https://bugs.winehq.org/show_bug.cgi?id=38274
Bug ID: 38274 Summary: White water in Heroes VI Product: Wine Version: 1.7.39 Hardware: x86 OS: Mac OS X Status: UNCONFIRMED Severity: normal Priority: P2 Component: directx-d3d Assignee: wine-bugs@winehq.org Reporter: isakov-sl@bk.ru
Created attachment 51102 --> https://bugs.winehq.org/attachment.cgi?id=51102 Water is while places
Might and Magic Heroes VI - Shades of Darkness
Water on map shown as white place, not transparent and no details like on screen 1. I was considering this as a missing feature but I found some engine version shows water like on screen 2. So this is possible!
I will do regression test to find the best revision and corresponding changes.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #1 from Sergey Isakov isakov-sl@bk.ru --- Created attachment 51103 --> https://bugs.winehq.org/attachment.cgi?id=51103 Water is partially present
It was a version 1.7.2 with some additional commits after it. I will find the best commit later.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #2 from Sergey Isakov isakov-sl@bk.ru --- I made regression test. Just to clarify "good" means water is white place. "bad" means water is transparent as it should be. So good and bad in this test reverted because 1.7.2 and early has white water. As well as 1.7.3+ More or less visible water was in several revision between 1.7.2 and 1.7.3. So I found first commit that developed water ---- ac37f7a2ac09c77f7bd9db39c3180307696fdc6c is the first bad commit commit ac37f7a2ac09c77f7bd9db39c3180307696fdc6c Author: Stefan Dösinger stefan@codeweavers.com Date: Wed Sep 18 12:24:56 2013 +0200
wined3d: Implement locking of block-based volume textures.
:040000 040000 4260ffc1ea88ab11c4e4580b621992c6d3e59204 0f87593a636e70dc51923d9030b4884179bc8150 M dlls
---- Should I made regression test to know when this feature disappears again?
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #3 from Sergey Isakov isakov-sl@bk.ru --- And now regression test to know when the water disappeared again --- 2680f3301585e2263e3800454f51c8f09314926f is the first bad commit commit 2680f3301585e2263e3800454f51c8f09314926f Author: Stefan Dösinger stefan@codeweavers.com Date: Wed Sep 18 12:24:57 2013 +0200
wined3d: Check box dimensions in volume_map.
:040000 040000 0f87593a636e70dc51923d9030b4884179bc8150 af9aef50708ee5a22625124e26ccb7ad1c1733d9 M dlls ---
https://bugs.winehq.org/show_bug.cgi?id=38274
Stefan Dösinger stefan@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |stefan@codeweavers.com
--- Comment #4 from Stefan Dösinger stefan@codeweavers.com --- Yes, it is possible that this game needs compressed volumes. What are its minimum requirements? Is it supposed to work on a Geforce 7 card?
Does the game use d3dx9_xy.dll? Our builtin DLL doesn't handle format fallbacks well, so if you're running this game with builtin try the native library instead.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #5 from Sergey Isakov isakov-sl@bk.ru --- (In reply to Stefan Dösinger from comment #4)
Yes, it is possible that this game needs compressed volumes. What are its minimum requirements? Is it supposed to work on a Geforce 7 card?
No this game installed on computer with Radeon card
Does the game use d3dx9_xy.dll? Our builtin DLL doesn't handle format fallbacks well, so if you're running this game with builtin try the native library instead.
When I first install the game it works without textures. Paladin looks like a silhouette filled by light clouds. Have no screenshot to show. Then I installed Directx9Redistributable and see very good textures. Game is almost playable except three bugs: 1. Crash at begin - resolved by DXTn volume disable. 2. White water - needed for DXTn volumes enable? 3. Whole luminance can suddenly switch off (Bug 37705).
I have logs with different tracings d3d, d3d_surface, d3d_texture, d3dx. I also implemented DXTn support without libtsc_dxtn but still no volume support. Strange but the testing utility DxTex.exe from DirectxSDK is not working on this computer with any wine version, may be there is my mistake.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #6 from Sergey Isakov isakov-sl@bk.ru --- I propose to make uncompress volume_textures before send GL_EXTCALL(glTexImage3D(GL_TEXTURE_3D, volume->texture_level, srgb ? format->glGammaInternal : format->glInternal, volume->resource.width, volume->resource.height, volume->resource.depth, 0, format->glFormat, format->glType, mem)); checkGLcall("glTexImage3D");
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #7 from Stefan Dösinger stefan@codeweavers.com --- (In reply to Sergey Isakov from comment #5)
Yes, it is possible that this game needs compressed volumes. What are its minimum requirements? Is it supposed to work on a Geforce 7 card?
