Poster | Thread |
TRIPOS
| |
Proposed MorphOS GFX/Driver evolution path Posted on 29-May-2022 21:14:52
| | [ #1 ] |
|
|
|
Super Member |
Joined: 4-Apr-2014 Posts: 1205
From: Unknown | | |
|
| Over at Morph.zone, bigfoot is proposing a bounty based evolution path of current MorphOS that could happen in several steps/targets.
Here is one:
Quote:
Target 5: R600 driver
A TinyGL driver for the R600, R700, Evergreen and Northern Islands families of Radeon chips. Writing this driver would depend on the first target listed above, implementing the fixed-function OpenGL pipeline as shaders, to avoid doing duplicate work.
Such a driver would enable OpenGL support for many commonly available PCI Express graphics cards, which can be used in for example the Sam460, the X5000 or PCI Express Powermac G5s. This hardware also supports many more features required for OpenGL 3 and beyond. |
Please go there and read the full post and its subsequent posts and clarifications, and then contribute with some input!
https://morph.zone/modules/newbb_plus/viewtopic.php?topic_id=13285&forum=49&start=37 |
|
Status: Offline |
|
|
t0lkien
| |
Re: Proposed MorphOS GFX/Driver evolution path Posted on 30-May-2022 18:08:43
| | [ #2 ] |
|
|
|
Regular Member |
Joined: 25-Sep-2004 Posts: 203
From: SPAIN | | |
|
| This is soooo cool! Thanks for the info! |
|
Status: Offline |
|
|
Rob
| |
Re: Proposed MorphOS GFX/Driver evolution path Posted on 1-Jun-2022 19:44:08
| | [ #3 ] |
|
|
|
Elite Member |
Joined: 20-Mar-2003 Posts: 6385
From: S.Wales | | |
|
| @TRIPOS
There are AGP versions of Radeon cards up to at least HD5670 so in theory it should open up options for G4 and G5 systems with an AGP slot. Has anyone tested such cards in 2D? |
|
Status: Offline |
|
|
TRIPOS
| |
Re: Proposed MorphOS GFX/Driver evolution path Posted on 1-Jun-2022 20:48:47
| | [ #4 ] |
|
|
|
Super Member |
Joined: 4-Apr-2014 Posts: 1205
From: Unknown | | |
|
| @Rob
Quote:
Rob wrote: @TRIPOS
There are AGP versions of Radeon cards up to at least HD5670 so in theory it should open up options for G4 and G5 systems with an AGP slot. Has anyone tested such cards in 2D? |
Scroll down the GFX card section a bit, and you will see what kind of support there currently is for various Radeons:
https://www.morphos-team.net/hardware |
|
Status: Offline |
|
|
Rob
| |
Re: Proposed MorphOS GFX/Driver evolution path Posted on 1-Jun-2022 21:26:38
| | [ #5 ] |
|
|
|
Elite Member |
Joined: 20-Mar-2003 Posts: 6385
From: S.Wales | | |
|
| @TRIPOS
I mean has anyone tested any of the AGP versions on AGP Mac hardware. The cards carry a premium in AGP format but if say A 4650 works in a G4 mac it might be worth gettin Bigfoot to focus on R700 first. |
|
Status: Offline |
|
|
t0lkien
| |
Re: Proposed MorphOS GFX/Driver evolution path Posted on 2-Jun-2022 17:41:39
| | [ #6 ] |
|
|
|
Regular Member |
Joined: 25-Sep-2004 Posts: 203
From: SPAIN | | |
|
| |
Status: Offline |
|
|
pavlor
| |
Re: Proposed MorphOS GFX/Driver evolution path Posted on 3-Jun-2022 15:28:31
| | [ #7 ] |
|
|
|
Elite Member |
Joined: 10-Jul-2005 Posts: 9636
From: Unknown | | |
|
| @TRIPOS
Nice! I hope there will at least some of these cards for sale, when the driver is ready (this issue is becoming more and more frustrating, at least here where I live). |
|
Status: Offline |
|
|
smf
| |
Re: Proposed MorphOS GFX/Driver evolution path Posted on 3-Jun-2022 21:37:00
| | [ #8 ] |
|
|
|
Regular Member |
Joined: 15-Mar-2003 Posts: 333
From: Växjö, Sweden | | |
|
| @TRIPOS
Ehhw. I have already payed for my morphos license once and now i have to pay even more just so i can play doom3? I dont expect the other ~60 active morphos users to pay the rest so it's up to me or some rich mother ####er but i will not pay a dime again until morphos is ported to arm. Bigfoot used to be cool but now he is just greedy.
