Originally posted by drag
View Post
Announcement
Collapse
No announcement yet.
What Was Decided With S3TC & Floating Points For Mesa
Collapse
X
-
-
Originally posted by FireBurn View PostIf I understand the process correctly Mesa has to be rebuild either way if the S3TC work is brought in tree or left as a library
If it's left as a library and it isn't present at compile time Mesa won't have S3TC support again as per my understanding (and I might be wrong)
I'm guessing it would be better to bring it in tree to prevent the library falling out of sync with the rest of Mesa
Comment
-
Originally posted by elanthis View Post"Open" means that the specification is available to read and study free of royalty; nothing else. You are confusing "open specifications" with "open implementations."
And again, the definition of "open standard" varies according to whom makes that definition. Only last week or so the UK government provided the terms it uses to define open standards, which require that they have "intellectual property made irrevocably available on a royalty free basis". This puts the UK in line with other countries, most notably the EU.
So the confusion only exists if you decide to give more credit to the world views of Nvidia, AMD, Apple and the likes than to what common sense dictates and quite a few of (presumably) democratically elected governments throughout the world understand.
And no, it couldn't have been done any other way. There is an actual _reason_ why people want OpenGL 3 support including all of its mandatory features; you realize this, right? Floating point textures are _essential_ for many modern rendering techniques. You cannot do them without floating point textures and render buffers. Period.
PS. I believe you are confusing "open implementations" with "free to implement".
Comment
-
Originally posted by ?John? View PostOriginally posted by drag View Post...
i even have read original mesa-dev discussion before this news was posted and have opened this forum page before any comments were posted but just could not say it good enough.
now i will put this post into my citation collection where it earned its place.
Originally posted by smitty3268 View PostNo one has publicly come out against merging the hardware support, but apparently it sounds like there was a lot of negative reactions to it in private. I think if anyone merged it, they'd get yelled at and it would just come right back out.
So we're basically where we were when this whole thing started. Nowhere. And with no one even publicly saying why.
in my understanding, even if some "patents"(i would call them 'papers with offer you can't refuse') are grossly "infringed", "patent-holder" is not going to scaremongering and threat free&open voluntary community project. that is some black PR. if he's going to get someone, he's going for the users (remember how MS fucked bunch of small companies with imaginary "Linux patents" ?). and it's users who should decide _and bother at all_ to enable or disable that functionality.
and all that _only if_ threat is legitimate _and_ patent-holder is really an extortionating bastard.
MS example actually says even more - you _can fuck people with patents even if you don't have them_ or they highly dubious ("dubious", if we all imagine that we live in a world where patent system is something more than lawful racketeering system). so - you can be fucked either way, so better that you make it worthwhile.
most interesting part of "discussion" was Ian Romanick's:
"If we're going to make any significant changes to Mesa's S3TC support,
we might want to consider changing to a maintained library. For
example libsquish (http://code.google.com/p/libsquish/) might be a good
choice.
One advantage of having S3TC in Mesa would be that we could (finally) do
hardware accelerated compression. There's quite a bit of work published
about, for example, DXT5 compression on GPUs."
it actually says something useful, lays down a plan and puts something more than fucking FUD. apparently there is an up-to-date, maintained, _supported by Nvidia_ DXTn library and we didn't even knew it.
the worst part is Corbin Simpson's:
"Stop messing with S3TC. I think things like my previous attempt to
language-lawyer S3TC compatibility in without any patented code are a
better use of time than silly patches to shoehorn it in to the
detriment of the rest of us."
it's really interesting to know more about those "previous attempt to
language-lawyer S3TC compatibility in without any patented code" and the "detriment of the rest of us".
Comment
Comment