No this game installed on computer with Radeon card
I meant if the game is supposed to work when you run it with a Geforce 7 GPU. According to http://www.game-debate.com/games/index.php?g_id=742&game=Heroes%20of%20M... it needs at least a geforce 8, which is d3d10 capable. This card may have DXTn volume support.
Strange but the testing utility DxTex.exe from DirectxSDK is not working on this computer with any wine version, may be there is my mistake.
It's been a while since I looked at DxTex.exe, but if my memory serves me right it tries to use ddraw blits to convert compress and decompress DXTn textures. We don't implement this because the compression and decompression algorithms are covered by U.S. patent 5956431. All we can do is pass the data to the OpenGL implementation and let the GPU vendor deal with the patent.
There may also be other problems that prevent DxTex from working. A long time ago it had problems with child window d3d rendering, and the ddraw version of this utility had problems with our DDSCAPS_* handling when loading / saving compressed textures. Child window d3d should work these days, and I don't know if the version from the last dx sdk still uses ddraw.dll for handling .dds files.
(In reply to Sergey Isakov from comment #6)
I propose to make uncompress volume_textures before send GL_EXTCALL(glTexImage3D(GL_TEXTURE_3D, volume->texture_level, srgb ? format->glGammaInternal : format->glInternal, volume->resource.width, volume->resource.height, volume-> resource.depth, 0, format->glFormat, format->glType, mem)); checkGLcall("glTexImage3D");
Patches are welcome. Properly integrating conversion like this is a challenge of course. You can also try to re-order the 2D blocks into 3D blocks that make GL_NV_texture_compression_vtc happy. The format conversion infrastructure in utils.c is a good place to start, but you'll have to add the ability to limit conversion to volumes.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #8 from Sergey Isakov isakov-sl@bk.ru --- Convert volume front slice to surface and let OpenGL render it...
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #9 from Stefan Dösinger stefan@codeweavers.com --- I haven't checked the spec, but I don't think this is allowed. OpenGL only allows block-aligned updates, except in cases where the resource is smaller than the block size (e.g. mipmaps). Even then you have to supply a full block of data of course.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #10 from Sergey Isakov isakov-sl@bk.ru --- (In reply to Stefan Dösinger from comment #7)
textures. We don't implement this because the compression and decompression algorithms are covered by U.S. patent 5956431. All we can do is pass the data to the OpenGL implementation and let the GPU vendor deal with the patent.
Does it mean that we are not allowed even to look these textures? Or decompress to look? For my understanding we can't compress our images by this algo. But if such data is in input then we can somehow to get data. And if there is a difference using libtsc_dxtn or its sources as they are opened just keeping original copyright and license agreement? --- /* * libtxc_dxtn * Version: 1.0 * * Copyright (C) 2004 Roland Scheidegger All Rights Reserved. * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the "Software"), * to deal in the Software without restriction, including without limitation * the rights to use, copy, modify, merge, publish, distribute, sublicense, * and/or sell copies of the Software, and to permit persons to whom the * Software is furnished to do so, subject to the following conditions: * * The above copyright notice and this permission notice shall be included * in all copies or substantial portions of the Software. * * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL * BRIAN PAUL BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ ---
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #11 from Stefan Dösinger stefan@codeweavers.com --- We're avoiding decoding them as well. The copyright notice you posted is about the license of libtxc_dxtn. Having this library doesn't grant you a right to use the patent.
The most realistic option is to use a library called Nvidia texture tools. Nvidia has an agreement with the owner of the patent (no idea who that is these days) that allow others to use this library without licensing the patent: https://code.google.com/p/nvidia-texture-tools/wiki/FAQ . Note that despite the name this library doesn't depend on Nvidia GPUs.