|
|
Status: Offline |
|
|
t0lkien
| |
Re: Proposed MorphOS GFX/Driver evolution path Posted on 4-Jun-2022 5:11:03
| | [ #9 ] |
|
|
|
Regular Member |
Joined: 25-Sep-2004 Posts: 203
From: SPAIN | | |
|
| @smf
I payed for two 3.0 MorphOS licenses and get 10 years with 17 free updates. I think it is not so bad. But hey! you are free to contibute or not. |
|
Status: Offline |
|
|
recedent
| |
Re: Proposed MorphOS GFX/Driver evolution path Posted on 4-Jun-2022 9:09:59
| | [ #10 ] |
|
|
|
Regular Member |
Joined: 28-Jan-2010 Posts: 227
From: Tarnów | | |
|
| @smf
Quote:
it's up to me (...) but i will not pay a dime again until morphos is ported to arm |
Oh heavens! Whatever will we do?
In the meantime - we're already at 18% without any rich mother ####er contributing. I guess there's still a chance you will play Doom 3 with no added cost (unfortunately that's not quite true as you'll have to buy a copy of game to do that - tough luck). |
|
Status: Offline |
|
|
kamelito
| |
Re: Proposed MorphOS GFX/Driver evolution path Posted on 4-Jun-2022 9:22:07
| | [ #11 ] |
|
|
|
Cult Member |
Joined: 26-Jul-2004 Posts: 832
From: Unknown | | |
|
| @t0lkien
This also mean no money for full time developer so a slower pace of development. IMO they should have switched to X64 first. Last edited by kamelito on 04-Jun-2022 at 09:22 AM.
|
|
Status: Offline |
|
|
t0lkien
| |
Re: Proposed MorphOS GFX/Driver evolution path Posted on 4-Jun-2022 9:53:33
| | [ #12 ] |
|
|
|
Regular Member |
Joined: 25-Sep-2004 Posts: 203
From: SPAIN | | |
|
| @kamelito
And perhaps thats why the reason this bounty. But we are never happy.
Yes, a x64 would be nice. |
|
Status: Offline |
|
|
t0lkien
| |
Re: Proposed MorphOS GFX/Driver evolution path Posted on 9-Jun-2022 18:57:52
| | [ #13 ] |
|
|
|
Regular Member |
Joined: 25-Sep-2004 Posts: 203
From: SPAIN | | |
|
| Bigfoot: Funding is now complete! thanks to all of you who contributed, and "of course a special thanks to pampers who contributed a massive 5,000 euro today, immediately making the funding cross the finish line!" |
|
Status: Offline |
|
|
t0lkien
| |
Re: Proposed MorphOS GFX/Driver evolution path Posted on 4-Sep-2022 21:27:51
| | [ #14 ] |
|
|
|
Regular Member |
Joined: 25-Sep-2004 Posts: 203
From: SPAIN | | |
|
| |
Status: Offline |
|
|
t0lkien
| |
Re: Proposed MorphOS GFX/Driver evolution path Posted on 8-Sep-2022 9:51:40
| | [ #15 ] |
|
|
|
Regular Member |
Joined: 25-Sep-2004 Posts: 203
From: SPAIN | | |
|
| |
Status: Offline |
|
|
bennymee
| |
Re: Proposed MorphOS GFX/Driver evolution path Posted on 9-Sep-2022 21:01:19
| | [ #16 ] |
|
|
|
Cult Member |
Joined: 19-Aug-2003 Posts: 697
From: Netherlands | | |
|
| @t0lkien
But what is new ?
A link to a .lha download on a non Amiga device does not work... Last edited by bennymee on 09-Sep-2022 at 09:01 PM.
|
|
Status: Offline |
|
|
t0lkien
| |
Re: Proposed MorphOS GFX/Driver evolution path Posted on 10-Sep-2022 16:56:02
| | [ #17 ] |
|
|
|
Regular Member |
Joined: 25-Sep-2004 Posts: 203
From: SPAIN | | |
|
| @bennymee
I can download the file even with my trusty Android phone but here are the changes:
Changes from 2022-09-04 to 2022-09-08 ------------------------------------- tinygl.library: - Library version is now 53.2. - Reading gl_ModelViewProjectionMatrix in a shader now works correctly. - Shaders now work correctly after window resize. - Similarly fixed a corner case issue with textures after window resize. - Fixed a bug making program objects without both a vertex and a fragment shader present fail.
r300.library: - Library version is now 53.6. - Now supports shaders that read back from output registers. - Now supports shaders which use write masks in combination with texture instructions. - Fixed the fragment shader compiler to ignore source register swizzles for unused source register components in texture instructions. - Optimised the swizzle emulation of the fragment shader compiler, resulting in smaller and faster compiled shaders.
