Editor, Experienced Forum User, Published Author, Active player
(296)
Joined: 3/8/2004
Posts: 7469
Location: Arzareth
Just a tip: Watching the original video series (which I did, entirely), to see how this game was intended to be played ( http://www.nicozon.net/watch/sm19220627 ) makes one appreciate this TAS so much more. Especially the stage with shell antics, which already was hilarious on its own.
EDIT: I revise the above statement. Watching the original video series made me appreciate this hack so much more. It is actually quite genius design, with lots of intelligent puzzles, and not just a death trap hell. But after watching the original video series, this TAS felt quite a let-down due to all the Yoshi shortcuts. In my revised opinion, this TAS is better enjoyed without watching the original video series.
Editor, Experienced Forum User, Published Author, Active player
(296)
Joined: 3/8/2004
Posts: 7469
Location: Arzareth
FractalFusion wrote:
which is an abbreviation of 自作の改造マリオを友人にプレイさせる (Making my friend play through my own Mario hack)
Which can be romanized as: Jisaku no kaizōmario o yūjin ni purei saseru. The "mario hack" part in that sentence (kaizōmario, meaning lit. remodeled Mario) is where the name "kaizo" / "kaizou" came from.
Editor, Experienced Forum User, Published Author, Active player
(296)
Joined: 3/8/2004
Posts: 7469
Location: Arzareth
I tried a bit at level 14 (computer gets 9 handicap). I have a feeling that it is possible, but my first few tries led into a failure. There seem to be certain actions that lead the CPU into spending a lot of moves accomplishing practically nothing, which helps a lot, but just not enough.
Editor, Experienced Forum User, Published Author, Active player
(296)
Joined: 3/8/2004
Posts: 7469
Location: Arzareth
SmashManiac wrote:
It says that the CPU goes first at level 7+, but you're obviously the one going first.
The handicap (black stones already present on the board) are counted as black going first. Black gets to play a few stones first, and only then white gets to reply.
Re: What the game looked like and on the level selection, I have similar thoughts.
Editor, Experienced Forum User, Published Author, Active player
(296)
Joined: 3/8/2004
Posts: 7469
Location: Arzareth
I just finished watching this, and I was about to post exactly the same as Ace of Hearts posted above:
Ace Of Hearts wrote:
I'd also like to note that I appreciate your taking a bit right before the final battle to go into your menu specifically to prove you collected your entire quota. For a TAS of this length, it is, in my opinion, aesthetically pleasing, and worth it.
Repeatedly stuffing the characters (especially Robo) full of pills was funny and disturbing. I had no idea there's such an abundance of those tabs in the game, when you steal them from bosses!
I was somewhat disappointed to find out that you discarded equipment when exiting Blackbird. I was worried that your "100% completion will not have all equipment in the end after all. I guess that was a needless worry.
Also surprised to find out that the question of Prism equipment becomes moot from 100% completion perspective when you can steal the other type of item from Zeal.
Why did you not complete Black Omen three times, though? Is it not possible to get tabs from each one of them? Doesn't not doing that actually undermine the "collecting all Power/Magic/Speed Tabs" goal?
I'm pleased to say this TAS is everything one could hope from a 100% completion of the game. It is entertaining to watch, despite its length. I did fastforward some scenes, but I did watch it to the end, which is something I can't say even for many short TASes.
Editor, Experienced Forum User, Published Author, Active player
(296)
Joined: 3/8/2004
Posts: 7469
Location: Arzareth
FODA wrote:
Here's something I did on the 3ds:
Boy that's a fluffy tree. I also like the little HDRI effect you have going on near the tree, even though it might have been unintentional. Reminds me of Thomas Kinkade.
Editor, Experienced Forum User, Published Author, Active player
(296)
Joined: 3/8/2004
Posts: 7469
Location: Arzareth
I also agree it is quite entertaining to watch in real time. The text being completely readable is one factor to this.
I share the worry about apparently arbitrary decisions on which treasures are collected and which are left alone, but the explanation posted before (some expensive duplicates are collected for money to buy some rare items later) is good.
What was done about the prism stuff that you can only choose one of?
