675010e1b5911

675010e1b713a
1 Guest is here.
 

Topic: SS1 EE Reactor code randomization fix
Page: « 1 [2]
Read 15350 times  

675010e1b756evoodoo47

Acknowledged by: dertseha
675010e1b776c
You probably have already checked the sources, but if you have not code is randomized in PLAYER.C.

675010e1b784eJosiahJack

675010e1b78a3
What exactly happened to break the randomization in EE?

675010e1b7c80dertseha

675010e1b7cd5
What exactly happened to break the randomization in EE?
Unknown.
Wild theories follow:
Assuming that the Enhanced Edition was created without the sources, so, based on reverse engineering / patching the executable, it may be that some initialization order and/or pointers were messed up. The randomization is based on the effects-randomizer. So, it could be that this randomizer is affected.
Or - Remember when the OpenSSH library was broken by satisfying a static code checker to initialize a variable? Something similar could have happened here: Instead of reading to-be-believed random memory as a seed for the randomizer, this memory locaion is now pre-set, with always the same value.

675010e1b7de0TurboSnail

675010e1b7e2b
Ok,

Well the Data Reader is behaving better but the Automap on level two is refusing to work now. It will not draw the map or move the arrow.
I am using "archive.dat" [CRC-32: 214f9e1f]
I just checked the older "archive.dat" and found it also had an Automap issues. The arrow was duplicated (One purple stationary and one red which moved), the arrows would not rotate and the map did not auto scroll but it did draw.
Very strange. Thank you for your effort!

675010e1b7efddertseha

675010e1b7f49
Thank you for testing!

This whole thing is getting weirder and weirder. I might have time on the weekend to investigate.

[edit] First test: This seems to affect only level 2 (tested R up to 6), and only the left MFD. The right one is fine.
« Last Edit: 13. April 2018, 05:29:39 by dertseha »
675010e1b802a
There must be something very wrong with the executable. That sounds very strange.

675010e1b80e9dertseha

675010e1b8136
I doubt that the cause is with the executable as I mainly work with the original CD release, where I reproduced it.
I do have a vague suspicion based on my excursions through the original code, though it is very remote and I'd want to first do some analysis before making it real by wording it ;)
675010e1b8218
Oh, OK. I thought the problem was with enhanced edition only.

I'm very intrigued!
1 Guest is here.
This thing was so bright that it almost appeared to be neon
Contact SMF 2.0.19 | SMF © 2016, Simple Machines | Terms and Policies
FEEP
675010e1bb3c9