Posts: 4,307
Threads: 63
Joined: May 2003
Blitz I am not even suggesting you to create patches for any video card. All I am saying is you make a better program which can really test UGL's compatibility on a video-card. I mean even if the card identifies itself as a VESA 1.2 compatible it may not have implemented all the features which leads to UGL crashing. This crash is mostly due to the card not supporting scanline reprogramming. So if you could make a function or include a test program which would check for this specific feature, not using VBE functions, by actually trying to reprogram the scanlines. If it doesnt work, the program will crash otherwise it will continue and write the "Sucess" result in a file which UGL will read while initializing. Thus it will allow other programs, which use UGL, to exit gracefully instead of crashing.
v3cz0r, too bad you decided to take Linux as an example. All the h/w that you have is supported. You just need to know *how* to. I too had issues with my HP printer, soundcard but I learnt that you shouldnt blame the manufacturers or the Linux developers for it. I just had to install the proper s/w and all my devices worked fine.
One of the things I admire about the Linux developers is that they always provide walkarounds which M$ doesnt. Hence I dislike M$.
I think you should too try and provide some strong checking like one i described above.
Posts: 788
Threads: 53
Joined: Nov 2002
That's the whole point man, radeons crash upon setting a vesa mode or changing scanline width. It doesn't return anything, it just crashes. There's nothing you can do, try it yourself. Voodoo's do not crash, they just return an error code and so does ugl on voodoos.
oship me and i will give you lots of guurrls and beeea
Posts: 3,368
Threads: 195
Joined: Jan 2003
Quote:I dunno, i never had a radeon, but i know that in xp the whole screen gets messed up after entering real mode vesa and going back to the desktop. Anyways, try that jpeg viewer antoni posted a few posts back. Does it work ?
Happens with an ATI: RAGE PRO TURBO AGP 2X, too. (win98)
Messes up entirely when trying to do it uncompiled, messes up paging when compiled.
Peace cannot be obtained without war. Why? If there is already peace, it is unnecessary for war. If there is no peace, there is already war."
Visit www.neobasic.net to see rubbish in all its finest.
Posts: 3,279
Threads: 170
Joined: Nov 2003
The problem is the video card manufacturers. Everyone who says differently is oblivious to this fact. Stop arguing over this moot point, weirdos. 
I'd knock on wood, but my desk is particle board.
Posts: 2,020
Threads: 24
Joined: Jun 2002
Interesting general note on the matter:
I downoaded the latest drivers for my Radeon, and it non longer garbles up the screen after I try and enter an unsupported mode, but it now actually enters the mode and says "far heap corrupt" in green letters. Upon quitting via alt+enter and clicking 'x' on the dos box, the screen looks perfectly fine.
Maybe the situation will settle itself for ati users...
i]"I know what you're thinking. Did he fire six shots or only five? Well, to tell you the truth, in all this excitement, I've kinda lost track myself. But being as this is a .44 Magnum ... you've got to ask yourself one question: 'Do I feel lucky?' Well, do ya punk?"[/i] - Dirty Harry
Posts: 3,368
Threads: 195
Joined: Jan 2003
No, it won't.
I downloaded my newest ATI driver recently.
Peace cannot be obtained without war. Why? If there is already peace, it is unnecessary for war. If there is no peace, there is already war."
Visit www.neobasic.net to see rubbish in all its finest.
Posts: 451
Threads: 16
Joined: Feb 2003
Quote:Oh shut the f*ck up! It's developers like you that make buggy software. If you know that there is a issue with your software working on some hardware then you must take precautions and not blame the @£^$ card makers.
And wtf will the vendors do about it? It's not the vendors but the manufacturers you want to talk to.
You're a nice example of this saying: "A bad worker blames his tools"
I am quite sure you are or will work at Microsoft.
Actually, quite the opposite. I loathe bugs. I rarely get programs released on time because I make sure v1.0 has *no* bugs in it. It's idiots who say, "Well, the vendor/manufacturors didn't do it right so I have to take up the role of fixing thier slack-off job and not get my actual project done." That need to have thier head examined. If I follow all the rules and make sure my code is 100%, why the fuck should I go and then start writing "special case" code to handle every bug in every piece of hardware?
"hrm, Pentium fdiv bug, let's handle that; hrm, VESA inconsitency, let's handle that; hrm, 486SX cache timing error, let's handle that; hrm, can't auto-detect SB 1.x settings, let's handle that...(list goes on)."
Or, you can make sure your code is correct then submit a bug report to the people who's fault it is that there is a bug.
And actually, ATI is a vendor/manufactoror, but, they also just sell chips and driver templates to vendors who make thier own products (like my video card). So, the vendors *can* do something about it as they have the source to the driver/BIOS. They just *choose* not to since it's not worth thier time to fix such obscure bugs.
[/end rant]
Life is like a box of chocolates', hrm, WTF, no it isn't, more like, 'life is like a steaming pile of horse crap.'
Posts: 1,272
Threads: 36
Joined: Feb 2003
TBBQ:
The qb community loosely resembles the open source community - we help and teach each other, and freely exchange source code. We do it because it's a rewarding experience. Blitz and/or v1ctor wouldn't have written UGL for everyone to use if they hadn't derived some pleasure from doing so.
Now how would you like it if somebody kept arguing with you about one of your projects? You're not going to get any favors from Blitz insulting him like that. Blitz has no obligation to you, so just give him a break, ok? Please?
They haven't explicitly put UGL's source code under any sort of GPL-like license, so you'd need their permission first, but maybe you could try what you propose yourself or convince somebody else to (by being polite and pursuasive, and not insulting them).
Posts: 4,307
Threads: 63
Joined: May 2003
Whatever...let UGL be as incompatible as before =(. I was just suggesting so as to improve UGL. But it seems no-ones listening so might as well drop it.
Posts: 788
Threads: 53
Joined: Nov 2002
Sure, i'm listening, but are you? What is your suggestion? Not change scanline width? That's impossible, it would require the lib being completly rewritten from scratch and the product would be something slow and big as said before. Neither v1ctor or me has the time nor will to do that. And that still won't help the problem of radeons not working at all, even flib doesn't work with radeons on nt/2k/xp. Dunno about 9x. If you are able to fix the problem in someother way wich would only require me to add a small patch i will add it for sure at once.
Btw Antoni, your jpeg viewer didn't work on voodoo or radeon. Eric tried it on his.
oship me and i will give you lots of guurrls and beeea
|