Changes from 2022-08-05 to 2022-09-04 ------------------------------------- tinygl.library: - Library version is now 53.1. - Added support for OpenGL shaders. - Fixed glMultiTexCoord#?(), which was broken in the previous beta release. - Fixed glClientActiveTexture() to properly validate app input and report the required OpenGL errors. - Fixed glBindTexture() to disallow binding a texture to a target where the texture's type doesn't match. - Reworked how cube map textures are handled, cube map texture support should now be a bit more robust. - The ARB program parser now respects driver limits when parsing ARB programs. - Increased the number of supported texture units from 4 to 8. - No longer skips array-based draws where the coord array isn't enabled. This is useful for program/shader-based rendering which pass vertex coords in a generic array. - New extensions advertised, depending on driver support: o GL_ARB_fragment_shader o GL_ARB_shader_objects o GL_ARB_shading_language_100 o GL_ARB_vertex_shader
r200.library: - Library version is now 53.3. - Added support for GL_TEXTURE_WRAP_R.
r300.library: - Library version is now 53.5. - Added support for cube maps. - Fixed ARB programs to work correctly regardless of which fixed function texture units are enabled. - Added support for reading the fragment window position in fragment programs. - Increased the number of supported texture units from 4 to 7. - Fixed the handling of killing/discarding a fragment in fragment programs. - Fixed the fragment program compiler to emulate unsupported instructions and source operand modifications. - Fixed to handle outputting colour values in fragment program stages other than the last stage.
Changes since MorphOS 3.17 -------------------------- tinygl.library: - Library version is now 53.2. - Cleaned up the OpenGL extensions advertised by TinyGL: o GL_EXT_texture_compression_s3tc was advertised with the wrong name, EXT_texture_compression_s3tc o The following non-existing extensions are no longer advertised: * GL_ARB_color_table * GL_ARB_compiled_vertex_array * GL_ARB_draw_range_elements * GL_EXT_color_table o The following extensions do exist, but are not supported, and thus no longer advertised: * GL_EXT_draw_range_elements o New driver-independent (or supported by all current drivers) extensions advertised: * GL_ARB_imaging * GL_ARB_texture_border_clamp * GL_ARB_texture_env_dot3 * GL_ARB_texture_mirrored_repeat * GL_ARB_texture_non_power_of_two * GL_ARB_transpose_matrix * GL_ARB_window_pos * GL_EXT_abgr * GL_EXT_blend_color * GL_EXT_blend_equation_separate * GL_EXT_blend_func_separate * GL_EXT_blend_minmax * GL_EXT_blend_subtract * GL_EXT_multi_draw_arrays * GL_EXT_texture_env_dot3 * GL_SGI_color_table * New driver-dependent extensions advertised. Availability depends on the driver in use: * GL_ARB_fragment_program * GL_ARB_fragment_shader * GL_ARB_shader_objects * GL_ARB_shading_language_100 * GL_ARB_texture_cube_map * GL_ARB_vertex_program * GL_ARB_vertex_shader * GL_EXT_texture_cube_map - Every OpenGL 2.1 function now exists in the tinygl.library base. Note that this doesn't mean that TinyGL supports all of OpenGL 2.1's functionality yet, just that the functions exists. - Added support for cube mapping. (extensions: GL_ARB_texture_cube_map, GL_EXT_texture_cube_map) - Added support for OpenGL ARB programs. (extensions: GL_ARB_fragment_program, GL_ARB_vertex_program) - Added support for OpenGL shaders. - Many OpenGL functions that previously only had some variants of otherwise supported functions available now support all variants. These are functions whose names typically end with a single digit (1, 2, 3 or 4) followed by 1 or two characters to indicate argument type (b, d, f, i, s, ub, ui, us) and then optionally a v, for example glColor3ub(). These functions are: o glColor#?() o glEvalCoord#?() o glMultiTexCoord#?() o glNormal#?() o glRasterPos#?() o glTexCoord#?() o glVertex#?() o glWindowPos#?