summaryrefslogtreecommitdiffstatshomepage
diff options
context:
space:
mode:
author Dirk Best <mail@dirk-best.de>2019-06-12 13:47:27 +0200
committer Dirk Best <mail@dirk-best.de>2019-06-12 13:47:37 +0200
commit9bb0c483bfa8ef12a38e526c9380686933c72ad4 (patch)
tree6766b8557d808e120e41733ae959d5b7ba4285a7
parentd8110b3499aaa9b4dc2e841665989702e7d2d05f (diff)
argus: Add some info about the ROM checksum tests
-rw-r--r--src/mame/drivers/argus.cpp15
1 files changed, 15 insertions, 0 deletions
diff --git a/src/mame/drivers/argus.cpp b/src/mame/drivers/argus.cpp
index f9d288b02a7..abd8d0df62c 100644
--- a/src/mame/drivers/argus.cpp
+++ b/src/mame/drivers/argus.cpp
@@ -105,6 +105,21 @@ Note :
argus
valtric
+- Information about the internal ROM tests (see also MT03219):
+ * argus: Checksum routine at $7fc9 (for banks at $7fc0). Checksum is a
+ simple sum of the contents. Our dump gives a result of 0x95 while the
+ game expects 0x9b, therefore it displays a checksum error. Checksums for
+ the banked ROMs match.
+ * valtric: Checksum routine at $987c (for banks at $f000). Checksum is a
+ XOR over the contents. The expected checksums are stored in ROM vt_06.bin
+ starting at $d000 (main ROM first, then banks). For our dump, the
+ expected checksums are all 0x00, but the calculated checksums differ,
+ therefore displays a checksum error for all ROMs. This has been validated
+ on real hardware to also fail there.
+ * butasan: Checksum routine is at $e0a8 (for banks at $ec74). Checksum is a
+ simple sum over the contents. The test seems to be broken (or hacked) as
+ it will only fail when the checksum is exactly 0x00.
+
Known issues :
===============