Experienced Forum User, Published Author, Experienced player
(516)
Joined: 6/7/2017
Posts: 35
GJTASer2018 wrote:
chatterbox, have you asked on a Steins;Gate focused forum or Discord server for some insights about the messages you're getting with the Konami code or how the hack creator might've intended the "true ending" to be reached? I'm sure that they would have some insight about these things you don't have right now...
No I haven't. They would be familiar with the messages that w7n quoted and could advise on their meaning, but I'm not sure they could suggest a solving.
Instead, I found these comments on Bilibili, a Chinese video sharing website where w7n and other said:
(Translator used)
Someone: I remember seeing somewhere that w7n said he forgot how he designed the hack.
w7n: Unfortunately I changed the machine code directly, and now I've forgotten what I did. " You're not welcomed" was originally intended to prevent some people from cheating, ...
Many others were also looking forward adding the true ending, but he appears to have stopped work on the hack since long ago. Besides, he hasn't been positive about continuing the developing, despite many people's hopes.
Experienced Forum User, Published Author, Experienced player
(516)
Joined: 6/7/2017
Posts: 35
Darkman425 wrote:
I assume Peach is supposed to represent Mayuri if this is based on Steins;Gate.
Reading the readme.txt file that came with the patch hints at the Konami code having other effects if used more than once. Do you know what that does, exactly?
Repeating the Konami Code gives sentences about Steins;Gate or Chrono Trigger, which I can't quite make sense of. If you input the code 8 times, the world name changes to Hyogo and to 60-x0 instead of α-x. I have no idea what these numbers 60 and x0 mean exactly.
World α-4 opens once you complete the game or enter the code. But it's just a repeat of α-3, and I tried to beat the game once more with this condition and didn't see anything change.
The main difference is Mouser throwing a lot of bombs, Clawgrip throwing a lot of rocks, faster death and bonus chance win, and a softlock fix.
Superfast SMB3 warpless is still broken, entering any ship will softlock in infinite scrolling.
Superfast SMB2J slowly corrupts graphics but otherwise should work.
Zelda1fast is fully playable.
I want to send the updated file to RHDN, but then the only way to sync this submission would be to grab it from his site again, and we can't know how long it remains available. chatterbox, can you check why the new file desyncs and say how hard it would be to resync on it?
I noticed the movie is not synced in the new patch (when using the ladder glitch in 1-1 cave), and I'm not exactly sure what the technical reason is. Trying every possible combination of inputs and using different characters didn't solve the issue, and I found that the same bug occurred in all other ladders, not just the one in the movie that was desynced.
So I made a new run with a revised route for 1-1 for now. Although the record is slower as a result, but I think this is a more truthful record as more accurate patch became available. https://tasvideos.org/UserFiles/Info/638190750274557357
Experienced Forum User, Published Author, Experienced player
(516)
Joined: 6/7/2017
Posts: 35
I recalled that a similar crash was seen in Suerfast Zelda, which was only playable if it was patched in Rev A. The ROM used in this submission was patched in PRG 0. I tried patching it in several other versions, including PRG 1, Rev 0, and Rev 1, but all of them either failed to pass the first door or had the same issue at 5-3. I'm also wondering if he knows the solution.
Experienced Forum User, Published Author, Experienced player
(516)
Joined: 6/7/2017
Posts: 35
feos wrote:
chatterbox wrote:
Glad to see this TAS revisited as the hack was registered on RHDN. I've known an improvement to this TAS for quite some time and it improves 10 frames from 4-2 while preserving most of the input they've done. https://tasvideos.org/UserFiles/Info/638130218040182824
In 4-2, when Mario climbs down from the vine, R+L+D input allows him to resume his move earlier. Inputs in 8-3 and 8-4 have been edited to match the changed RNG, the rest of the inputs are the same as done by the authors.
Nice job! Replaced the file and added your name.
Thank you for your works :) And nice find for Denial. Though I hadn't tried further optimization, but I believed there were potential improvements to be made.
Experienced Forum User, Published Author, Experienced player
(516)
Joined: 6/7/2017
Posts: 35
Glad to see this TAS revisited as the hack was registered on RHDN. I've known an improvement to this TAS for quite some time and it improves 10 frames from 4-2 while preserving most of the input they've done. https://tasvideos.org/UserFiles/Info/638130218040182824
In 4-2, when Mario climbs down from the vine, R+L+D input allows him to resume his move earlier. Inputs in 8-3 and 8-4 have been edited to match the changed RNG, the rest of the inputs are the same as done by the authors.
