[Bf-taskforce25] defaults & tweaks list

[William Reynish]

Cheers Brecht.

Great points.


On 11 Jul, 2009, at 2:57 PM, Brecht Van Lommel wrote:

> Hi,
>
> I've been keeping a list of small things I think could change. Some  
> are
> default changes, others are small code tweaks. Most of these are quite
> simple, only a few are more work.
>
> Also something I noticed is that will some defaults have changed in
> the .B.blend, these are now out of sync with the defaults in the code
> when creating a new datablock (when not copying from an existing one).
> This is just a note to myself and others who do such changes, to keep
> these in sync.
>
> Brecht.
>
>
> Rendering
> ---------
>
> * Do Composite and Do Sequence enabled by default. But, they will only
> operate if there are nodes or sequencer strips respectively.

Good idea.

> * Stamp time,data,frame,scene,filename metadata into image file (not
> render). Rename it "metadata" perhaps?

Also good idea!

> * Larger bake margin.
> * Normal map default to tangent space for baking.

Yes, makes sense. Tangent space is what you want 99% of the time

> * Render to image window should render to an already open image  
> window,
> if there is no image open in that window
> * Panorama should be a camera, not scene setting.

Interesting idea. Makes sense.

> * Organize output formats by image/multilayer/movie in the menu.

Would be cool, yes. Could even add image/movie icons ;)

>
> More work:
> * Rather than FFMPEG and Quicktime, the format menu could have entries
> like Xvid, H.264, etc with simple quality controls, in addition to
> Custom Quicktime and Custom FFMPEG for people who like to do fine
> tuning.

Very nice idea.

>
> Shading
> -------
>
> * Texture rgb color something else than the purple. Probably best to
> still use a color, but a nicer one?

I heard lime green is the new 'in' color of 2009? ;)

> * Material preview sphere+sky could use nicer sky colors.
> * Default mist depth non-zero (maybe 8, half number of grid lines).
> * Object color in preview .blend should be white like the default.
> * Object material looks different with and without materials slots,  
> with
> no material assigned.
> * Texpace Mesh could be an option in the texture tab as an option for
> Orco, to clarify what it does. Probably not actually make it a  
> material
> option.

Yes, that would make much more sense.

>
> More work:
> * Texture influence, individual amount for each option.

Oh yes! Would be fantastic.

> * Replace sticky coordinates by UVs + texture mapping mode. Have not
> verified this is possible, but MSticky also consists of two  
> coordinates,
> so perhaps it is just a matter of interpolating UVs differently? In  
> any
> case, I don't think people use this option much if at all.
>
> Particles
> ---------
>
> * Particle system normal velocity default to 1.0.
> * Particle size much smaller (1.0 -> 0.05).
> * Random distribution by default.
> * Color by material by default.
> * Fields strength 0.0 -> 1.0.
> * Render emitter by default.

Great! This means you'd actually get a visual result when enabling  
particles.

>
> Files
> -----
>
> * Hide .dot files in file browser by default, and also hide tilde~
> files.
> * Enable/disable .blend file compression should be a property of
> the .blend file, not a user pref. a user pref could be to enable/ 
> disable
> it by default on new files.

Yes,


> Animation
> ---------
>
> * Autokey & autokey insert available enabled by default.

Hmm, I'd argue it's best to have 'only auto-insert keys on channels  
that are already animated' on. Otherwise you'd accidentally add  
keyframes when just moving object around which is usually not what you  
want. But objects that you've already animated you *are* likely to  
want to insert new keys for.

> * Graph/Dopesheet/NLA show selected only by default.
>
> Other
> -----
>
> * Wave modifier half default speed, and smaller UI range.
> * Remove Ev.M.Stack setting for collision, can be controlled with
> collision modifier position.
>
> More work:
> * Memory limit for Undo. But this needs some kind of auto-detection to
> obtain a reasonable limit depending on the computer?
>
>