Ultimate Amiga
Games Corner => The Crypt of Bloodwych => Bloodwych Editors and Modifications => Topic started by: Hungry Horace on April 05, 2014, 12:20:54 PM
-
Here's a challenge foor anyne playing around in the Bloodwych code.
Its easy to derive a table of all the Monsters, $00-$6D or whatever it is... but how about changing those monsters?
we know that the sprites are broken up into a number of body parts (arms, legs, heads torsos) most of whcih can be 're-coloured' ... so in theory there must be a tble that defines each monster (and character) .... use X arms, colour 4, Y legs Colour 5 etc.
I challenge anyone to find this table and open it up for editing! it would be wonderful to create a few new exclusive monsters (i would like to make a white witch who uses Eleanors head for example).
-
I've been trying to crack this but no luck yet :(
I did find the table of how monsters are pieced together with there Sprite x,y,Height,Width but I don't think that helps anyone :)
-
I've been trying to crack this but no luck yet :(
I did find the table of how monsters are pieced together with there Sprite x,y,Height,Width but I don't think that helps anyone :)
it might be very near that.... can you tell me where it is, how it works, and the example data so that i can find it on different versions?
edit: (again) ... plus i would like to ensure we document everything possible! (this will help disassemble the entire code)
is there anything else like this you have found which is not documented?
-
i think i have foudn the bit you are talking about.... right after the spellbook text? ("maryhadalittlelamb ... etc")
-
Yeah that's it, you can actually change the graphics in front of the player also ie I was messing with the first shop keeper and replaced the open wooden door with a wooden wall, not the map data but what graphic in memory is used, also not sure if you know but in the monster type ie 15 = first shop keeper you can actually replace the type to say a fireball I'll find the exact values you can use tomorrow.
Sent from my iPad using Tapatalk
-
Oh also like you I want to try map out what we know and what we don't, problem I have is I do most mine using steem the ST emulator with Cheat Engine and all the data is backwards :s would be good though to have the complete disk image as a reference and say from x to y is switch data, then unknown, then x to y is map info. I can start look at the bits we don't know yet
Sent from my iPad using Tapatalk
-
Why is the data backwards??
I like using the amiga binary myself, as this is the most "compact" set of day for the game... A disk image or memory dump will have unneeded odd sections or even just empty data.
If there isn't a free hex editor I can find which allows marking up / highlights, then I can always write a VBA spreadsheet again that will do it for us.
-
Hi Horace,
Sorry I didnt mean work off a memory dump :) I just mean work from a clean file and so I can search for the values in memory and try and work out what they are doing.
Im not sure on the reason STEEM reads the data backwards it seems to be a unique thing to that emulator it just happens to be the nicest emulator i find to use.
I know you work off an Amiga Image so do you have one we can work on together and mark out what we know and dont know.
-
I have attached the 'last' version of amiga BW, which i will now adapt my editor to use instead of the older version i had before (i suspect there may be bug fixes in this one)
I would love to verify if the data extracted from the files i sent you patches directly back into the ST version without problem and/or compare for any differences!
Also added the BEXT file from the amiga too :)
Do you use OSX or Windows or Linux? I am trying to find a good hex editor we can share a highlighting mark-up system on....
-
I use Windows :)
Sent from my iPad using Tapatalk
-
Horace: I've found this http://www.wxhexeditor.org/home.php
It allows tagging (Import and Export) so hopefully it does what we need. There also seems to be a Windows/OSX and Linux Version
Sorry that apps pretty crappy, Im using Hex Editor Neo which is a 14 day trial but has nice bookmarking. I'll see what I can find in 14 days :)
-
i tried that one you linked to... it has some ok functions but it really needs tidying up, and it sometimes slows-up really bad on OSX.
I have Parallels on my machine, so i can run windows applications, it is just not ideal!
-
Attached is my Hex Editor Neo Bookmarks
I've found a few extra bits so I now know where the Graphics data is stored (but not the end yet) and some other bits. I want to try roughly map out the areas then we can document them in more detail.