But still, we maybe don't need to decompress the dxtn blocks for volumes, "only" shuffle them around in memory.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #12 from Sergey Isakov isakov-sl@bk.ru --- What is the difference how can we view the image, by block manipulating or by decompressing? The final result is the same "we can view images created with DXTn". It is a problem of those who uses the compression to encode images but it is not a problem for those who bought a game to play. I think it's a time to call lawyer to resolve a question.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #13 from Sergey Isakov isakov-sl@bk.ru --- OpenSource Mozilla can view GIF patented images?
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #14 from Dmitry Timoshkov dmitry@baikal.ru --- (In reply to Sergey Isakov from comment #13)
OpenSource Mozilla can view GIF patented images?
GIF patent has expired long time ago.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #15 from Sergey Isakov isakov-sl@bk.ru --- (In reply to Stefan Dösinger from comment #11)
The most realistic option is to use a library called Nvidia texture tools. Nvidia has an agreement with the owner of the patent (no idea who that is these days) that allow others to use this library without licensing the patent: https://code.google.com/p/nvidia-texture-tools/wiki/FAQ . Note that despite the name this library doesn't depend on Nvidia GPUs.
Probably good idea. They said the library licensing compatible with GPL. It also gives us DirectX10 and 11 capabilities. But it requires a time to implement into wine. I just didn't understand what is the condition to use this library? Place Nvidia copyright?
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #16 from Sergey Isakov isakov-sl@bk.ru --- (In reply to Dmitry Timoshkov from comment #14)
(In reply to Sergey Isakov from comment #13)
OpenSource Mozilla can view GIF patented images?
GIF patent has expired long time ago.
Mozilla exists long time ago.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #17 from Stefan Dösinger stefan@codeweavers.com --- We "view" the image when running a game by passing it to OpenGL and let the driver developers (Nvidia, AMD) bother about the legal stuff. In case of Mesa OpenGL passes that to the user in the sense of "Install this library. It's your responsibility now to care for the patent, not ours."
In theory we could also use OpenGL to compress and decompress the textures without rendering it, but the API architecture, and in particular the requirement to create an OpenGL context to do that, makes this difficult.
Still, I don't think our ability to create or decompress s3tc textures breaks your game. It most likely breaks dxtex though.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #18 from Sergey Isakov isakov-sl@bk.ru --- dxtex crashes on open new windows irrespectively of textures formats
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #19 from Sergey Isakov isakov-sl@bk.ru --- Nvidia Texture Tools has a license on the patent. But implementing their files into other project will lose the license. Only products created by Nvidia itself can use this patent. Again, I will insist that decompressing is free. Or wine project have to pay for the license.
https://bugs.winehq.org/show_bug.cgi?id=38274
Matteo Bruni matteo.mystral@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |matteo.mystral@gmail.com
--- Comment #20 from Matteo Bruni matteo.mystral@gmail.com --- (In reply to Sergey Isakov from comment #19)
Nvidia Texture Tools has a license on the patent. But implementing their files into other project will lose the license. Only products created by Nvidia itself can use this patent.
That's not how I read https://code.google.com/p/nvidia-texture-tools/wiki/FAQ (specifically last point and second comment below).
FWIW e.g. many games made with Unreal Engine 3 include nvtt.dll.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #21 from Sergey Isakov isakov-sl@bk.ru --- (In reply to Stefan Dösinger from comment #17)
Still, I don't think our ability to create or decompress s3tc textures breaks your game. It most likely breaks dxtex though.
Those bug 38178 are closed so I reported here new found. no traces, no warn - crash after logo. trace+d3d, d3d_texture, d3d_surface - same warn+d3d - no crash .............
Can anybody explain me what is happen when I switch WARN_ON?
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #22 from Sergey Isakov isakov-sl@bk.ru --- log instead crash warn:d3d:state_debug_monitor token: 0xbaadcafe. warn:d3d:state_cb_warn Constant buffers (STATE_CONSTANT_BUFFER(WINED3D_SHADER_TYPE_PIXEL)) no supported. warn:d3d:state_cb_warn Constant buffers (STATE_CONSTANT_BUFFER(WINED3D_SHADER_TYPE_VERTEX)) no supported. warn:d3d:state_cb_warn Constant buffers (STATE_CONSTANT_BUFFER(WINED3D_SHADER_TYPE_GEOMETRY)) no supported. warn:d3d:context_bind_shader_resources No resource view bound at index 0, 0. warn:d3d:context_bind_shader_resources No resource view bound at index 0, 0.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #23 from Sergey Isakov isakov-sl@bk.ru --- (In reply to Matteo Bruni from comment #20)
(In reply to Sergey Isakov from comment #19)
Nvidia Texture Tools has a license on the patent. But implementing their files into other project will lose the license. Only products created by Nvidia itself can use this patent.