Experienced Forum User, Published Author, Experienced player
(516)
Joined: 6/7/2017
Posts: 35
half_witted wrote:
I loved the video. But I have to be pedantic here. The category is "full inventory" but if you were to hit pause before getting Zelda, you would see you had 0 bombs (missing from the inventory). Just wondering if there was a way to save 1 bomb at the end.
PS: The lack of rupees doesn't bother me but I guess the case could also be made to save 1 to fulfill the category?
I checked the TAS I exemplified also left 1 bomb and 1 rupee. I'm sorry I have submitted a poorly prepared run. The solution to that is simple, by the way. Beacause I purposely wasted them since I wanted to finish the game with 0 rupees, 0 bombs. Even if I make a new run correcting known improvements, the possibility of leaving more than one of rupees or bombs is sufficient. Thanks for your comments.
Experienced Forum User, Published Author, Experienced player
(516)
Joined: 6/7/2017
Posts: 35
Chanoyu wrote:
I think I can answer those two questions.
1) When screenscrolling to the left, you can pass an obstructed tile. Screenscrolling to the right, this is impossible and you get stuck.
2) In level 8, the dragon boss needs to be killed to open the door.
And I have two questions of my own:
1) In the Money Making Games, you sometimes grab the left rupee. Would manipulating the center rupee to be the reward just take too long in these cases?
2) You can hit an enemy with both sword and wand at the same time, by pressing both buttons at the same time. Why is this not done on the Gleeoks in level 5, 6 and 7?
Thanks for answering on my behalf, and letting me know what I missed (question #2). I believe it won't work and I misunderstood at the first time. As for the 1st question, what you thought is right. Manipulating 50 rupees to the center sometimes caused a big delay than placing it on the left or right.
Experienced Forum User, Published Author, Experienced player
(516)
Joined: 6/7/2017
Posts: 35
HappyLee wrote:
I like the goal, which I've tested myself some years ago. I didn't manage to make this movie mostly because I haven't figured out the best way to solve the "8 stars" requirement.
I have trouble playing the movie file on FCEUX. It starts on 1-1 instead of A-1. The author didn't give any instructions in the submission text, so I don't know what to do.
Play any movie to complete the game 8 times (you still have to wait until you return to the title even if you have reached the ending). That way you need to create a .sav file that has collected 8 stars for your emulator. I don't think you haven't tried this, but I wrote it, just to be sure.
It's nice to see the unique full-fpg of FDS version that doesn't feature in the SNES warpless run. I was especially impressed with the careful on-screen X manipulation on A-2. Great work!
Experienced Forum User, Published Author, Experienced player
(516)
Joined: 6/7/2017
Posts: 35
Darkman425 wrote:
I think there's a Lua script to show the floor plan in the dark rooms, mostly as a visual aid to see the floor without a fire source.
He's right. The game has the concept that you can't grasp what structure a dark room is until you light it up. So I recorded the temp encoding with a lua script so that viewers can easily understand what's going on in that dark rooms.
Experienced Forum User, Published Author, Experienced player
(516)
Joined: 6/7/2017
Posts: 35
There were a lot of unexpected strategies so I enjoyed watching this. The quality of the movements and techniques is also excellent. But apart from these evaluations, I think "ABC" is a bit arbitrary. SM64DS "jumpless" run actually beats the game without jumping, and SMB "walkathon" run actually beats the game without pressing B. It's questionable whether the goal of minimizing the number of button presses that must be used to beat the game is acceptable. The authors' efforts are very valuable and deserve high praise, however, I don't know whether it meets the conditions for publication. I abstained.
Experienced Forum User, Published Author, Experienced player
(516)
Joined: 6/7/2017
Posts: 35
HappyLee wrote:
zdoroviy_antony wrote:
Can somebody confirm is it new WR or what? Down press 1 frame earlier in walljump room, but idk what it is.
https://www.youtube.com/watch?v=3sJ6oexgfZQ
Edit: Looks like cheating and I feel like it already happened before
I'd assume it's simply cheating. He used almost all my inputs, and probably changed Mario's speed after the wall jump.
