The 03.09.12 at 23:20, Jan Elders wrote:
* cancel sax2 * goto init 5 (so, with the nvidia-XF86Config) just to try -> black screen
Instead of that, just do "startx" from the terminal (better as user, not root). Errors get printed on the console, and it's easier to kill. And, "startx fwmn" or similar is faster than kde.
NOW HERE IS A STRANGE THING : When (in init 5 mode) I start sax2 from within a terminal, then I get a normal - good - sax screen. Then I cancel sax2, go down to init 3, and start sax2 in command mode, and then I get the fuzzy screen again I mentioned before.
I think that is related to this text (sax2 --help) that I posted here the other day, and that I said I do not fully understand : [ -s | --sysconfig ] this option tell SaX2 to import the system wide config file even if SaX2 was started from a textconsole which normaly will import the SaX2 HW detection data See? It seems that sax2 uses different data if started from a console or inside X.
BTW, someone told me that with TFT screens (like mine) there is no vsync/hsync. He said there is only a 'reaction time'. Is that not true ?
That must be true. On a glass CRT, there is an electron beam that is swept left to right, up to down, so as to "illuminate" every dot (pixel) on the screen. Parameters such as vsync/hsync, frequencies, times, and such relate to the times needed to "move" that electron beam, and how to synchronize it to the signals (ie, to know where the beam is anytime). They have a very physical meaning. A TFT screen is a type of a matrix display, with electronics activating whatever they activate (leds, LC orientation, whatever), in some kind of matrix, indexing or whatever (I lost track of the actual technologies long ago). It could be better used with a bus, instead of the VGA connector, and a specific video card tailored to that type of addressing (perhaps with no memory). To use the VGA connector they must use intermediate electronics that "read" the signals on the VGA cable and "convert" it to the matrix needed to activate the dots on the display. This is rough sketch of an explanation, and probably incorrect on some places and details, but it will suffice. The thing is that the frequencies supported are more or less arbitrary, probably not really very much related to the real refresh rate of the screen, and thus limited, much more limited that those available on a glass CRT. The more frequencies available, more money.
Anyway, I could not find any information about vsync/hsync in the documentation of my notebook.
Makes sense. Probably has a fixed frequency(s), supplied as VESA info. For example, mine (a cheap Proview monitor) is running at H:60KHz/V:75Hz, 1024*768 (15"), and absolutely no flicker whatsoever, even at lower frequencies. A CRT at this frequency would tire my eyesight.
Also, modelines is a new phenomenon to me. I'm afraid I'll have some studying to do before I know how to manipulate this.
I'm uncertain of the effect they would have there. Anyway, on mine I don't bother: it has an "auto" button that automatically adjusts everything fast. When I use my bigger and "better" CRT I sorely miss it :-) -- Cheers, Carlos Robinson