Author Topic: [PC] HRC Resizer  (Read 19521 times)

New Pagodi

  • *
  • Posts: 11
    • View Profile
    • tumblr
Re: [FF7] HRC Resizer
« Reply #25 on: 2016-04-22 17:46:12 »
So I tested few of the edits for the settings I need. The Vertices setting does only need a fix Vertices button, because mode 2 is probably the flag the game use to hide a model. So the program should set it only to mode 1. The texture flag is harder to get rid of as I expected, because it needs to make the NumTexCoords to 0 as also it needs to remove the array were the text coords are stored. Here are two test files I made:https://mega.nz/#!udlDSQBb!OjM026qUn6gOiII0pzrNCpQc6Wx9pND_wRr5PLVyQfY
The array starts at 000027C0x8.
Is it actually possible to cut out code from a file?

I tried writing a viewer to see if I could at least read the data in those files.  Since I don't know what the numbers are supposed to be I don't know if I'm reading them right.  Can you try it and see? 

I'm pretty sure that I'm at least reading the floats wrong, because in the bounding box section of one of the files you uploaded, the utility the utility reads the maxz as less than the minz.  That obviously can't be right.

One problem I already see is in the "Hundreds" section of the file.  The link you gave says it's a "Complex structure that specifies render state (texturing, blend modes, culling, filtering, depth testing, depth masking, shademode etc)" 

Do you have the details of that structure?  It seems to be of variable size.  In one of the files you uploaded, it appears to be 108 bytes and in the other it's 104 bytes.  In the viewer, I simply output the bytes in hex and decimal since I don't know what the bytes are supposed to mean.

edit: updating the link.
« Last Edit: 2016-04-22 21:06:17 by New Pagodi »

New Pagodi

  • *
  • Posts: 11
    • View Profile
    • tumblr
Re: [FF7] HRC Resizer
« Reply #26 on: 2016-04-22 21:17:02 »
After looking at a few more pfiles, I guess the hundreds structure is not variable size but always 100 bytes (hey, it's right there in the name). 

But there seems to be an extra 4 bytes between the groups items and the final 2 items (Bounding boxes and the Normal index table). 

At least with these 2 assumptions, the numbers the viewer spits out seem to make more sense.

Also, there seems to be an extra 4 bytes (all zeros) at the end the Swordtest.p file that was uploaded.  That's why I originally thought the hundred table was variable size.  Are those extra bytes supposed to be there?