Editor, Experienced Forum User, Published Author, Active player
(296)
Joined: 3/8/2004
Posts: 7469
Location: Arzareth
Dwedit wrote:
Does open bus really work that way? Some emulators just read back the low byte of the address as the value.
The CPU bus just returns whatever was last placed there, if no device sets it into a particular value. If you read the PPU, you will not get the open value from CPU's bus, but an open value from the PPU's bus, which has whatever was last written to/read from the PPU's registers (where applicable).
What happens in the posted example (Battletoads, NES) is:
- The opcode was an absolute indirect jump, which causes a read from RAM, and the read value is $75BD, which puts $75 in the bus.
- Opcode is read from $75BD. Nothing is mapped there, so bus value $75 is returned.
- $75 is interpreted as opcode (75 75, "ADC $75,x"). Equation $75+x produces $75; RAM address $75 is read, and value $6F is returned. Bus now has value $6F.
- Opcode is read from $75BF. Nothing is mapped from there, so bus value $6F is returned.
- Value $6F is interpreted as opcode (6F 6F 6F, "RRA $6F6F"). This causes a read from $6F6F, which produces the value from open bus, $6F because nothing is mapped at $6F6F. (It also writes at $6F6F, but I am not sure why the written value $37, or $38, is not placed on the bus. 38 is SEC, by the way. Maybe this emulator does not properly emulate the RRA opcode. If it did, the code would still work but the timing would be different, because SEC takes 2 cycles per byte whereas RRA takes 5 cycles per 3 bytes. 37 37 would be "RLA $37,x", which would read $37 and also write back.) Stable loop ensues, repeated until the end of undefined region.
Editor, Experienced Forum User, Published Author, Active player
(296)
Joined: 3/8/2004
Posts: 7469
Location: Arzareth
Warp wrote:
I find it curious that the Gameboy would execute anything from RAM at all. I would assume that the most rational design for a handheld console (or any console where the program code is not read into memory but executred directly from ROM cartridges) would be for the game ROM and the RAM being in completely separate address spaces.
It is not really that surprising. Examples of other well-known platforms where there is no difference between RAM and ROM from the viewpoint of the CPU's instructions and addressing modes*:
- NES
- SNES
- 8086 PC (and all of its descendants, including modern x86_64 PCs)
- Commodore 64
You can even execute memory mapped I/O if you are so inclined. I have a test for NES which does exactly that: Ensures that a combination of I/O ports will return predictable values, and then transfers control to those addresses. The CPU will execute whatever opcodes corresponding to the values read from those I/O ports.
On the PC you might execute code from the CGA/VGA video memory. (I believe this has been done in some demos.)
*) Sure, on the NES there are addressing schemes which can only access a particular 256-byte region of the RAM, and the stack can only be in a very specific location in the RAM, but it doesn't mean you are limited to using those schemes for addressing the RAM, nor that all access is done through those schemes.
Editor, Experienced Forum User, Published Author, Active player
(296)
Joined: 3/8/2004
Posts: 7469
Location: Arzareth
Dacicus wrote:
Bisqwit wrote:
Dacicus = /dɑkikus/
The i and u are supposed to be short, which would make the pronunciation more like /dɑkɪkʊs/ according to the Latin textbooks I've seen. That might be an artifact of the multiple sounds possible for the same letter in English, however.
In English there are indeed several different "i" phonemes (at least /ɪ/ and /iː/). In Finnish, though there are short and long vowels separately, they are still the same phonemes, distinguished in IPA by adding the colon /ː/ character after it. I was transcribing Finnish pronounciation, and there short and long i are indeed /i/ and /iː/ respectively. One of those tiny things that make Finnish accent in speaking English noticeable. Our pronounciation for the English words "bit" and "beet" truly only differ in the vowel's length, unless practiced hard. It is even clearer in words like "could" and "boot" (/kʊd/, /buːt/), where both get pronounced with /u/, and "bed" and "take" (/bɛd/, /tʰeɪk/) where both get /e/ and no /ʰ/ will occur; "bat", "father" (/bʌt/ or /bat/, /fɑːðə/) where both get /ɑ/ and the /ð/ will probably turn into /θ/, /t/ or /d/ and of course the /ə/ usually turns into /ø/.
creaothceann = /kɹiːɔtʃan/ probably.