That's not how I read https://code.google.com/p/nvidia-texture-tools/wiki/FAQ (specifically last point and second comment below).
FWIW e.g. many games made with Unreal Engine 3 include nvtt.dll.
nvtt.dll made by nvidia so it can use DXTn
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #24 from Matteo Bruni matteo.mystral@gmail.com --- (In reply to Sergey Isakov from comment #23)
(In reply to Matteo Bruni from comment #20)
(In reply to Sergey Isakov from comment #19)
Nvidia Texture Tools has a license on the patent. But implementing their files into other project will lose the license. Only products created by Nvidia itself can use this patent.
That's not how I read https://code.google.com/p/nvidia-texture-tools/wiki/FAQ (specifically last point and second comment below).
FWIW e.g. many games made with Unreal Engine 3 include nvtt.dll.
nvtt.dll made by nvidia so it can use DXTn
Yeah, so we can use it too just fine.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #25 from Sergey Isakov isakov-sl@bk.ru --- Fine! We just need interfaces.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #26 from Sergey Isakov isakov-sl@bk.ru --- (In reply to Sergey Isakov from comment #5)
Strange but the testing utility DxTex.exe from DirectxSDK is not working on this computer with any wine version, may be there is my mistake.
Resolved. There is a trick "Use Mac driver instead of X11" set true. Now the utility works!
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #27 from Sergey Isakov isakov-sl@bk.ru --- See license ------ NVIDIA Texture Tools 2.0 is licensed under the MIT license.
Copyright (c) 2007 NVIDIA Corporation
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #28 from Sergey Isakov isakov-sl@bk.ru --- Hey, I have some progress in implementing DXTn --- $ wine Z:./dlls/d3dx9_36/tests/d3dx9_36_test.exe.so volume err:process:__wine_kernel_init boot event wait timed out fixme:d3d_surface:volume_init format 0x74 size=1048576 fixme:d3dx:D3DXLoadVolumeFromMemory convert volume tex from 0x15 to 0x15 fixme:d3dx:D3DXLoadVolumeFromMemory Unhandled filter 0x80004. fixme:d3dx:D3DXLoadVolumeFromMemory format converted from 0x15 to 0x15 fixme:d3dx:D3DXLoadVolumeFromMemory convert volume tex from 0x15 to 0x15 fixme:d3dx:D3DXLoadVolumeFromMemory format converted from 0x15 to 0x15 volume.c:127: Test failed: Got color 0xffffffff, expected 0x00000000 volume.c:127: Test failed: Got color 0xffffffff, expected 0x00000000 volume.c:127: Test failed: Got color 0xffffffff, expected 0x00000000 volume.c:127: Test failed: Got color 0xffffffff, expected 0x00000000 volume.c:127: Test failed: Got color 0xffffffff, expected 0x00000000 volume.c:127: Test failed: Got color 0xffffffff, expected 0x00000000 volume.c:127: Test failed: Got color 0xffffffff, expected 0x00000000 fixme:d3d_surface:volume_init format 0x74 size=1048576 fixme:d3d_texture:volumetexture_init create dxtn texture 0x31545844. fixme:d3d_surface:volume_init format 0x31545844 size=32 fixme:d3dx:D3DXLoadVolumeFromMemory convert volume tex from 0x15 to 0x31545844 volume.c:208: Test marked todo: D3DXLoadVolumeFromMemory returned 0x8876086c, expected 0 fixme:d3d_texture:volumetexture_init create dxtn texture 0x33545844. fixme:d3d_surface:volume_init format 0x33545844 size=32 volume: 262209 tests executed (1 marked as todo, 7 failures), 0 skipped.
