home *** CD-ROM | disk | FTP | other *** search
- a
- Area 0: Users Netmail Area
- From: Grant Erickson Pvt Rec'd
- Orig: Unlisted system (0:153/911)
- To: Chris Padgett Msg #3, May-16-95 14:50:00
- Subject: CIV speed-up
-
- I got this from another board on the Internet
-
- --------- start of message ----------------------------------------------
-
- Here's an earlier post of mine about disabling pointer changing in
- CIV-ECS. It also works for the AGA version, although you need to change
- the 2nd and 3rd instances. In the ECS it's only one instance of 2 you'll
- find. (Can't remember which.) Also, the AGA version is powerpacked with
- version 4.0, which is commercial. Use "XFD" on Aminet to unpack it.
-
- ---
- Use a disk/file editor to search for the following hex sequence:
-
- 4EAEFEF2 (in asm, JSR -$10E(A6) )
-
- You will find two (2) of these. Change one of the sequences to
-
- More [Y,n,=]? =
-
- 4E714E71 (in asm, NOP NOP )
-
- Run the program. If the graphics flicker wildly, then you changed the
- wrong sequence. (WaitVBlank, which is needed!) Restore the executable, and
- try the other sequence!
- ---
-
- Even though you boot in ECS, the ROM code is still written for AGA
- sprite pointers. There are alot more custom registers to playwith, and the
- copper list is 5-10x as big as a ECS one on KS 1.2/2.0. Civilization
- likes to change pointers ALOT, sometimes as many as 5-10 times a VBlank!
- This is ofcourse unneccessary, as the pointer does not interfere with the
- underlying GFX, unlike IBM cards! WHY IN THE HELL DIDN'T MICROPROSE
- REMOVE THESE POINTER CALLS IN THE AMIGA VERSION??? The world may never know.
-
- -------- end of message --------------------------------------------------
-
- Actually I found 4 instances of '4EAEFEF2' in the patch version I believe
- and I changed the 2nd and the 3rd, the first will make it flicker wildly,
- the fourth I did not see what it had to do with anything, so I changed it
- back. It could be that this fix was originally made for the original not the
- patch version, but I used it on the patch version ECS.
- There is supposed to be a way of getting rid of the screen fades also,
- but I was only able to get a hold of the AGA version way, I tried running
- CIV on a source level debugger, but was unable to find the screen fades. If
- I eventually find or hear how, I'll let you know.
- If you do not have a file editor I have a share-ware one I could encode
- and send out. I think it LHA's down to 185k encodes to 265 or so, but I could
- break it up into smaller chunks if necessary.
-
-
- Next Msg [Y)es, N)o, A)gain, R)eply, *)NextArea]: