Material Test Suite

Commander McLane · Post by **Commander McLane** » Tue Jun 15, 2010 7:21 am

[log.header]: Opening log for Oolite version 1.74 (x86-32 test release) under Mac OS X Version 10.4.11 (Build 8S2167) at 2010-06-14 17:43:27 +0300.
Machine type: MacBookPro1,1, 1536 MiB memory, 2 x x86 (Core/Yonah) @ 2000 MHz.
Oolite options: procedural planet textures, docking clearance, wormhole scanner, target incoming missiles, spoken messages, JavaScript console support, Debug plug-in support, OXP verifier, localization tools, debug GraphViz support.
...
[rendering.opengl.version]: OpenGL renderer version: 2.0.0 ("2.0 ATI-1.4.58")
Vendor: ATI Technologies Inc.
Renderer: ATI Radeon X1600 OpenGL Engine
[rendering.opengl.extensions]: OpenGL extensions (94):

Everything's fine.

JazHaz · Post by **JazHaz** » Tue Jun 15, 2010 10:03 am

On my non-shader computer all the tests ran and completed correctly.

In my log between the two markers, I got depreciation messages between the test reports. Is that expected?

Code: Select all

[materialTest.start]: Starting material test suite 1.0 under Oolite 1.74 and Windows (x86-32 test release) with OpenGL renderer "Mirage Graphics3", vendor "SiS"; shaders are not supported.
[materialTest.runTest]: Running test fixed-function:1 (diffuse_map).
[script.javaScript.warning.ooliteDefinedError]: ----- JavaScript warning (Automatic Chaff System 1.0): Ship.hasEquipment("EQ_ACS") is deprecated, use ship.equipmentStatus("EQ_ACS") == "EQUIPMENT_OK" instead.
[materialTest.runTest]: Running test fixed-function:2 (diffuse_map + emission_map).
[script.javaScript.warning.ooliteDefinedError]: ----- JavaScript warning (Automatic Chaff System 1.0): Ship.hasEquipment("EQ_ACS") is deprecated, use ship.equipmentStatus("EQ_ACS") == "EQUIPMENT_OK" instead.
[materialTest.runTest]: Running test fixed-function:3 (emission_map only).
[script.javaScript.warning.ooliteDefinedError]: ----- JavaScript warning (Automatic Chaff System 1.0): Ship.hasEquipment("EQ_ACS") is deprecated, use ship.equipmentStatus("EQ_ACS") == "EQUIPMENT_OK" instead.
[materialTest.runTest]: Running test fixed-function:4 (diffuse_map + illumination_map).
[script.javaScript.warning.ooliteDefinedError]: ----- JavaScript warning (Automatic Chaff System 1.0): Ship.hasEquipment("EQ_ACS") is deprecated, use ship.equipmentStatus("EQ_ACS") == "EQUIPMENT_OK" instead.
[materialTest.runTest]: Running test fixed-function:5 (illumination_map only).
[script.javaScript.warning.ooliteDefinedError]: ----- JavaScript warning (Automatic Chaff System 1.0): Ship.hasEquipment("EQ_ACS") is deprecated, use ship.equipmentStatus("EQ_ACS") == "EQUIPMENT_OK" instead.
[materialTest.runTest]: Running test fixed-function:6 (diffuse_map + emission_map + illumination_map).
[script.javaScript.warning.ooliteDefinedError]: ----- JavaScript warning (Automatic Chaff System 1.0): Ship.hasEquipment("EQ_ACS") is deprecated, use ship.equipmentStatus("EQ_ACS") == "EQUIPMENT_OK" instead.
[materialTest.runTest]: Running test fixed-function:7 (diffuse_map + emission_and_illumination_map).
[script.javaScript.warning.ooliteDefinedError]: ----- JavaScript warning (Automatic Chaff System 1.0): Ship.hasEquipment("EQ_ACS") is deprecated, use ship.equipmentStatus("EQ_ACS") == "EQUIPMENT_OK" instead.
[materialTest.complete]: Shader test suite complete.

