Deathlike2 said:
Thanks for the links! I guess I have a problem with Derek French's explanation for two reasons:
(1) Although it's no longer an issue, I still can't forget the incorrect and conflicting info that came out of Bioware back when the NWN and shiny water issue first came up. The very first thing Bioware did IIRC is to flatly state that the effect was not possible on R300. Later, that was moderated to something closer to the truth: that it wasn't supported on R300 because ATi's drivers didn't support the nV_ extensions the engine required for the effect (duh, naturally.) And then finally the issue was resolved via a Bioware patch, along with an ATi driver update, which presumably did not require the ATi driver to make use of any nV_ extensions to render the effect, because Bioware had included vendor-neutral code to support the effect in the patch (Which is claimed for KoTOR at present in regard to soft shadows support.) So all of that damages BioWare's credibility for me on the present issue.
(2)
More importantly, though, what French states in the Bioware thread you've linked above just doesn't make any sense to me...
OK, so if "soft shadows" actually works on ATi hardware, albeit very slowly, why hardcode the game engine to disable the feature completely when it detects Radeon hardware? It makes a lot more sense to simply ship the game so that the feature is automatically disabled upon the installation routine detecting a Radeon, so that people can enable it voluntarily if they choose. Upon seeing that the game runs extremely slow--unplayable, really--when "soft shadows" is turned on with a Radeon--what, pray tell, does French think people might do OTHER than immediately read the "troubleshooting" section which he's said "he's afraid" nobody would read? Well, if he doesn't think anyone would read it, why did BioWare include such a *long* "troublehooting" section in the first place? Just doesn't make much sense, I'm afraid.
It gets even more interesting, because in the long "troubleshooting" section BioWare ships with the game, which contains a lot of card-specific, vendor-specific information I guess French has no fear of people "not reading," there's *not one word* printed there which states what French alleges--not one word at all about the code being "vendor-neutral" and that soft shadows "runs slowly because of a bug in ATi's drivers"--not one word at all. Additionally, there's not a single word to the effect that, "Although with soft shadows enabled the game runs very slowly with the current Catalysts, ATi has assured us that a near-term driver update will solve the problem, so check your Catalyst driver updates for references to a fix for this game, which should correct the problem without the need for a specific BioWare game patch at the same time."
Not one single word to that effect--nothing beyond BioWare's statement that "Soft shadows is not supported on these [ATi] chipsets." Evidently, French had no "fear" of people "not reading" *that*, did he?...
So I guess in his mind it was better to misrepresent the situation in the Troubleshooting read.me he "feared" no one would read, and have to write numerous posts on the BioWare web site explaining what the read.me "should have said," instead of just putting it all into the read.me in the first place. And instead of fibbing about their being "no support" in the game itself for "soft shadows" for Radeons. Oh, sure. It always saves time to fib to your customers in troubleshooting files, and to omit pertinent information from them, especially in "troubleshooting" files you are "afraid no one will read." Right....Good idea, to include such files with games, when you think no one will read them--makes a lot of sense. IMO, Derek's got to explain why it was only this information, out of all of the other information printed in the troubleshooting file, that he was "afraid" that "nobody would read." Pretty tall order, I think.
Next, there's specific info in the troubleshooter about what hardware features a 3d card has to have to support FBE and soft shadows, and about the necessity of hardware PS/VS support. I guess he had no fear of people with GF2's not reading that. Then there's the part about how soft shadows won't run very well on a slow/older 3d card (GF2, anyone?)--again, he had no "fear" that people in that category would fail to read the troubleshooting file. It was just the
Radeon owners he feared would not read the troubleshooting file if they had trouble, which of course only makes sense to French...
It's ridiculous, frankly, and insulting, for someone to state he left out pertinent info that belonged in a troubleshooting file, and materially misrepresented something in the read.me, simply because he was "afraid" that people wouldn't read the troubleshooting file. If that was the case then there'd be no troubleshooting file at all for the game, would there? I mean, why bother, if no one's going to read it?
The clincher for me is this: The game engine disables the feature apparently if either specific cards or specific OpenGL extensions are not present and recognized by the game when it boots (don't know which.) Even if I manually edit the KoTOR.ini and turn on "soft shadows," the game engine is coded to turn it right back off and disable the feature when it boots. Of course, this means that contrary to French's statements in the BioWare thread, not only will the "fix" require a new driver from ATi, but *it will also require a new KoTOR patch from Bioware at the same time*. It won't do me much good for ATi to fix the driver bug if KoTOR keeps disabling the feature when it boots, will it???? So BW will have to do a patch just to allow me to turn the feature on at all, bug-fixed driver or no.
So, in conclusion, while French states that "soft shadows" actually does work with Radeons, just very slowly (a fact which if true is directly contradicted by the KotOR "troubleshooting" file which ships with the game which states that "soft shadows" is simply not supported by Radeon hardware at all in the game), French further alleges that all of the OpenGL support in the game is "vendor neutral" and that when ATi fixes the Catalyst bug causing the problem the feature will work without any need for a corresponding patch from BioWare. Great, except for the fact that BioWare has disabled the feature in such a way that a BioWare patch will be *required* when ATi fixes the Catalyst bug Bioware blames for the situation, in order for people with Radeons to enable the feature at all. Wow--good thinking--for a "vendor-neutral" software problem that doesn't require patching from BioWare to fix--good thinking from BioWare to require a BioWare patch *anyway.* Yea, man, how sharp can you get?
Heh...
I wasn't really hot about this until I read the BioWare thread, and French's remarks in particular, which I found drenched and dripping in BS, frankly, and presented from a high and mighty position I particularly disliked. I don't have a problem with people who admit honest shortcomings or being sloppy or being rushed, etc.--who *tell the truth* about such things. I just can't stand it when somebody thinks they can spew this kind of nonsense and expect people to believe it and accept it. So many things could be avoided if people simply learned how to tell the truth. Either way, according to French, BioWare has lied about this. The company either lied in the read.me when it said the feature wasn't supported by Radeons in the game, or French is lying when he says that was never true, that Radeons always supported the feature--just too slowly to suit Bioware. French's excuse for lying in the troubleshooting file was the flimsiest imaginable--his "fear" that nobody would read the troubleshooting file. Oh, man, you can't get any more pathetic than that.
About the only thing we do know for certain is that no matter what it's going to take another Bioware patch to open up this feature support for Radeon users. Be smart, French, do it now and let us see for ourselves how effective a Catalyst bug fix is in resolving this absolutely needless, ridiculous, and unnecessary mess.
