Are IHVs going to eventually get rid of trilinear filtering?

[jvd]

yea but do you know how long the game will last ?

It would be months before we even get into the game !!!!!!

 

[vb]

yea but do you know how long the game will last ?

It would be months before we even get into the game !!!!!!

But by the time we finish it we will be able to play Half-life 2 demo!!

 

[jvd]

yea but do you know how long the game will last ?

It would be months before we even get into the game !!!!!!

But by the time we finish it we will be able to play Half-life 2 demo!!

 

[Martillo1]

I say screw mipmaps... Lets just have the base textures at full resolution on everything. 8)

I don't know if to laugh or to "cray"

 

[Kristof]

A wieghted mipmap blend would probably be better than Trilinear.

Aren't they the same thing, or am I missing something important here

Yes, Dave... Trilinear is a weighted MipMap blend...

K-

 

[Simon F]

A wieghted mipmap blend would probably be better than Trilinear.

Aren't they the same thing, or am I missing something important here

Yes, Dave... Trilinear is a weighted MipMap blend...

K-

And here's the abstract of the paper that introduced it all. I can't find the actual paper online (except for the ACM site which requires subscription) but it goes on to coin the phrase "MIP map".

 

[Quitch]

I think mip-mapping and filtering should be left to hardware/driver, and the app should only supply the base texture + filtering hints.

I strongly disagree.
The driver will never guess what should the mipmaps look like when the texture contains custom data.
Even simple normal map mipmaps are non-trivial.

You can always ask for point sampling and/or supply textures in sets. but on "classic" textures the driver/hardware should be able to choose what is appropriate for the hardware implementation.

You could leave it to the driver. Or you could leave it to the app.

Or, you could leave it to the end-users. Or are they too stupid to know what they want?

I'd say that you're right, end-users are stupid. How many of them do you think know what trilinear filtering does? How many know what its down-side is? How many know it has a down-side?

Relying on end-users for anything technical should always be a last resort.

 

[glw]

From the conclusion to the paper Pyramidal Parametrics
(Williams - Computer Graphics - July 1983 - Vol. 17 No. 3):

By continuously varying the detail with which data are resolved,
pyramidal parametrics provide economical approximate solutions
to filtering problems in mapping texture and illumination onto surfaces,
and preliminary experiments suggest they may provide flexible
surface representations as well.

'Approximate' - worth remembering when trilinear is described as
some sort of gold-standard in filtering.

It's also interesting to note that even in 1983 displacement mapping
was considered.

 

[nutball]

I'd say that you're right, end-users are stupid. How many of them do you think know what trilinear filtering does? How many know what its down-side is? How many know it has a down-side?

Relying on end-users for anything technical should always be a last resort.

I was including the folks who post on this board as "end-users".

Yeah, the vast majority of end-users kow nothing of tri-linear filtering, etc., etc., etc. So this new algorithm is great for them, let it be on by default. Fine.

I'd still like the opportunity to judge for myself ATIs assertion that they know what I want better than I do. I don't care if the option is a registry hack, or some tick box 7 levels down from the control panel, I want it (and it is my money after all).

 

[Quitch]

I'd say that if the reaction on this board were anything to go by, I would never include the option. In the first few hours there was a real knee jerk reaction against this method, without a scrap of proof that it was inferior (and many inferences that it was in fact, quite superior).

Including that option is going to lead to more stupid reviewer statements than my poor little mind can handle. The whole brilinear episode has led to trilinear being considered the ultimate in IQ, even though it quite painly is no such thing. When this level of confusion exists, who in their right mind would want to put control of such a thing in the hands of the end-user?