Commander McLane · Post by **Commander McLane** » Wed Jun 16, 2010 12:22 pm

JazHaz wrote:
On my non-shader computer all the tests ran and completed correctly.

In my log between the two markers, I got depreciation messages between the test reports. Is that expected?

Only if you had other OXPs with deprecated code installed next to the test suite.

JensAyton · Post by **JensAyton** » Fri Jun 18, 2010 9:43 am

Come on, you people. I know there’s at least one more GPU on which full shaders don’t work. *taps foot*

An additional request: please also report if your system supports full shaders, but runs significantly slower that way.

Pluisje · Post by **Pluisje** » Fri Jun 18, 2010 10:58 am

Ok, on my MacMini the test was successful. Except for the full shader mode of course. Cube 13 did look a little strange. It's the specular only I think. So it gives very low contrast except when the light hits it just right. I think that is how it should be. The normal maps looked really nice. The full shader cubes aren't all black by the way. Sometimes the sides are red, blue and green with a lot of black.

I'm sorry to say I find the space dust when not in torus drive quite annoying. When sniping at ships I occasionally hunt a space dust particle instead of one of the ships in the faraway group

. The dust particles look 25 pixels square (I mean 5x5) which is not much smaller than a sidewinder at long range, or even at mid range.

Edit: as I posted the above while working on my laptop, I didn't have all data available. Here's the log.

[log.header]: Opening log for Oolite version 1.74 (x86-32 test release) under Mac OS X Version 10.5.8 (Build 9L30) at 2010-06-14 22:06:07 +0200.
Machine type: Macmini2,1, 1024 MiB memory, 2 x x86 (Core 2/Merom) @ 1830 MHz.
[rendering.opengl.version]: OpenGL renderer version: 1.2.0 ("1.2 APPLE-1.5.48")
Vendor: Intel Inc.
Renderer: Intel GMA 950 OpenGL Engine
[rendering.opengl.extensions]: OpenGL extensions (77): ...

I see the GMA 950 was already tested on page 1 in almost the same setup. Still another one is always nice to have, right..

Post by **Eric Walch** » Fri Jun 18, 2010 12:05 pm

Pluisje wrote:
Ok, on my MacMini the test was successful. Except for the full shader mode of course. Cube 13 did look a little strange. It's the specular only I think.

That specular map should also be visible on nr 14 and 15. Even on cube nr 16 but there it is much fainter with me and only visible on one of the faces. On the other faces it is to faint to be seen. (My mac uses a GeForce 9400)

But what graphical processor does your mini have? Its shown at the end of the test. Mac Mini by itself is not conclusive. The just new released mac Mini ships with a NVIDIA GeForce 320M. Those since early 2009 have a NVIDIA GeForce 9400M. Both are full shader capable so you must have an older version.

I think the only graphic chip used before on the mini was the intel GMA950?

Pluisje · Post by **Pluisje** » Fri Jun 18, 2010 12:58 pm

Eric Walch wrote:
Those since early 2009 have a NVIDIA GeForce 9400M. Both are full shader capable so you must have an older version.

I think the only graphic chip used before on the mini was the Intel GMA950?

Yes, I have the 2007 version, it was half price in 2009 and being Dutch and all, I couldn't resist it at the time. I've edited my original post to have everything in one place.

Rebecca · Post by **Rebecca** » Fri Jun 18, 2010 6:09 pm

[materialTest.start]: Starting material test suite 1.0 under Oolite 1.74 and Windows (x86-32 test release) with OpenGL renderer "GeForce 8600M GT/PCI/SSE2", vendor "NVIDIA Corporation"; shaders are fully supported, texture image unit count is 32.

all worked for me!!!!!!

B.

drew · Post by **drew** » Fri Jun 18, 2010 6:21 pm

Not so great here, but I do have a 10 year old Dell X200

OpenGL renderer version: 1.3.0 ("1.3.0 - Build 4.14.10.3889")
Vendor: Intel
Renderer: Intel 830M

...