-
hi guys,
I have been adapting my editor a bit further as a result of the recent findings for the above (i.e. finding monster colours etc) - obviously, my concentration has been on advanced monster editing.
Firstly, I have noticed that the X coordinate value for the Lord Of Entropy in the Zendick Tower has 'bit 7' set. It appears that using the chaos crystal in any chaos crystal socket clears this bit, and therefore causes the creature to be "summoned" into being (usually about 10-20 seconds later)
As far as i can tell, this is limited to Monster Number 0 in Zendicks tower only.
Another thing i noticed is that on the Extended levels a lot of the monsters have Bit 8 set in their "type field"...
from what i can tell the bits for MOnster type are as follows;
Bit 0 = denotes if spell-caster or not.
Bit 1 = denotes if "drone" or not (i.e.. does not hunt down players)
Bit 2 = "arc bolt shooter" (all other bits ignored i think, if used)
Bit 3 =. .... ??? Mystery setting on extended levels!!
Any thoughts on what this is used for? It appears a *lot* - i was wondering if it was connected to special items to be dropped by the monsters?
Any thoughts / research welcomed!
-
i have also just discovered that on Extended levels the colour data for each character is extended by an extra 4 bytes.
This probably determines the colours used on the characters Avatar when recruited into your team, but I am only guessing at this.
-
another set of palette data found.
It starts at $9AD8 technically, but most useable values are at $9ADB (which is where the code links to i think)
The first four values are for an Illusion, and the values afterwards get used for the Summon creatures, but appear to be re-used for other monsters (dragons, beholders etc) so there must be other tables which tell the game "when this creature reaches level X, use colour combination Y"
Data as follows;
00000708 0003040E 0008040E 0005060E
00090A0B 00090C0B 000A0B0D 000B0D0E
000C0B0D 00020304 0008040D 0005060D
00070804 0007080D
-
at $9A34, 8 bytes.... determine which colour map to use for each change with the Summon creatures :)
00 01 02 03 04 05 ...
no idea where i might find the data for the other monsters though.... (pretty sure i can find it with time, even if i have to do it the hard way!)
edit:
untested, but i suspect these are the other locations....
$9F20 … crabs
$A1AC … beholder
$A3A6 .. Dragons
$A52E .. Behemoth
(somewhere the 'level change' values must be there too!)
-
Nice one Horace and excellent timing :)
-
looks like the locations for each different versions (BW SPS 439/1927 , BW SPS 102 and BEXT SPSP43) are these:
439 102 bext
palettes: $9AD8 $9B8E $A8EE
summon: $9A34 $9AEA $A84A
beholder: $9E28 $9EDE $AC1E
behemoth: $A1AA $A260 $AFc2
crab: $9B9C $9C52 $A9B4
dragon: $A022 $A0D8 $AE3A
-
more extended levels data.... the colours of the Entropy / Chaos Guardians:
@ $AFCA in the BEXT binary
00 02 05 06 00 05 06 0D 00 09 0A 0B 00 0A 0B 0D 00 07 08 04 00 08 04 0D 00 09 0C 0B 00 0C 0B 0D
There's a 8 bytes of data before it, that looks like more palette stuff, but it could be anything.
-
Horace any chance you can see if you can find the palettes for the amour?
-
i assume you mean the palette used by the armour being worn, on the characters (i.e. in the 3D view) rater than the actual armour object held in people's pockets?
shouldnt be too difficult now i know the format/style.
Been trying to establish other items for possible edits. I'd like to know how it determines that the traders and zendick are positioned in the centre of the floor rather than in one of the 4 edge spaces, but my suspicion is that this will be hard-coded (which doesnt mean it cant be changed, it just wont suit "normal" editing)
Similarly, I want to establsh the "level offset" at which each monster starts changing colour as they go up a level...
Thoughts /requests on possible areas for exploration are welcome.
-
i assume you mean the palette used by the armour being worn, on the characters (i.e. in the 3D view) rater than the actual armour object held in people's pockets?
shouldnt be too difficult now i know the format/style.
