O

OpenCBM

Settings | Report Duplicate

0

I Use This!

Very Low Activity

Commits : Listings

Analyzed about 16 hours ago. based on code collected 1 day ago.

Commit Message	Contributor	Files Modified	Lines Added	Lines Removed	Code Location	Date
May 17, 2023 — May 17, 2024 Showing page 48 of 51 Search / Filter on:
- WINDOWS/buildone.inc.sh: Allow to specify $CA65_FLAGS externally. - WINDOWS/ddkbuild_local.bat.sample: Added $CA65_FLAGS definition.	Spiro Trikaliotis	More...				over 18 years ago
Added debugging output: The port number for the parallel port is printed, too.	Spiro Trikaliotis	More...				over 18 years ago
- DDKBUILD_LOCAL.BAT contains settings for the CC65 build process, now. - ddkbuild_local.bat.sample added as sample of how DDKBUILD_LOCAL.BAT can be constructed.	Spiro Trikaliotis	More...				over 18 years ago
- New version 0.1.0.35: Synchronized with mnib. - n2g.c: removed >0x2000 from n2g. this was an error from debugging. - mnib.c: When I broke out the drive reset routine (bailout) I was not unloading MNIB code from the drive, causing the drive not to release +after a transfer. Fixed	Spiro Trikaliotis	More...				over 18 years ago
New version 0.1.0.34.	Spiro Trikaliotis	More...				over 18 years ago
- mnib/mnib.c: moves to track density boundaries during adjust_target. seems more accurate - mnib/write.c: * Removed unneccesary align_offset (0x100 + density * 0x100) prewrite "delay". I don't believe it's possible to align tracks by timing the track writes. I've tried all sorts of experiments and the timing results are just random. BurstNibbler uses a delay here of 175390ms if you calculate the number of cycles it wastes and calculate for NTSC/PAL timings, etc. I don't think Burstnibbler aligned tracks very well this way either but I'd like to be proven wrong. I've noticed that the Rapidlok copiers have a special routine that writes a sync and bumpsvery quickly to the next track and writes another, then triggers the writes on those sync marks. We could try this by rewriting the drive code but I don't get good copies of the Rapidlok games this way anyway. It's hit and miss, same as just plain luck. * added logic to handle short tracks so that all data is overwritten when reusing a disk. also extended G64 handling from 40 to 41 tracks- we still cant write 42 on a normal drive * Fixed include/WINDOWS/mnibarch.h msleep() to be Sleep(x/1000) instead of just Sleep(x). msleep is supposed to me microseconds, and 176000 seconds is a long time to wait between tracks. :) - mnib/mnib.c, mnib/mnib.h, mnib/version.h, mnib/write.c: * Added back in the old Burstnibbler-style disk alignment routine. It can be activated with '-b'. It essentially formats each track in a tight loop (175390ms between each, as calculated with the CPU cycles and speed of a 1541 drive) with a short sync. The write routine then waits for this sync mark before writing each track. This does not take into account the capacity() measurement, and I've never had any better luck at writing aligned disks with it, but it's now there again to play with. - mnib/write.c: * fixed align_offset and short track logic again. heh. have to make sure we overwrite all data that was on the track before which was the other reason why it was there. that's what I get for working on this late at night. it should be fixed for all cases now. :) - mnib/gcr.c, mnib/mnib.c, mnib/prot.c, mnib/write.c: * Fixed V-MAX remastering to look for both header sequences. * Fixed SYNC search to look for 9 bits instead of 10. It appears that MNIB misses approximately one bit of every byte of sync, so when the drive brings up the SYNC flag, we might see only 9 bits and miss it. This is why some V-MAX tracks show up as NOSYNC when scanned for 10-bit SYNCS but the drive thinks the track has sync. This will probably fix lots of other protections. :) * Added a pad to make all tracks the exact same length, and extended the amount of data we prewrite to wipe all tracks and align them as best we can. Without this pad, V-MAX