Wineskin Command Line Wine Test
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #29 from Bruno Jesus 00cpxxx@gmail.com --- I didn't read the whole thread but this this dxtn discussion reminds me of the dxtn patches in wine-staging, do they help in any way for your problem? https://github.com/wine-compholio/wine-staging/tree/master/patches
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #30 from Sergey Isakov isakov-sl@bk.ru --- Sure, I based on that patches but also adding volume DXTn preliminary support.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #31 from Sergey Isakov isakov-sl@bk.ru --- I got new knowledge. 1. Install the game. All textures are wrong. 2. Install DirectX9 redistributable. All is good except three bugs. 3. Use Winetricks to set alldlls=builtin. Textures are wrong. 4. With winecfg set d3dx9_36.dll to native. Textures are good. That's all.
It means: 1. Builtin d3dx9_36 is far from good. But we can use native one. Why not??? 2. The bug with white water and with the game crashing lies out of the library. Probably in wined3d.dll.
Next experiment. MacOSX 10.9.5 with native drivers. AMD Radeon 6670. Use DxTex.exe from DirectX9SDK. The utility can manipulate with textures different forms and formats. I created Volume Textures DXT5 with MipMapLevel=9 and save it. SIC! There is DXTn support! This test was done with unmodified Wine-1.7.41 with native d3dx9_36. The utility is able to create and save volume DXT5 textures with mipmaps. It is attached. Where is my mistake?
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #32 from Sergey Isakov isakov-sl@bk.ru --- Created attachment 51530 --> https://bugs.winehq.org/attachment.cgi?id=51530 DDS volume texture compessed as DXT5 created by unmodified wine
trace:loaddll:load_native_dll Loaded L"C:\windows\system32\mfc100.dll" at 0x78b60000: native trace:loaddll:load_builtin_dll Loaded L"C:\windows\system32\shell32.dll" at 0x42a10000: builtin trace:loaddll:load_builtin_dll Loaded L"C:\windows\system32\winspool.drv" at 0x41fa0000: builtin trace:loaddll:load_builtin_dll Loaded L"C:\windows\system32\comdlg32.dll" at 0x42910000: builtin trace:loaddll:load_builtin_dll Loaded L"C:\windows\system32\opengl32.dll" at 0x7a810000: builtin trace:loaddll:load_builtin_dll Loaded L"C:\windows\system32\wined3d.dll" at 0x42cb0000: builtin trace:loaddll:load_builtin_dll Loaded L"C:\windows\system32\d3d9.dll" at 0x42c60000: builtin trace:loaddll:load_builtin_dll Loaded L"C:\windows\system32\d3dx9_43.dll" at 0x418d0000: builtin trace:loaddll:load_builtin_dll Loaded L"C:\windows\system32\msvcrt.dll" at 0x42df0000: builtin trace:loaddll:load_native_dll Loaded L"C:\windows\system32\d3dx9_36.dll" at 0x400000: native trace:loaddll:load_builtin_dll Loaded L"C:\windows\system32\imm32.dll" at 0x41fe0000: builtin trace:loaddll:load_builtin_dll Loaded L"C:\windows\system32\winemac.drv" at 0x42f10000: builtin
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #33 from Sergey Isakov isakov-sl@bk.ru --- d3dx9_36_test succeeded too --- $ wine c:/tests/d3dx9_36_test.exe.so volume fixme:winediag:start_process Wine Staging is a testing version containing experimental patches. fixme:winediag:start_process Please report bugs at http://bugs.wine-staging.com (instead of winehq.org). fixme:ntdll:NtCreateNamedPipeFile Message mode not supported, falling back to byte mode. fixme:ntdll:NtCreateNamedPipeFile Message mode not supported, falling back to byte mode. fixme:ntdll:NtCreateNamedPipeFile Message mode not supported, falling back to byte mode. fixme:ntdll:NtCreateNamedPipeFile Message mode not supported, falling back to byte mode. fixme:ole:RemUnknown_QueryInterface No interface for iid {00000019-0000-0000-c000-000000000046} fixme:d3d:quirk_apple_glsl_constants There are 0 GLSL constants reserved fixme:d3d_surface:volume_init format 0x74 size=1048576 fixme:d3d_surface:volume_init format 0x74 size=1048576 fixme:d3d_surface:volume_init format 0x31545844 size=32 volume.c:208: Test succeeded inside todo block: D3DXLoadVolumeFromMemory returned 0, expected 0 fixme:d3d_surface:volume_init format 0x33545844 size=32 volume: 262209 tests executed (0 marked as todo, 1 failure), 0 skipped.
