWORD, no change between versions, maybe just a terminator (41 00 00 00) |
| 0584h - 0687h | Hud_SplitTimes |
| 0688h - 0733h | Unknown, changes between versions |
| 0734h - 0B1Dh | Unknown block with repeating structure every 108 bytes, much more data in 1.01b |
| 0B1Eh - 0C43h | Empty space (0x00 filled) |
| 0C44h - 0FC0h | Unknown block with repeating structure every 108 bytes, much more data in 1.01b |
| 0FC1h - 146Fh | Unknown, changes between versions |
| 146Fh | End of CMOS area |
Entry structure:
| diff : uint8 | null (0x00) | null (0x00) | null (0x00) |
|---|---|---|---|
| byte 0 | byte 1 | byte 2 | byte 3 |
Entry structure:
| boat : byte | initials : char[3] | seconds : single float |
|---|---|---|
| byte 0 | byte 1 | byte 2 |
Table structure:
| 'Ship Graveyard' high scores | 'Lost Island' high scores | ... | 'LOOP3' high scores |
|---|---|---|---|
| 1st | 2nd | ... | 10th |
| 0 | 123 | 4567 | 0 |
Entry structure:
| seconds : single float | |||
|---|---|---|---|
| byte 0 | byte 1 | byte 2 | byte 3 |
Table structure:
| 'Ship Graveyard' split times | 'Lost Island' split times | ... | 'LOOP3' split times |
|---|---|---|---|
| Split 1 | Split 2 | ... | Split 5 |
| 0123 | 0123 | ... | 0123 |
- The CMOS data contains data for 2 extra inaccessible tracks 'TEST' and 'LOOP3'.
- Tracks that have less than 5 split check points just leave the extra unused ones with their default time ~225 s (85 AB 61 43).
- More documentation of hidden data
- Provide examples of data
- Determine how data validity check is done and impliment it for modified scores