Correct :)
Been trying to establish other items for possible edits. I'd like to know how it determines that the traders and zendick are positioned in the centre of the floor rather than in one of the 4 edge spaces, but my suspicion is that this will be hard-coded (which doesnt mean it cant be changed, it just wont suit "normal" editing)
I believe your right, though im not sure where to find this but ill have a look
-
looks like the locations for each different versions (BW SPS 439/1927 , BW SPS 102 and BEXT SPSP43) are these:
439 102 bext
palettes: $9AD8 $9B8E $A8EE
summon: $9A34 $9AEA $A84A
beholder: $9E28 $9EDE $AC1E
behemoth: $A1AA $A260 $AFc2
crab: $9B9C $9C52 $A9B4
dragon: $A022 $A0D8 $AE3A
Horace can you post the palette line for summons so I can check I have the right values, also what's the different 439/102? Different disks?
I think I've accidentally found the table on how the summons are put together
-
the 102/439/1927 numbers refer to the SPS preservation disks:
http://www.softpres.org/
the game binary on 102 seems to be slightly different and an earlier version, so i prefer to work from the 439 version... the 1927 version has only a small difference in a few addresses but seems mostly the same.
the summons palette i posted above?
00000708 0003040E 0008040E 0005060E
00090A0B 00090C0B 000A0B0D 000B0D0E
000C0B0D 00020304 0008040D 0005060D
00070804 0007080D
i know this is the right set, because 00 00 07 08 is black, black, blue, blue, as used by the Illusion versions :)
Also my editor uses these to create its graphics set!!
No luck on the armour palettes (so far) i'm sorry to say...
-
Thanks Horace,
If you add 189 Bytes to the Summon table you get the details of how the summon is put together
20 00 67 0c 98 32 20 00 9a 32 20 01 60 00 0f 80
4e 75 00 01 02 03 04 05 06 07 15 11 0d 0c fe 00
2e 26 1f 1a 15 10 00 01 80 01 14 14 10 10 10 0b
0c 0c 0a 0b 0b 08 09 f9
09 f9 = Left Arm X/Y
I believe they are for different distances and also the sprite height, it has some strange results messing around with the data as it looks like they use some kind of calculation to work out the positioning and playing around here can mess it up.
-
Sorry Horace but any chance of a new dump from your editor of clean data or if you can post up the palettes Im having issues finding them :(
-
Hi mate,
I've attached a clean export for you... i think these are right, but i kind of rushed adding the export function to the editor to do this, so they are un-checked!! (they will be the final 9 files created)
-
Cheers Horace :)
-
Horace: looks like all your colour/palette files are wrong :(
-
I've double checked and they seem right? (although there is one small mistake I will fix)
Are you sure you are familiar with how to use them?
monster.colours is a list of possible palettes, each entry of which are four bytes long.
00 00 07 08 | 00 03 04 0E | 00 08 04 0E | 00 05 06 0E | 00 09 0A 0B | 00 09 0C 0B | 00 0A 0B 0D | 00 0B 0D 0E | 00 0C 0B 0D | 00 02 .... errr.
The first entry, I call "palette -1" because it is used only on the Illusion creatures.
00 00 07 08
- colour 0 becomes colour entry 0
- colour 4 becomes colour entry 0
- colour 8 becomes colour entry 7 (blue)
- colour C becomes colour entry 8 (another blue)
The next entry is palette number 0, and is used on the most basic (grey) summon creature
00 03 04 0E
- colour 0 becomes colour entry 0
- colour 4 becomes colour entry 3 (mid grey)
- colour 8 becomes colour entry 4 (lightest grey)
- colour C becomes colour entry E (white)
I think I may have left-off some bytes though, as I don't seem to have enough entries above, but the entries are definitely correct.
The file Summon.colours (or any other mosnter name.colours) then tells you which palette to use for each "level" (or 'grade') of Summon.... this is how I know the files are right, because this one is really easy to spot!
00 01 02 03 04 05 06 07
For the lowest grade summon, use palette entries 0 (mostly grey)
For the number 1 grade summon, use palette entries 1 (mostly blue/grey)
For the number 2 grade summon, use palette entries 2 (mostly green)
etc. etc.