() - Updated glGetDoublev, glGetFloatv and glGetIntegerv to fix bugs in numerous already supported tags, and added many previously unsupported tags. The following tags are newly supported: o GL_BLEND_COLOR o GL_BLEND_EQUATION_ALPHA o GL_BLEND_EQUATION_RGB o GL_COLOR_WRITEMASK o GL_CURRENT_RASTER_SECONDARY_COLOR o GL_DEPTH_CLEAR_VALUE o GL_DITHER o GL_FOG_COLOR o GL_FOG_DENSITY o GL_FOG_END o GL_FOG_MODE o GL_FOG_START o GL_LIST_BASE o GL_LIST_INDEX o GL_LIST_MODE o GL_MAX_COMBINED_TEXTURE_IMAGE_UNITS o GL_MAX_TEXTURE_COORDS o GL_MAX_TEXTURE_IMAGE_UNITS o GL_PROGRAM_ERROR_POSITION_ARB o GL_STENCIL_CLEAR_VALUE o GL_STENCIL_FAIL o GL_STENCIL_FUNC o GL_STENCIL_PASS_DEPTH_FAIL o GL_STENCIL_PASS_DEPTH_PASS o GL_STENCIL_REF o GL_STENCIL_TEST o GL_STENCIL_VALUE_MASK o GL_STENCIL_WRITEMASK o GL_TRANSPOSE_MODELVIEW_MATRIX o GL_TRANSPOSE_PROJECTION_MATRIX o GL_TRANSPOSE_TEXTURE_MATRIX - Added support for the following state getter functions: o glGetBooleanv() o glGetLightiv() o glGetMaterialiv() o glGetTexGendv() o glGetTexGenfv() o glGetTexGeniv() o glGetTexLevelParameterfv() - Added support for the following state setter functions: o glFogiv() o glLightiv() o glLightModeliv() o glMaterialiv() o glPointParameterf() o glPointParameterfv() o glPointParameteri() o glPointParameteriv() o glTexGeniv() o glTexGendv() - Added support for the following glGetLight#?() tags: o GL_CONSTANT_ATTENUATION o GL_LINEAR_ATTENUATION o GL_QUADRATIC_ATTENUATION o GL_SPECULAR o GL_SPOT_CUTOFF o GL_SPOT_DIRECTION o GL_SPOT_EXPONENT - Added support for glMultiDrawArrays() and glMultiDrawElements(). - Added support for glFinish(). - glGetError() is now implemented and will - for some functions - return standards compliant error codes. - Added glRectsv(). - Fixed some of the default states in freshly created TinyGL contexts to match what is required by OpenGL: o Secondary colour o Raster colour o Raster secondary colour o Texture coordinates for texture units > 0 - glGetTexLevelParameter#?() now supports queries of the GL_TEXTURE_1D target. - Fixed glClientActiveTexture() to properly validate app input and report the required OpenGL errors. - Fixed glBindTexture() to disallow binding a texture to a target where the texture's type doesn't match. - Increased the number of supported texture units from 4 to 8. - No longer skips array-based draws where the coord array isn't enabled. This is useful for program/shader-based rendering which pass vertex coords in a generic array. - A function to obtain OpenGL function pointers at runtime, tglGetProcAddress(), has been added. R200: - Library version is now 53.3. - Added support for cube maps. - Added support for GL_TEXTURE_WRAP_R. - Texture combiner support has been fixed to be be standards compliant: o Added/fixed the support for the GL_ADD, GL_ADD_SIGNED, GL_DOT3_RGB, GL_DOT3_RGBA, GL_INTERPOLATE and GL_SUBTRACT texture combiner functions. o Added support for texture combiner output scaling. o Any and every combination of GL_SRC_COLOR, GL_ONE_MINUS_SRC_COLOR, GL_ALPHA and GL_ONE_MINUS_SRC_ALPHA operands now work.
R300, R400 and R500: - Library version is now 53.6. - Added support for cube maps. - Increased the number of supported texture units from 4 to 7. - Texture combiner support has been fixed to be be standards compliant: o Added/fixed the support for the GL_ADD, GL_ADD_SIGNED, GL_DOT3_RGB, GL_DOT3_RGBA, GL_INTERPOLATE and GL_SUBTRACT texture combiner functions. o Added support for texture combiner output scaling. o Any and every combination of GL_SRC_COLOR, GL_ONE_MINUS_SRC_COLOR, GL_ALPHA and GL_ONE_MINUS_SRC_ALPHA operands now work.
R300/R400: - Added support for OpenGL ARB programs. - Added support for OpenGL shaders.
R500: - Fixed a bug that would cause wrong rendering if an application had gaps in its enabled texture units, for example if only texture units 0 and 2 were enabled.
OpenGL test suite: - Added texture combiner tests to my OpenGL test suite. All drivers and hardware combinations now pass all tests.