Yes, an experienced SMB players know well no improvement can be found on 8-4. I think it is fake, either.
Experienced Forum User, Published Author, Experienced player
(516)
Joined: 6/7/2017
Posts: 35
feos wrote:
chatterbox wrote:
I remember the actual rerecord count is around 9,000.
I'll need to update the movie file with the one with manual rerecord count. Should I use 9000 or unknown?
Sorry for the delay in my replying. I've been finding for the original movie file for about week but I couldn't find it. I want this movie to be published with that rerecord count maintained (yes, 19890727 is the date when Mother was released in Japan), but if this is an unacceptable request, please correct it as "unknown".
Experienced Forum User, Published Author, Experienced player
(516)
Joined: 6/7/2017
Posts: 35
jlun2 wrote:
How does the RNG work in this game? Is it only incremented by delay?
Nitrodon had explained how the RNG was worked and used in the game as follows, but I can't quite understand it. I will need to find out more about how the memory addresses relates to the actual RNG.
Nitrodon wrote:
Technically it's not a counter, but an RNG which is only changed when it is needed, or if you are in a menu, or if some text is waiting for you to press A. The same RNG is used to choose which enemy you fight, what the enemy does, the result of the action, and so forth. The RNG seed is located at 0026 (2 bytes), and each use of the RNG multiplies it by 0x0101 and adds 0x6375.
jlun2 wrote:
Nice. Will you be attempting to obsolete the other run in the future?
I had planned that. I found this video (https://youtu.be/bFOYWaVFZt8) while researching undiscovered glitch on NES EarthBound Zero. The video demonstrates the use of the Bread crumbs glitch, which was generally not known to be available in NES EarthBound Zero or GBA Mother 1+2. It was pretty shocking, but I immediately noticed the glitch was not working by corruption to the Bread crumbs because he had performed that by squeezing through certain trees (00:33 and 00:36 of the video) and breaking out of bounds. Naturally we cannot squeeze the trees in like that on the NES version because the character's movements are not smooth enough to squeeze through the narrow gap.
Perhaps I won't try to obsolete Nitrodon's run until I find a definite improvements. Or maybe "Uses a game restart sequence" could again serve as a notable improvement...
feos wrote:
Is it known how much of the rerecord count is manual/non-botted?
I remember the actual rerecord count is around 9,000.
Experienced Forum User, Published Author, Experienced player
(516)
Joined: 6/7/2017
Posts: 35
2 Players co-op hacks are always fun, however, I think this is not the case on this hack. Co-op hack should be more special and entertaining than any other hacks, it is true that entertainment is even less than the original game.
But certainly, the optimization is got better than previous ones. I wish you to prepare a better project on this occasion.
Experienced Forum User, Published Author, Experienced player
(516)
Joined: 6/7/2017
Posts: 35
Doomsday31415 wrote:
So this game can probably be beaten in an hour? Crazy.
Is the nature of the crash at Giygas if you don't do the prerequisites known? Obviously it's known that if you do XYZ, the crash doesn't happen, but perhaps there's a faster way?
Since this movie was submitted, two methods have been discovered by pirohiko: overwriting barricades in Onett, cliping walls in Winters.
He even devised a method to overwrite zombies blocking the stairs from Threed to Grapefruit Falls, but the possibility of being realized is inaccurate, and there is no way to defeat Ness' Nightmare without Paula as far as I know.
Perhaps there is no more major skip; we should be satisfied with the possibility that the game can be completed within an hour.
Experienced Forum User, Published Author, Experienced player
(516)
Joined: 6/7/2017
Posts: 35
Kenttsu wrote:
Yeah, the goal feels pretty arbitrary. Avoiding a checkpoint doesn't make the game any different and many of them are unskippable anyway so skipping just some of them seems pointless.
Same as me. I doubted my eyes when I saw the branch "Least Checkpoints". Also, since [2597] Windows VVVVVV by Scepheo in 08:45.87 with Telejumping has obsoleted [2368] Windows VVVVVV by Masterjun in 13:30.25, the "No Telejumping" is will not be accepted. Assuming that this movie is accepted, I think Masterjun's one should have already been published individually first.