[materialTest.start]: Starting material test suite 1.0 under Oolite 1.74 and Windows (x86-32 test release) with OpenGL renderer "Intel 830M", vendor "Intel"; shaders are not supported.

First 7 spinning cubes looked good. Couldn't shoot them though.

Cheers,

Drew.

Post by **Disembodied** » Sat Jun 19, 2010 10:25 am

[log.header]: Opening log for Oolite version 1.74 (x86-32 test release) under Mac OS X Version 10.4.11 (Build 8S2167) at 2010-06-19 10:57:10 +0100.
Machine type: iMac5,1, 1024 MiB memory, 2 x x86 (Core 2/Merom) @ 2000 MHz.
...
[rendering.opengl.version]: OpenGL renderer version: 2.0.0 ("2.0 ATI-1.4.56")
Vendor: ATI Technologies Inc.
Renderer: ATI Radeon X1600 OpenGL Engine
[rendering.opengl.extensions]: OpenGL extensions (94):

Situation nominal ...

Post by **Getafix** » Sat Jun 19, 2010 1:12 pm

Code: Select all

    O/S: Windows Vista 32bit & Linux Ubuntu 8.04 64bit
    gfx: nVidia GeForce 8600 GT/PCI/SSE2
 OpenGL: v3.2.0
 driver: Windows 197.45 & Linux 195.36.24
shaders: full support

All material tests seem good.
Dust fades when simple or full shaders,
and goes Black & White (cheers!) when shaders are off.

Post by **Getafix** » Sat Jun 19, 2010 1:30 pm

Code: Select all

    O/S: Windows XP SP3 32bit
    gfx: Intel Cantiga (GMA 4500MHD)
 OpenGL: 2.0.0
 driver: 6.14.10.5068
shaders: full support

All material tests seem good.
Dust fades (not as smoothly as in the nVidia box) when simple or full shaders,
and goes Black & White (cheers again!) when shaders are off.

Post by **Getafix** » Sat Jun 19, 2010 2:09 pm

Code: Select all

    O/S: Linux Ubuntu 8.04 32bit
    gfx: Intel 845G
 OpenGL: 1.3.0
 driver: 7.0.3-rc2 (Matched GPU configuration "Intel GMA 900/950 family")
shaders: No

Only first seven material tests can be executed.
No fade effect for space dust particles.

Post by **Getafix** » Sat Jun 19, 2010 3:15 pm

Code: Select all

     O/S: vBox3.2.0 Linux Ubuntu 10.04 32bit (with Guest Additions installed)
     gfx: InnoTek Systemberatung GmbH VirtualBox Graphics Adapter
host gfx: Intel Cantiga (GMA 4500MHD) 
  OpenGL: 2.0.0
  driver: Chromium 1.9.0, vendor Humper
 shaders: full support

All material tests can be executed but many effects fail.
A screenshots tarball of all tests can be downloaded from here (8.1MB file).

Space dust particles do not appear at all in simple or full shader mode.
When shader-effects are off, space dust particles appear normally, without the fading effect, of course.

JensAyton · Post by **JensAyton** » Sat Jun 19, 2010 3:41 pm

Getafix wrote:
Code: Select all
     O/S: vBox3.2.0 Linux Ubuntu 10.04 32bit (with Guest Additions installed)
     gfx: InnoTek Systemberatung GmbH VirtualBox Graphics Adapter
host gfx: Intel Cantiga (GMA 4500MHD) 
  OpenGL: 2.0.0
  driver: Chromium 1.9.0, vendor Humper
 shaders: full support
All material tests can be executed but many effects fail.
A screenshots tarball of all tests can be downloaded from here (8.1MB file).

Space dust particles do not appear at all in simple or full shader mode.
When shader-effects are off, space dust particles appear normally, without the fading effect, of course.

This looks oddly familiar, as if we discussed it on the mailing list on the tenth and eleventh of June. ;-)

The VirtualBox driver is lying about supporting shaders. Complain to them about it. Also, you’ll find the space dust hiding over at the witchpoint.

Oolite Bulletins