Wineskin Command Line Wine Test
https://bugs.winehq.org/show_bug.cgi?id=38274
Béla Gyebrószki gyebro69@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Keywords| |download Status|UNCONFIRMED |NEW URL| |http://www.gamershell.com/d | |ownload_78182.shtml CC| |gyebro69@gmail.com Ever confirmed|0 |1
--- Comment #34 from Béla Gyebrószki gyebro69@gmail.com --- Water is replaced with glowing white textures here as well, tested on Nvidia card with nouveau and binary drivers 340.76. Reducing detail level (including shader quality), native d3dx9* or disabling GLSL doesn't make it better.
Based on the info in comment #2, I did a git checkout ac37f7a2ac09c77f7bd9db39c3180307696fdc6c but all I saw was the same white water with some additional glitches.
The problem can be reproduced in the demo version as well (has a mere 4.3 GB download size). When you start the tutorial campaign in the demo, you should see the white water issue during the intro cinematic (videos are rendered in real-time in the game). If you start the only available skirmish map in the demo, water is located south of the player's starting position.
In the terminal I see only fixme:win:EnumDisplayDevicesW ((null),0,0x35dc748,0x00000000), stub! fixme:win:EnumDisplayDevicesW ((null),0,0x35dc418,0x00000000), stub! fixme:ddraw:ddraw7_Initialize Ignoring guid {aeb2cdd4-6e41-43ea-941c-8361cc760781} fixme:d3d9:Direct3DShaderValidatorCreate9 stub
wine-1.7.49-41-g36a39ce Fedora 22 nouveau/mesa git on NV92
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #35 from Sergey Isakov isakov-sl@bk.ru --- Thanks for confirming the bug.
I can say that the bug presents with all HeroesVI versions and with all Wine versions including wine-stage. Replacing libraries d3dx*** to native doesn't help. The bug present. The bug is not related to DXTn support. I made it and the bug is still here. The bug seems to be in wined3d.dll interpreting volume textures. I saw that the game assumed XZ slice (up view) while wined3d uses XY slice (front view). May be this is the root.
The bug is absent in real Windows 7.
https://bugs.winehq.org/show_bug.cgi?id=38274
Stefan Dösinger stefandoesinger@gmx.at changed:
What |Removed |Added ---------------------------------------------------------------------------- CC|stefan@codeweavers.com |stefandoesinger@gmx.at
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #36 from Sergey Isakov isakov-sl@bk.ru --- Still the issue with 1.9.10
https://bugs.winehq.org/show_bug.cgi?id=38274
fjfrackiewicz@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |fjfrackiewicz@gmail.com
--- Comment #37 from fjfrackiewicz@gmail.com --- (In reply to Sergey Isakov from comment #36)
Still the issue with 1.9.10
Is this still an issue with Wine 1.9.13?
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #38 from Sergey Isakov isakov-sl@bk.ru --- Created attachment 54986 --> https://bugs.winehq.org/attachment.cgi?id=54986 white places instead of water
Tested wine-1.9.13-28-g8bfcd7c.
What is the specific reason to ask again? Is there any movement in implementing Volume textures? DXTn textures?
All three weird bugs in Heroes VI are still present: 1. White water 2. Flying squares 3. Fall into darkness - play impossible after that.
https://bugs.winehq.org/show_bug.cgi?id=38274
Józef Kucia joseph.kucia@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- See Also| |https://bugs.winehq.org/sho | |w_bug.cgi?id=39253 CC| |joseph.kucia@gmail.com Summary|White water in Heroes VI |White water in Heroes VI | |(DXTn volume textures)
--- Comment #39 from Józef Kucia joseph.kucia@gmail.com --- The hack from bug 39253 should bring back water rendering. Visual glitches are expected though.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #40 from Sergey Isakov isakov-sl@bk.ru --- Sorry, no. It will be too simple solution. Tested wine-2.0-rc4-80-gacd6243 + patch from 39253 patching file dlls/wined3d/texture.c patching file dlls/wined3d/utils.c
Same picture with white water as before.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #41 from Józef Kucia joseph.kucia@gmail.com --- Created attachment 56812 --> https://bugs.winehq.org/attachment.cgi?id=56812 Hack
(In reply to Sergey Isakov from comment #40)
Same picture with white water as before.