For the other monsters, they aren't quite as sequential, but the principle is the same.
You should bare in mind though, that a Grade 0 monster, does not relate to a level 0 monsters.... each one has its own "starting offset" value, i.e. they might not start changing colour until level 3 etc.
Hope this helps
-
monsters.colours I have accidentally made 26 bytes in length.
I *think* it should be 38 (maybe less)
the correct full set of values are:
00 00 07 08 | 00 03 04 0E | 00 08 04 0E | 00 05 06 0E
00 09 0A 0B | 00 09 0C 0B | 00 0A 0B 0D | 00 0B 0D 0E
00 0C 0B 0D | 00 02 03 04 | 00 08 04 0D | 00 05 06 0D
00 07 08 04 | 00 07 08 0D
:)
-
Thanks to the disassembly of the 68k source by BruceUncle, I've made some interesting finds.
I was right, that there is some hard-coding going on to determine if a character is placed in the central position:
009728 cmpi.b #$15,-$0017(a3)
00972E beq.s lbC009750
009730 cmpi.b #$16,-$0017(a3)
009736 beq.s lbC009750
009738 cmpi.b #$40,-$0017(a3)
00973E beq.s lbC009750
009740 cmpi.b #$67,-$0017(a3)
009746 bcc.s lbC009750
There is also some similar code which would allow me to have any special characters I produce to respond with a "special" name... like Zendik names himself as lord of creation. There is also again similar code that tells Zendik to 'hold' the permit which he drops.
Now, I really like the idea of having one special character like this per tower. The problem is, this cannot be added to my editor, and therefore cannot be added to other (non amiga) versions, without writing bespoke code... thankfully, the code required isn't that difficult, and using WHDload to create unique routines for this purpose would actually make this quite easy....
however, I am concerned about the impact of such a heavy edit, and wondered what the thoughts of 'the public' were?
-
Thanks Horace my mistake, I'll check and try again :)
-
I have been watching far too much Game of Thrones....
Character $11 adapted to have their own name text... (yes, i had to write new code into the game for this)
I am going to need some help form BruceUncle though to be sure i put it into an area that wont be wiped. (preferably i would like to expand the executeable size slightly)
-
Looking good!
Looks like your be busy now :) unfortunately the stuff Bruce posted means nothing too me :( I'm hopeful Bits data will be more to a format I can understand but this is all excellent
-
well... I am thinking, it is possible that some code can be shared between Amiga and ST versions, however, this will depend heavily on memory locations, and whether there is any "spare" room in the game binaries to do this.
MadMunky - I think you may have sent me it before, but do you have the main game binary from the ST version at all you can send me? I do not mind if it is a cracked version either.
I am thinking I will however maybe sacrifice the idea of having Book of Skulls work on both machines, in favour of having a more involved update/mod, unless there are any strong objections to this?
It would be perfectly possible to write some code that looks for the correct ASM and translates it back into useable parameters to be read by an editor or your remake
(so you would know that there are names for characters 10,1A,40 etc, and they are connected to specific text.) and this is something I will allow for by adhering to specific "rules" in my new game code.
-
I cant remember what I gave you before :(
I've attached the ST Blood files but they look like they are compressed, I honestly cant remember how/if I uncompressed them before :(
Or maybe I did a memory snap shot and extracted the data (attachment is not a zip but raw ST as I had to rename it to upload)
-
Horace are you able to find anything in the code about how many hit points are assigned to which monster level?
-
Horace are you able to find anything in the code about how many hit points are assigned to which monster level?
I would need to understand the "unpacked" monster data... which if $10 in length.
it contains the monsters location, type, current status ie. are they attacking, or communicating etc) , their direction, what object they hold and various other bits like this.
It must also contain their hit-points, which will be assigned during the "unpacking" (which it think is done when you enter each tower)
If I had this block understood more, I could start looking at what happens in the unpack process....
Labeling up the ASM source code should make this much easier, so I am hoping BruceUncle can give me some pointers as to how to add new labels to his resource file.
-
Horace: good find on the other thread I didn't want to pollute it with comments so will post here,
Also do you know why the characters stats have the Spell points set to say 9 for blodwyn but as soon as the game starts this gets changed to 6, do you know if there was any reasoning to this.
-
Horace: good find on the other thread I didn't want to pollute it with comments so will post here,
Also do you know why the characters stats have the Spell points set to say 9 for blodwyn but as soon as the game starts this gets changed to 6, do you know if there was any reasoning to this.
no idea yet, but it will be easy enough to find the code that does it.... like the monsters it needs the character stats section correctly mapped out though.
-
01 23 11 0D 0D 23 23 1F 1F 06 09 05 80 00 00 00
C7 FF 00 00 00 00 07 09 02 00 03 00 00 00 00 00
This is Blodwyn's starting data, unedited from the champions.stats file.
immediately after starting the game (as blodwyn) the data becomes this:
01 23 11 0D 0D 23 23 1F 1F 06 06 05 80 00 00 00
C7 00 00 FF 00 00 07 09 02 E0 03 00 0B FF 00 00
in the other thread I will post the code that does this. The address which this happens at is stored in that code at A4 .... this is important. Each of the above items is then referenced like this:
0(a4) = first byte above (level)
5(a5) = 6th byte (stamina)
a(a5) = 10th byte (magic)
we should really have an understanding of what every byte means in the 'live' data. maybe the code can give some insight into this.
From what i can see, it simply moves the value from the 11th byte, to the 10th byte address... rendering the value of '9' for Blodwyn's magic as meaningless. (unless the value is 'read' for some purpose prior to this routine)
-
We already know what pretty much all of this does it's attack timers and cool down timers,xp and amount of spells owed by the fairy :)
It's only this value we don't really understand
-
can you provide that info here please? It might help me understand the code more.
I am trying to work out a few bits atm, but i still have a lot of blanks
Some of it is probably the same with the monsters too :)
-
Here we go http://www.ultimateamiga.co.uk/index.php/topic,8806.msg45087.html#msg45087
-
MM - i moved your post to a thread specifically about the champion data, and will try and match your info above to what i've posted in that thread, hopefulyl all the blanks can be cleared.
regarding monsters, there wont be another table. It will simply do some maths in that code i posted before to derive a strength / HP value based on the level.
Edit: just re read your message on this... Sorry, I thought I covered it before but there are two types of monster tables. One is the "stored" version with 6 bytes per monster. Pretty easy to handle stuff in those (except the special set bit on extended levels which I don't understand yet) and the "unpacked" monster data which is 16 bytes long per monster. "Live" spells are also stored in this block. (They are items with a value >$80)
I suspect the data held in the unpacked block is similar to the champion data, including remaining HP, current action (arms), speed etc
It certainly contains info about held items and teaming and x/y etc which are all very easy to understand
-
Horace,
I've found the Amour palette, I'm still trying to work out how its structured
I'm also not sure yet if that's the right beginning/end of the block of data but (0B 0A 09 0B = Leather amour)
Offset in 439 is A81E
00 08 06 0b 0b 0a 09 0b 09 0a 0b 00 82 83 00 00
0b 0a 09 0b 09 0a 0b 81 09 0a 0b 00 81 09 0a 0b
80 09 0a 0b ff ff ff ff 0e 04 02 03 82 04 02 83
82 04 83 0c 0e 04 02 03 81 04 02 80 03 04 0e 00
81 02 04 0e 80 03 04 0e ff ff ff ff 0e 04 03 03
82 04 04 83 00 04 08 0c 0e 04 03 03 81 04 04 80
03 04 0e 00 81 03 04 0e 80 02 04 0e 0b 03 03 04
0b 0a 07 08 06 05 0b 0d 0b 0a 09 0c 06 05 05 06
0b 0a 09 0c 0e 00 0b 0d 0b 0a 05 06 0b 0a 07 08
0b 0a 0b 0d 0b 0a 05 06 0b 0a 09 0c 0b 0a 05 06
00 00 07 08 09 00 07 08 09 00 09 0c 08 04 04 04
0b 0a 0b 0d 81 0a 09 09 81 02 04 82 81 03 04 82
82 09
-
Nice one!