SDK: - Please note that any software compiled with the updated SDK will require tinygl.library version 53.0 or higher, and thus either MorphOS 3.18 or MorphOS 3.17 + this separate TinyGL update. If you manually open tinygl.library, you _must_ update your code to request version 53! - TinyGL's automatic libopen function now requires at least version 53 of tinygl.library. - TinyGL's link library now comes in both regular and baserel flavours. - Newly supported OpenGL extensions that introduce new functions to OpenGL are disabled by default. To enable the new extensions, call tglEnableNewExtensions(0). The passed argument _must_ be zero. If you call tglEnableNewExtensions(), then you _must_ use tglGetProcAddress() to obtain pointers to OpenGL functions present in OpenGL extensions. You _must not_ use a custom function for this purpose. - Includes have been updated to fully support OpenGL 2.1, plus the following extensions: o GL_ARB_color_buffer_float o GL_ARB_depth_texture o GL_ARB_draw_buffers o GL_ARB_fragment_program o GL_ARB_fragment_program_shadow o GL_ARB_fragment_shader o GL_ARB_framebuffer_object o GL_ARB_half_float_pixel o GL_ARB_imaging o GL_ARB_matrix_palette o GL_ARB_multisample o GL_ARB_multitexture o GL_ARB_occlusion_query o GL_ARB_pixel_buffer_object o GL_ARB_point_parameters o GL_ARB_point_sprite o GL_ARB_shader_objects o GL_ARB_shading_language_100 o GL_ARB_shadow o GL_ARB_shadow_ambient o GL_ARB_texture_border_clamp o GL_ARB_texture_compression o GL_ARB_texture_cube_map o GL_ARB_texture_env_add o GL_ARB_texture_env_combine o GL_ARB_texture_env_crossbar o GL_ARB_texture_env_dot3 o GL_ARB_texture_float o GL_ARB_texture_mirrored_repeat o GL_ARB_texture_non_power_of_two o GL_ARB_texture_rectangle o GL_ARB_transpose_matrix o GL_ARB_vertex_blend o GL_ARB_vertex_buffer_object o GL_ARB_vertex_program o GL_ARB_vertex_shader o GL_ARB_window_pos o GL_ATI_fragment_shader o GL_ATI_separate_stencil o GL_EXT_abgr o GL_EXT_bgra o GL_EXT_blend_color o GL_EXT_blend_equation_separate o GL_EXT_blend_func_separate o GL_EXT_blend_logic_op o GL_EXT_blend_minmax o GL_EXT_blend_subtract o GL_EXT_color_subtable o GL_EXT_color_table o GL_EXT_compiled_vertex_array o GL_EXT_convolution o GL_EXT_draw_range_elements o GL_EXT_fog_coord o GL_EXT_histogram o GL_EXT_multi_draw_arrays o GL_EXT_packed_pixels o GL_EXT_paletted_texture o GL_EXT_point_parameters o GL_EXT_rescale_normal o GL_EXT_secondary_color o GL_EXT_separate_specular_color o GL_EXT_shadow_funcs o GL_EXT_shared_texture_palette o GL_EXT_stencil_two_side o GL_EXT_stencil_wrap o GL_EXT_texture3D o GL_EXT_texture_compression_s3tc o GL_EXT_texture_cube_map o GL_EXT_texture_env_add o GL_EXT_texture_env_combine o GL_EXT_texture_env_dot3 o GL_EXT_texture_filter_anisotropic o GL_EXT_texture_lod o GL_EXT_texture_lod_bias o GL_EXT_texture_sRGB o GL_HP_convolution_border_modes o GL_NV_blend_square o GL_NV_texgen_reflection o GL_S3_s3tc o GL_SGIS_generate_mipmap o GL_SGIS_multitexture o GL_SGIS_texture_edge_clamp o GL_SGIS_texture_lod o GL_SGI_color_matrix o GL_SGI_color_table o GL_SUN_multi_draw_arrays
|
|
Status: Offline |
|
|
ferrels
| |
Re: Proposed MorphOS GFX/Driver evolution path Posted on 10-Sep-2022 17:15:27
| | [ #18 ] |
|
|
|
Cult Member |
Joined: 20-Oct-2005 Posts: 922
From: Arizona | | |
|
| @t0lkien
Impressive change log. Keep up the good work! |
|
Status: Offline |
|
|
bennymee
| |
Re: Proposed MorphOS GFX/Driver evolution path Posted on 11-Sep-2022 14:13:32
| | [ #19 ] |
|
|
|
Cult Member |
Joined: 19-Aug-2003 Posts: 697
From: Netherlands | | |
|
| |
Status: Offline |
|
|
pavlor
| |
Re: Proposed MorphOS GFX/Driver evolution path Posted on 12-Sep-2022 4:51:46
| | [ #20 ] |
|
|
|
Elite Member |
Joined: 10-Jul-2005 Posts: 9636
From: Unknown | | |
|
| |
Status: Offline |
|
|