I guess that the hack from bug 39253 is not enough when GL_NV_texture_compression_vtc is not supported. This one might help.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #42 from Sergey Isakov isakov-sl@bk.ru --- It's a pity same result. Something new in log ----- fixme:d3d:texture3d_upload_data Not supported for block formats. fixme:d3d:texture3d_upload_data Not supported for block formats. fixme:d3d:texture3d_upload_data Not supported for block formats. fixme:d3d:texture3d_upload_data Not supported for block formats. fixme:d3d:texture3d_upload_data Not supported for block formats. fixme:d3d:texture3d_upload_data Not supported for block formats.
https://bugs.winehq.org/show_bug.cgi?id=38274
mo78@abv.bg changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |mo78@abv.bg
--- Comment #43 from mo78@abv.bg --- Same problem here with Wine 2.8. Water is white and the map is very dark hence it is impossible to play. In the tutorial campaign everything is allright. There are problems when water is present in the level, the first level from the Crag Hack campaign for example.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #44 from mo78@abv.bg --- Wine 2.21 - the problem is still here.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #45 from fjfrackiewicz@gmail.com --- (In reply to mo78 from comment #44)
Wine 2.21 - the problem is still here.
Have you tried Wine 3.0-rc6?
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #46 from mo78@abv.bg --- No, is there a patch in 3.0 RC6?
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #47 from mo78@abv.bg --- Wine 3.0 RC6 the problem is still there and it' worse. Now, in every level where we have water, the textures are very dark. The water still remains white. See the screenshot. If you run level without water, everything is working fine and all the graphics are correct.
https://s10.postimg.org/ksh8i80nd/Screenshot_20180118_131553.png
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #48 from mo78@abv.bg --- Created attachment 60275 --> https://bugs.winehq.org/attachment.cgi?id=60275 Very dark textures with white water. Wine 2.21 staging and Wine 3.0 RC6
Very dark textures with white water. Wine 2.21 staging and Wine 3.0 RC6. When you try a level without water, all the graphics seem to work properly.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #49 from mo78@abv.bg --- Very dark textures with white water. Wine 3.3 staging and Wine 3.3. When you try a level without water, all the graphics seem to work properly.
https://s10.postimg.org/mzmpzdq3t/Screenshot_20180304_190124.png
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #50 from mo78@abv.bg --- Wine 3.8 - the problem is still here. Ask for some logs I'll provide them. Just say what you need. Please let's get this bug fixed already :)
https://bugs.winehq.org/show_bug.cgi?id=38274
Stefan Dösinger stefan@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |stefan@codeweavers.com
--- Comment #51 from Stefan Dösinger stefan@codeweavers.com --- Mo78, this bug isn't easy to fix, logfiles won't help. The issue is that there's no equivalent OpenGL functionality for this d3d feature. Now that the S3TC patents expired we can probably write code to decompress the compressed volumes and use uncompressed textures. You are welcome to give it a try :-)
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #52 from mo78@abv.bg --- If I could, I would have done it :) Will VK9 help in this this case?
https://bugs.winehq.org/show_bug.cgi?id=38274
tokktokk fdsfgs@krutt.org changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |fdsfgs@krutt.org
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #53 from mo78@abv.bg --- Józef Kucia, aka The Wizard, is capable providing another miracle. I'm sure :)
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #54 from Béla Gyebrószki gyebro69@gmail.com --- Works for me in Wine 4.0 with Nvidia 415.22.05 Vulkan beta drivers.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #55 from mo78@abv.bg --- I can't tri it. With the stable NVIDIA driver 415.22.05 the game starts but the problems described in this topic ar all here. With version 415.22.05 Vulkan Beta the game doesn't start with the following error:
2019-01-24 22:47:22 [ 9] [DEBUG ] : starting launcher if needed X Error of failed request: BadValue (integer parameter out of range for operation) Major opcode of failed request: 151 (GLX) Minor opcode of failed request: 3 (X_GLXCreateContext) Value in failed request: 0x0 Serial number of failed request: 281 Current serial number in output stream: 282
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #56 from mo78@abv.bg --- Hmm, actually all non DXVK games won't start with the Vulkan Beta driver with the exactly same error. I just tried Stalker and Bloom Labyrinth and they doesn't start too. With the stable NVIDIA driver they works without problems at all. Do someone knows what is going on? Arch Linux here.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #57 from Béla Gyebrószki gyebro69@gmail.com --- https://source.winehq.org/git/wine.git/commit/3d4001f4fc84e73d83caf05550c411...
