|
#781
|
|||
|
|||
![]() Quote:
Serious question tho: Know any games that use the PREV combiner mode, for getting input from previous pixels, apart from Monster Truck Madness 64 of course? |
#782
|
||||
|
||||
![]()
Just start off using GCC, and follow the directions! I almost purposely wrote this thing to not be compiled on MSVC at one point. If you don't follow my build instructions then of course it won't compile.
hax, is this an HLE question? Because I don't remember or see anything about "prev" in the SETCOMBINE command.
__________________
http://theoatmeal.com/comics/cat_vs_internet |
#783
|
|||
|
|||
![]() Quote:
Because, I was wondering, if shunshit wanted to make "his" plugin faster, why not then go all out on inaccuracy and shed more things for the sake of speed? |
#784
|
||||
|
||||
![]() Quote:
I'm curious about doing game specific hacks in emulators / plugins. I noticed Jabo D3D8 1.6 seems to use game hacks, because if I changed the crc of SSB64 (U), I got these black scanlines in VS mode. If I play (J) version, I always see those black scanlines. If I make (J)'s crc = (U) version's crc, the scanline problem goes away. That's kinda unfortunate that in order to a not see the scanlines, you have to use a fake crc and bypass the crc check in the game (which gives messagebox errors on most emulators, but it still works). So my question is, is there a another good way to identify games, rather than using crc? |
#785
|
||||
|
||||
![]()
Jabo's plugins read from Project64.rdb for game settings, as they are (or were) components of the Project64 package.
The Project64 INI-based section headers identify a game using the 2 32-bit CRCs stored in the 64-byte ROM header, followed by the 8-bit country code. Personally I insisted that games should be recognized based on the 2-character game ID field. ROM hacks or bad dumps do not pose significant changes over the kinds of properties the hacky RDB file is supposed to amend. zilmar told me once over IM that "the rdb file screams of piracy". I think it's just the consequence of amending problems of different types with HLE.
__________________
http://theoatmeal.com/comics/cat_vs_internet |
#786
|
||||
|
||||
![]()
That issue with Jabo 1.6 sure is confusing lol. I already had the J version configured in the rdb.
Even if I'm not able to fix that issue with the rdb, I could always just use a different gfx plugin ![]() ![]() |
#787
|
||||
|
||||
![]()
Like I said if you're bad at APIs about as much as I am, use SDL.
![]() I wouldn't use XAudio2 for sound since it's not portable, but if you only care about a Windows solution then Xaudio probably is your bet.
__________________
http://theoatmeal.com/comics/cat_vs_internet |
#788
|
||||
|
||||
![]()
It's not that I'm bad, but rather have a hard time deciding between OpenGL, D3d, and DDraw. Also time will be an issue since I have too many goals lol. It sucks being busy & indecisive. I need to narrow things down. One interesting thing I've read is, using multiple API's simultaneously.
My primary focus will be performance over portability. Later on, I could always port code to different API's or better yet, have someone else do it ![]() Maybe when I decide to mess with input stuff, I'll use SDL for that. So I've been messing around with Zilmar's audio. It seems that increasing the buffer, adds delay and decreasing the buffer messes up the sound. I did notice an interesting option in Nemu's sound plugin, called prebuffering. I wonder how that works. |
#789
|
||||
|
||||
![]() Quote:
The larger you make the sound buffer, the more audio samples that play straight before the entire buffer has finished executing. The smaller you make the sound buffer, the fewer audio samples playing consecutively until the end of the sound buffer. Surely that what you're observing makes sense from a logical perspective, right? When you make the buffer big, the sound quality is good for large chunks of audio, but with lagging/delay in between the buffers. When you make the buffer small, the sound buffers are more linearly timed, but you hear the lapses in between them more frequently, as a greater number of sound buffers within a given unit of time have just elapsed (or something to that effect). I remember by upgrading to DirectSound9.0 (but with most features basically just DirectSound8 initialized to that version), I had already greatly improved the sound quality of zilmar's basic audio plugin. But I never played around with audio for very long until I got addicted with RSP interpreter development.
__________________
http://theoatmeal.com/comics/cat_vs_internet |
#790
|
||||
|
||||
![]()
Ya even the name makes sense. I'm just wondering how you can buffer ahead of time.
So besides timing, another thing I need to learn how to fix is that annoying cracking sound. |