Main > Software Forum

HiToText (Version 2010.11.4).

<< < (15/117) > >>

tspeirs:
Hi Firecrypts.

As HK says, I just bought another server, and am now able to give you the .net hosting, and help with whatever you need in regards to that. More than happy to do it.

I prefer to do things by email, so drop me a note that way.

Tom.

tspeirs:
I also have a coding suggestion which I think will help ensure success going forward, meaning less work for you and updating the source code less, and that is not to code the decipering in c# classes, but code a binary processing api in the app and a basic scripting format for decipering each game. Say each game would have a .hid file, which would be a little script that the app uses to process the binary data. From what I can make out from the completed games this should be doable. Youd basically have an algorithim/script for each game in a .hid file. So the source code would only be updated when the algorithim/scripting support needed to be improved to support a particular game. Im a software architect, and thats what Id recommend from a top level perspective. Although I know talking is easier than doing.

Fyrecrypts:
I'm not against the idea, as I'm sure it would allow a larger number of people to contribute. The only problem is so few games so far haven't required some sort of weird totally different piece of code here or there to work correctly. Marble Madness uses 2 bytes to store 3 characters for the name, 1942 stores the scores in whatever order it feels like storing them and using a byte to determine where it shows up on the list of scores, some games like Donkey Kong Jr. aren't symmetrical when it comes to scores, and the list goes on. :(

It would be an ideal solution if we could come up with a format that could deal with these quirks, some sort of byte mapping solution maybe. For example take Marble Madness. It's a 50 byte file with the current hiscore.dat, and stores 10 scores. It could be listed in a .hid file as:

"0-2", "Score1", "Reverse,HexToInt"
"3-4", "Name1", "MarbleName"
"5-7", "Score2", "Reverse,HexToInt"
....
"2E-30", "Score10", "Reverse,HexToInt"
"31-32", "Name10", "MarbleName"

While this seems like overkill matching up every single byte, it's probably necessary. If this seems like something everyone thinks is a good idea, I'll see what I can do to start implementing this. However, I still think certain games would have to be completely custom (1942).

NOP:
you could also hybrid it.

Games that are simple, or identical (like all the pacman clones) could all have .hid files, others that are wonky have custom code for deciphering.  As long as it's all open source in the end, I really don't mind either way it goes, since we're not losing any data.  I'd think that anyone able to read/write a .hid file is likely also to have enough skill to read/write actual code.

Fyrecrypts:
I tend to think the same way NOP, but code can be intimidating to learn for those that don't I suppose, whereas a .hid file would be a csv file basically.

I dunno, I think it could be worth it, many of the old 80's Nintendo games or games based off Nintendo games(Donkey Kong, Donkey Kong Jr, Crazy Kong, Mario) are all very similiar, and probably could be done easily in a .hid file.

It'd be a lot of work converting the 23 files we have to use this system, but for the 1000's of games left to decode, it could ultimately save a lot of time.

Navigation

[0] Message Index

[#] Next page

[*] Previous page