games do not load. - mnib/gcr.c, mnib/write.c: * Fixed weakbit/badgcr handling- all tracks are now scanned and fixed by default unless specified. * Moved pad routine into logical loop. - mnib/prot.c: * autogap returns the last 5 bytes of the gap instead of the boundary. this helps with some syncless track protections (pirateslayer) - mnib/gcr.c, mnib/mnib.c: * -D# now selects any arbitrary device ID. - mnib/kernel.c: * adjusted par_read and par_write timing per Spiro's findings - mnib/gcr.c, mnib/pscan.c: * improved compare_sectors() so that more states are recognized. This fixes the problem where 2 non-CBM sectors reported as matched when they both returned empty (unparseable returns empty) with a disk error (no header found, usually).	Spiro Trikaliotis	More...				over 18 years ago
- Increased initial timeout for mnib_par_read(), mnib_par_write() from 20 us to 200 us. This way, mnib works for Curt's machine. This might fix the problem on some Windows machines. - Instrumented mnib for performance evaluation.	Spiro Trikaliotis	More...				over 18 years ago
Use 1830 DDK (2K3 SP1) for compilation of instcbm if using VC++ 6.	Spiro Trikaliotis	More...				over 18 years ago
New version 0.1.0.33.	Spiro Trikaliotis	More...				over 18 years ago
Removed double ";" in main() function which makes this file non-valid C89.	Spiro Trikaliotis	More...				over 18 years ago
Included mnib upstream changes:	Spiro Trikaliotis	More...				over 18 years ago
close_disk(): Make sure we do not try to write the file if there is none. This fixes a problem when d64copy is terminated via Ctrl+C.	Spiro Trikaliotis	More...				over 18 years ago
Make sure allvs (vice build) compiles all mnib tools, but not install.	Spiro Trikaliotis	More...				over 18 years ago
Added archmnib_vice.c into opencbm.dll when compiling the VICE build.	Spiro Trikaliotis	More...				over 18 years ago
Make debian packages compile again (by calling make with -f LINUX/Makefile).	Spiro Trikaliotis	More...				over 18 years ago
Fixed the build problem for WIN32 with -mno-cygwin.	Spiro Trikaliotis	More...				over 18 years ago
New version 0.1.0.32.	Spiro Trikaliotis	More...				over 18 years ago
Removed superfluous CBMT_MNIB_READ_TRACK_IN.	Spiro Trikaliotis	More...				over 18 years ago
Removed 'mode' parameter from cbm_mnib_read_track() and cbm_mnib_write_track().	Spiro Trikaliotis	More...				over 18 years ago
Moved n2d/ into WINBUILD-n2d/, n2g/ into WINBUILD-n2g/, pscan/ into WINBUILD-pscan/, pshift/ into WINBUILD-pshift/. (the directory names were exactly the names the Linux executables had, which did not work).	Spiro Trikaliotis	More...				over 18 years ago
Moved n2d/ into WINBUILD-n2d/, n2g/ into WINBUILD-n2g/, pscan/ into WINBUILD-pscan/, pshift/ into WINBUILD-pshift/. (the directory names were exactly the names the Linux executables had, which did not work).	Spiro Trikaliotis	More...				over 18 years ago
Removed arch.h	Spiro Trikaliotis	More...				over 18 years ago
Handle the difference a.exe / a between Linux and DOS/Windows.	Spiro Trikaliotis	More...				over 18 years ago
- libopencbm.so depended on arch_filesize in the executable. This prevented mnib to build on Linux machines, as well as VICE compile with cbm4linux support. Fixed that. - cbm4linux compiles with mnib support now.	Spiro Trikaliotis	More...				over 18 years ago
Added -mno-cygwin to LD flags for win32 target. Anyway, it does not work (yet).	Spiro Trikaliotis	More...				over 18 years ago
New version 0.1.0.31.	Spiro Trikaliotis	More...				over 18 years ago
Define VERSION is called OPENCBM_VERSION now. Make cbm4linux compile again.	Spiro Trikaliotis	More...				over 18 years ago
Integration of all mnib tools into the opencbm build process.	Spiro Trikaliotis	More...				over 18 years ago
- New files .cvsignore and debian/.cvsignore - New file docs/WINDOWS/install.de.txt with german install instructions.	Spiro Trikaliotis	More...				over 18 years ago
New version 0.1.0.30	Spiro Trikaliotis	More...				over 18 years ago

←
1
2
…
43
44
45
46
47
48
49
50
51
→