^ That's the commit which fixed the problem for me.
wine-4.0-96-gfee112f90a OpenGL vendor string: NVIDIA Corporation OpenGL renderer string: GeForce GT 730/PCIe/SSE2 OpenGL core profile version string: 4.6.0 NVIDIA 415.22.05
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #58 from mo78@abv.bg --- Did it work with the stable NVIDIA driver?
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #59 from Béla Gyebrószki gyebro69@gmail.com --- (In reply to mo78 from comment #58)
Did it work with the stable NVIDIA driver?
I tested with the following Nvidia drivers, all of them rendered water correctly: 410.93 (current stable) 415.22.05 (Vulkan beta) 415.27 (current beta)
For the sake of completeness, I tested this bug with the Uplay version of the game on Archlinux using vanilla Wine (git). The following components were installed in the prefix, everything else was set up as built-in: corefonts (needed to launch Uplay) mfc90.dll (needed to start the game) d3dx9_40.dll (missing textures in the main menu with built-in).
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #60 from mo78@abv.bg --- Thank you very much. I'll apply the patch and will try it :)
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #61 from Henri Verbeet hverbeet@gmail.com --- (In reply to mo78 from comment #55)
I can't tri it. With the stable NVIDIA driver 415.22.05 the game starts but the problems described in this topic ar all here. With version 415.22.05 Vulkan Beta the game doesn't start with the following error:
2019-01-24 22:47:22 [ 9] [DEBUG ] : starting launcher if needed X Error of failed request: BadValue (integer parameter out of range for operation) Major opcode of failed request: 151 (GLX) Minor opcode of failed request: 3 (X_GLXCreateContext) Value in failed request: 0x0 Serial number of failed request: 281 Current serial number in output stream: 282
If you didn't reboot yet since updating your OpenGL drivers, I think that's something that's known to cause that particular error with the proprietary NVIDIA drivers.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #62 from mo78@abv.bg --- Yes I know and hence I rebooted twice. No go :)
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #63 from Sergey Isakov isakov-sl@bk.ru --- Created attachment 63375 --> https://bugs.winehq.org/attachment.cgi?id=63375 The water is present
Big thanks! The bug is resolved. Tested on wine-4.0. Can be closed resolved.
https://bugs.winehq.org/show_bug.cgi?id=38274
Sergey Isakov isakov-sl@bk.ru changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution|--- |FIXED
--- Comment #64 from Sergey Isakov isakov-sl@bk.ru --- The water is transparent and blinking as it should.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #65 from mo78@abv.bg --- Can you please start the Crack Hack's campaign? Is everthyng alright there? For me everything is dark with white areas:
https://i.postimg.cc/3xn7DZ3b/Screenshot-20180118-131553.png
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #66 from mo78@abv.bg --- Anybody tried the Crack Hack's campaign? And would someone tell me if the patch is implemented in Wine 4.0 or there is need to apply it additionaly? Because with vanilla Wine the Crack Hack's campaign is messed up.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #67 from mo78@abv.bg --- I made a video. I first start a campaign without water and everything is rendered fine. Then I start Crack Hack's campaign and everything gets very dark and it gets unplayable. Even if you quit that campaign the game still remains bugged. Changing the video settings doesn't help either. So this bug isn't fixed for me.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #68 from mo78@abv.bg --- Hello, would someone opt in please? Is that hard to try the campaign and help me?
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #69 from Béla Gyebrószki gyebro69@gmail.com --- (In reply to mo78 from comment #68)
Hello, would someone opt in please? Is that hard to try the campaign and help me?
This bug report was closed as fixed. The problem you are experiencing has already been reported in bug #37705.
https://bugs.winehq.org/show_bug.cgi?id=38274
--- Comment #70 from mo78@abv.bg --- Ah yes, I didn't realize they are different problems. Thank you very much!
https://bugs.winehq.org/show_bug.cgi?id=38274
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #71 from Alexandre Julliard julliard@winehq.org --- Closing bugs fixed in 4.1.