I guess there is a pattern relating to bodies/armour type etc?
I mean, it must have to do leather armour for each body type (Bloodwyn type, human male type, female type etc) then repeat for chain mail etc....
Just wondering what order that is in!
-
I found the data which determines what 'grade' / colour of monster to use for each level. As i suspected, its a simple offset value which is plugged into the D3 register when it looks for the colour data for each monster type.... as a result it's hard-coded for each one, that means "manual" extraction of the data for each game (BW, BEXT, and ultimately BOS)
the offsets for BW are;
- Summons, 2
- Beholders, 4
- Behemoths, 6
- Crab, 2
- Big Dragons, 9
- Small Dragons, 3
-
Horace,
I've found the Amour palette, I'm still trying to work out how its structured
I'm also not sure yet if that's the right beginning/end of the block of data but (0B 0A 09 0B = Leather amour)
Offset in 439 is A81E
00 08 06 0b 0b 0a 09 0b 09 0a 0b 00 82 83 00 00
0b 0a 09 0b 09 0a 0b 81 09 0a 0b 00 81 09 0a 0b
80 09 0a 0b ff ff ff ff 0e 04 02 03 82 04 02 83
82 04 83 0c 0e 04 02 03 81 04 02 80 03 04 0e 00
81 02 04 0e 80 03 04 0e ff ff ff ff 0e 04 03 03
82 04 04 83 00 04 08 0c 0e 04 03 03 81 04 04 80
03 04 0e 00 81 03 04 0e 80 02 04 0e 0b 03 03 04
0b 0a 07 08 06 05 0b 0d 0b 0a 09 0c 06 05 05 06
0b 0a 09 0c 0e 00 0b 0d 0b 0a 05 06 0b 0a 07 08
0b 0a 0b 0d 0b 0a 05 06 0b 0a 09 0c 0b 0a 05 06
00 00 07 08 09 00 07 08 09 00 09 0c 08 04 04 04
0b 0a 0b 0d 81 0a 09 09 81 02 04 82 81 03 04 82
82 09
Horace, you much better decoding these kind of things than me, this is what I know..
09 0a 0b - Male (V-Neck) Leather Legs
0a 0b 00 81 - Male (V-Neck) Leather Chest
80 09 0a 0b - Male (V-Neck) Leather Arms
0b 0a 09 - Blodwyn Style Leather Legs
0b 0a 09 0b - Blodwyn Style Leather Chest
81 09 0a 0b - Blodwyn Style Leather Arm
I believe all the bytes up to the FF FF FF FF are all Leather Armour values then Iron Value I cant though find the Females leg entries :(
00 08 06 0b 0b 0a 09 0b 09 0a 0b 00 82 83 00 00
0b 0a 09 0b 09 0a 0b 81 09 0a 0b 00 81 09 0a 0b
80 09 0a 0b ff ff ff ff
Iron Amour
0e 04 02 03 82 04 02 83
82 04 83 0c 0e 04 02 03 81 04 02 80 03 04 0e 00
81 02 04 0e 80 03 04 0e ff ff ff ff
Found another bit...
This seems to link a Armour to a Body/Palette type
00 00 03 04 04 02 03 03 03 02 04 02 01 02 03 03
03 03 01 0b 02 03 03 03 0c 0c 0c 0c 0c 03 03 03
08 08 04 04 04 0a 0d 0d 0d 0d 07 07 07 07 07 07
03 02 02 0b 0b 03 03 03 03 03 03 03 03 03 03 03
03 03 04 07 07 07 07 07 01 01 01 01 01 0b 07 02
02 00 00 00 00 00 00 16
00 = Blodwyn
01 = V-Neck
04 = Female Legs
05 = Male Chain
06 = Female Skirt
0A = Demon
0B = Demon Chain
FF = Mr Flay