- User Since
- Jan 2 2010, 2:22 AM (372 w, 4 d)
Feb 17 2016
This looks better :)
Feb 16 2016
It this done toward the end of drawing so that the full screen quad can take advantage of depth testing?
Feb 15 2016
How is this property different from bindId? Taking a quick look at the source code, it looks like this is meant to be an OpenGL name. However, it looks like there are some cases where it has not been set. Could we just make sure m_actTex gets set in more cases?
Feb 7 2016
Why is a checkbox needed under mapping? It looks like the Height Scale checkbox could be used for the same thing, especially since it looks like they are the same thing based on tooltips. The tooltips don't seem very helpful, especially since I don't know what SPOM is. Is there any documentation other than youtube videos?
Feb 4 2016
Overall it looks fine to me. Is this a setting that should be restricted to the Blender Game render engine? I'll need to apply the patch this evening to play around with it and get a feel for the user experience.
Feb 3 2016
Was the texture order issue sorted out? Also, I added some inline comments about some things that looked a little strange in the UI and shader to me.
Jan 21 2016
As far as I know, this is still a problem. You are welcome to try to reproduce the bug on a recent build and report your findings here.
Dec 7 2015
That derived mesh stuff is a little ugly, but, since immediate mode is being removed, I'll allow it.
Nov 4 2015
Looks good to me.
Aug 9 2015
My suggestion is to create a new API and deprecate the old one. This new API should be flexible to changes so this does not happen again if we want to add support for tessellation shaders.
Jun 26 2015
While I like the idea of more built in uniforms, I don't really see the need for the pre render filter. You can already use depth testing in a regular 2D filter to render a background, and it would be cheaper. I see some potential use with logic bricks, but I don't see many people using the built in filters specifically for a background (or at all really).
Jun 22 2015
I still want the shader APIs to be more consistent. Also, I don't see any documentation changes. With those documentation changes I would like to see a note that this is only partial geometry support since it does not support adjacency information.
Jun 6 2015
I am assuming in your earlier post you meant to say setSource instead of setSampler. I would like to see the setting of sources to be consistent. So either being able to set them all with setSource with keyword arguments, or adding a setVertexProg, setFragmentProg, and a function to call once all programs are set to actually handle the compiling. If possible it would be nice to replace one shader without replacing all of them. For example, being able to change the vertex shader and continue using the fragment shader that Blender generated. I haven't looked at that part of the code in a while to know how feasible that is.
May 12 2015
As long as the logic has been tested and works, this looks good to me.
May 3 2015
What is the purpose of this patch? What does it gain us? My intent for the LoD code was to keep it minimal in the BGE and to rely as much as possible on the BKE module to share as much code as possible between Blender and the BGE.
I do not have a Mac, just a Windows and Linux install. Are these changes sufficient, or do we need to do more to be properly aligned for Bullet? Also, what impact are you seeing with these changes?
Mar 31 2015
I'm not convinced that this is even necessary. What use cases do you see for this change? I am also not sold on the name. Something like "Write Depth", while still not great, might be a little more clear to a user what the option does. It also looks like you are displaying the option for both opaque and transparent materials, but you only have the option affecting transparent materials.
Feb 25 2015
Looks good to me.
Feb 23 2015
The UI still looks like it might be inconsistent with the check box on the Scene's LoD panel header controlling the use of global hysteresis, but the header label is "Level of Detail". It looks like that check box would disable/enable the use of levels of detail instead of controlling the hysteresis setting.
Feb 16 2015
This looks like a mix of refactoring as well as feature/functionality change. Could you separate this out into two patches make it easier to see what should be changing behavior, and what is simply cleanup/refactoring?
Feb 14 2015
The checkbox on the scene panel header seems a little confusing. The UI looks like it should be enabling/disabling LoDs, but the code seems to be enabling/disabling the global hysteresis setting.
Jan 22 2015
Looks good, I had one minor question that I inlined.
Jan 16 2015
I like those UI changes. It would probably be better to call the per object LoD check box "Hysteresis Override" to make it clear that it takes precedence over the global setting.
Jan 15 2015
Looks good to me, would be nice to get one more approval on this.
What about old files that already are using the combination of Character and Triangle Mesh?
Once that print is gone, you'll have my OK.
Hmm . . . fair point. We don't want to force existing projects to use this new feature. If you are only setting it to zero, I am pretty sure you don't even need to bother with a do_version, as I am pretty sure it will automatically be loaded as zero (probably due to a calloc).
Looks good to me. I'm busy this evening, but I can try to land this tomorrow.
Jan 14 2015
I believe I have been spelling it out in many user facing strings as "level of detail", and as "LoD" when I need to abbreviate it to save space. I vote for using "level of detail" in the documentation string (e.g. "index of the level of detail currently").
Looks like you might have screwed up Differential a bit, try uploading a diff that contains all of your changes for this (docs and the Python attribute). Currently it is only showing the doc change.
I think you meant to set hysteresis to 10 in the do_version, though. Otherwise, looks good to me.
Jan 9 2015
If the default value is 0, I don't think you need to bother with a doversion.
Looks good to me, just needs documentation.
Jul 30 2014
Jul 28 2014
Did an initial code review pass. There are a few methods I haven't looked through yet, including: bl_ConvertBlenderObject_Single, KX_BoneParentRelation:UpdateChildCoordinates, KX_NormalParentRelation:UpdateChildCoordinates, and KX_SlowParentRelation:UpdateChildCoordinates. Most of the in line comments are probably answered by commit messages, but it wouldn't hurt to answer them here as well.
Jul 26 2014
Is there any way for you to create a small file that reproduces the problem?
Jul 25 2014
Singletexture mode is no longer in the BGE.
◀ Merged tasks: T33151.
✘ Merged into T40684.
Meant to mark as incomplete instead of invalid.
Not enough information to reproduce the bug. As Sergey mentioned, really needs at least a file, but preferably more details about the problem (such as what is involved in "shooting at a monster").
✘ Merged into T33730.
◀ Merged tasks: T41167.
Looks like a duplicate of T33730
Jul 2 2014
Jun 27 2014
I don't think it is necessary. I experimented with adding and removing LoDs while in edit mode, and it did not appear to be a problem. To be honest, that poll method was most likely used because it was probably what the operator I was using as a reference used.
Something like this?
Jun 18 2014
Can not reproduce using Blender 2.70a on Windows 7 64bit.
May 12 2014
May 6 2014
Is this problem unique to the Game Engine, or does it apply to the viewport and the material system in general?
Since we no longer have Singletexture mode, is this still a valid bug report?
Can not reproduce using 2.70 on Windows 7 Pro 64bit and an NVIDIA GeForce GTX 260.
Apr 16 2014
SingleTexture mode no longer supported.
Feature is working as intended, no bug.
Can confirm the problem. Deleting the modifier works as well.
No clear steps to reproduce, no simple file demonstrating the problem, and the description is unclear as to what the problem is. Also, a more descriptive title would also be nice.
Ralson, could you please change the title to something more descriptive?
Closed by commit rB70f089b6f5b7.
Dec 5 2013
This sounds familiar. I build with MSVC 2012, but I will need to wait until I get home to double check this.