ok I've got 58-65 fps. Game works good now.... waiting for SPU2-X plugin...
I found some bugs.
Character direction bug in battle, broken polygon, Keyboard upside input bug in some area.
Tested on PCSX2 R628
CRC=6A4EFE60
[Plugins]
GSdx 0.1.14 R594 [PS3.0/NLoop on/Texture filtering on/Logarithmic Z on]
SPU2-X 1.1 [default options]
Lilypad 0.9.10 R630 [default options]
[Options]
EErec on/VU0 on/VU1 on/MTGS on
FPU Compare Hack off/VU add Sub hack off
EE sync hack default/INTC sync hack off/IOP x2 Cycle off/Waitcycles sync hack off
Advanced options default
Windows XP SP3
# Game Version: FFX German (SCES 50492, CRC 941BB7D9)
# PCSX2 0.9.6 (crash)
# PCSX2 SVNs: r612 (crash), r632 (black screen, no repsonse) - (r590 and r603 worked fine)
# Plugins (as of 0.9.6):
gsdx, zerogs
lilypad
spu2-x
# CPU: default (MTGS EERec VU0 VU1 on, limit @56)
# Advanced Settings: EE round mode negative, everything else default
# Speed hacks off
# Description:
In some Scenes the graphics get corrupted (starts as minor polygon glitches, within a few seconds whole screens gets garbled)) followed by emulator crashing (r612) or going for a black screen/no response (r632). Using r603 there's no corruption visible and emulator continues.
[
attachment=13638] [
attachment=13636] [
attachment=13639]
(r632, gsdx10 software)
I encountered this behavior in three different scenes so far (Talking to the Summoners in the Al Bhed Home, Seymor's first death in Macalania and the talk with Yunalesca in Zanarkand), and it happens at the exact same spot on every try. The only thing I can see that those scenes have in common is the heavy use of crossfades.
I somewhat excluded GS plugins as culprits (tried it with both gsdx and zerogs using various settings and versions - and the fact that it works in r603).
--------------
Also, I second the above poster's finds, although the directionality bug has been around for a while (e.g. 0.9.2).
[
attachment=13637]
(r603, gsdx10 hardware)
---------------
-----EDIT----
added screenshot of the error msg given by a r634 debug build. I don't know if that helps.
[
attachment=13650]
-----EDIT 2----
*tested with 0.9.6, still crashes
The reversed directions in battle thing can be fixed by setting EE round mode (actually COP2 round mode which is sadly the same setting as EE round mode) to negative or positive (try both, some may fix it partially, some fully). This is in advanced settings, by the way.
Setting EE Rec Round Mode to Negative fixes directionality problems.
thx nneeve!
I also updated my bug report with 0.9.6 info, crashes still occur.
(02-18-2009, 03:43 PM)lightchris Wrote: [ -> ] (02-18-2009, 12:40 PM)Bositman Wrote: [ -> ]I'm not getting that here. Rev 502 is the same as Rev 520,521 and 522.The blackness there is caused by GSdx's hardware renderer,if you use SW it looks ok (check pics)
Yes and no. 
The black/purpleness is also not present when VU1rec is disabled (also, Kimahris' hair is still bugged in GSdx software mode with VU1rec on). So I wonder whether the problem really lies in GSdx's hardware mode or in VU1rec.
(02-18-2009, 12:40 PM)Bositman Wrote: [ -> ]But this could also be because my CPU supports both SSSE3 and SSE4 so if yours doesn't,it will be using some code mine doesn't.
Edit: Hmm I see yours doesn't support SSSE3 or SSE4,will check it with someone who doesn't
As any Core 2 Duo it supports SSSE3, which I am using.
(02-18-2009, 02:40 PM)refraction Wrote: [ -> ]make sure you have the "advanced" options to defaults, no extra options on, also "Logarithmic Z" is selected in gsdx9
I did that. Maybe some error in my specific compile, I used the "release" option.
edit: ah, I see ramapcsx2 already reverted the changes of 521 which really caused bugs without SSE4.
edit2: Yep, the glitch is gone in 524.
Is this bug supposed to be fixed in 0.9.6? I get it no matter what settings I choose. Except maybe if I disable VU1rec but if I do that I get 5 fps and random crashes, so I'm not sure I'll ever even get far enough through the intro to check that.
Edit: Oops, was wrong gsdx version.
Edit 2: Is there any way to run it without that glitch in hardware mode? Older gsdx version, sse4 cpu or directx10 version of gsdx?
(03-02-2009, 03:38 AM)ZellSF Wrote: [ -> ]Is this bug supposed to be fixed in 0.9.6? I get it no matter what settings I choose. Except maybe if I disable VU1rec but if I do that I get 5 fps and random crashes, so I'm not sure I'll ever even get far enough through the intro to check that.
Edit: Oops, was wrong gsdx version.
Edit 2: Is there any way to run it without that glitch in hardware mode? Older gsdx version, sse4 cpu or directx10 version of gsdx?
If you're talking about the darkness (Tidus' necklace, hair): I know it worked in some very old version of PCSX2 and/or ZeroGS but that might cause other problems.
For now I'd say no, impossible.
The reversed controls in the airship room (and probably others) can be fixed by enabling "full" FPU clamping mode in the latest revision. Please say if it breaks something else.
[Bug Report] Final Fantasy X [PAL]
Version - 0.9.6
Graphics: GSdx 883 (SSE4.1)0.1.13 (And ZeroGS 0.97.1)
Cdvdrom - Gigaherz
Bios: Europe V1.6
Bug: after the game in Luca (story line game, not optional) there is a fight in the sphere pool and two outside, after which seymour begins to summon Anima. As the hook comes down to grab anima, Pcsx2 crashes and closes.
Newest Beta (720) does not solve the problem.
Attatched:
save state: wait around 12 seconds
Memcard, easiest save is to use save 3, (cheer from tidus once will allow first section of enemies to be killed in one hit) auran will kill 2nd battle in one hit. 3 enemy, hit with wakas dark ball and everyone elses attack. Once this enemy goes down it will begin a cutscene. After seymour does the 'prayer' screen pans up and hook begins to descend and emulator crashes.
latest version of gsdx is 0.1.14 i think.
my bios is different and FFX is NTSC - was no problem with that scene.
also, as a temporary solution, try 0.9.5svn to pass this scene and than continue playing on 0.9.6