Where would that be used? The x86 FPU has built-in commands to load seven constants. They are 0.0, 1.0, pi, ln(2), log2(e), log(2), and log2(10). The last four are used because the FPU can calculate only log2(x) and 2^x directly; multiplying with the right constant before/after the log2/2^x …

HunterZ wrote: Is there a way to hook it up so that the host OS cursor is always shown whenever the emulated one is in a hidden/invisible state? That could be confusing in fullscreen mode. it's a good idea, but IMHO only in windowed mode.

Qbix wrote: Think again. I didn't mean to step on anybody's toes. Sorry if I offended you or some other developer. Still, the font in there is so similar to the font on my card - more than 99% identical. How do you do that when designing it from scratch?

How erratic? I tried MS mouse.com version 9.01. The mouse pointer keeps jumping all over the place, and the mouse buttons usually don't respond at all. Does dosbox emulate mouse hardware at all, or only the (int 33 I believe) mouse driver interface? Also try another driver... Will try when I find …

Another patch, really too anal to put it on the patches pages, thus here. It blows up the math constants in fpu_types.h to 50 significant digits. It's overkill, I know, but these should work even on systems which do FP math in 128 bits accuracy.

A related question, is it normal that the mouse doesn't work when booting to a real copy of DOS through imgmount and boot? I tried adding in a real DOS mouse driver, result: now the mouse "works" but in a really erratic way.

Hello all, I uploaded two small patches to the sourceforge patches page of dosbox. One is a slightly improved font, taken from a real (Matrox G400) card. It's a true 8x16 (and 8x14) font, ideal for dosbox since dosbox doesn't do 9 pixel columns per character (720x400) in text mode, instead it does …

Tearex, Qbix said here http://www.vogons.org/viewtopic.php?p=75938 that one of the CVS changes as per jan. 30th, is because of functionality like the one Kronuz has implemented, so Kronuz' patch shouldn't be necessary anymore. Ok, great, I can't view the above posting (for some readon "Topic does …

One more thing, I noticed problems with the 40 character modes (say MODE CO40 or MODE BW40 in DOS) with this patch. The display becomes unresponsive and largely stuck, sometimes it changes a little bit when scrolling. Though I also added the SVGA chipset patch, maybe that one might be the culprit as …

Hello Kronuz, I'd like to notify you that the big changes in CVS as of 2006-01-30 make it very difficult to apply this great (I'd even say vital) patch. 29 Hunks fail. If you, Kronuz, have the time for it, would you consider updating it another time? It's too bad it isn't already in CVS. Thanks a …

Say xms=false and ems=false in your dosbox.conf. 6.22 can't install its XMS and EMS drivers if the memory isn't pristine and unoccupied by any other thing, such as the DOSBOX built-in drivers.

Thank you for your appreciation, I've put the patch in the sourceforge patches area as well.

See also the "date" thread I have just started. The MAKE_RETURN macro *must* be modified (macro parameter must be put in parantheses in macro body, really always a god idea :-)) for this to work! The weekday function should be corrected too, localtime uses 0-6, but RTC chip uses 1-7. And year should …

Hello this is a little patch against current CVS to get the date functions of the RTC and BIOS working. It's very annoying if you boot into a real DOS and the date is always January 4, 1980 because those functions aren't implemented. So here goes: