But if you hit the nearby saveblock, the stacked event pointers are overwritten. And weird stuff happens -> The saveblock moves. I remember seeing similar weirdness when I tested events back when I added the functionality to my luascript. Random sprites would move in random directions etc. Because the game jumped to a bad ROM pointer.
Now, if this could be done somewhere else in the game..... Or if it can be influenced..... Needs testing. You can hit the saveblock at close to the same time the wide block gets solid for some strange outcomes, e.g.: Screen may fade black without sending you to the save screen
If you hit the saveblock while it is moving, it will stop and the wide block will be alive forever. You can farm free money forever from this block since the timer already reached the end (but the block is not solid). It will turn solid when you leave the room, though.
Edit: Actually, the overwritten events aren't the cause of the bug, it seems. It seems to be a very minor oversight that the event is handled slightly differently. $004C14 turns different values between the "hit saveblock" and "don't hit saveblock" cases at the time the wide block is supposed to get solid. Not sure what this address does but it's related to the bug.
Wait, so is there any actual effect when event pointers get overwritten to something else? That seems like some exploit waiting to happen if so.
For anyone who has/are TASing 3D games (Super Mario 64 type of 3D, not N3DS), how exactly do you optimize movement? Currently, for things like general gameplay, I play through it slowly in real time, then based on the time obtained from that, slowly reduce it with savestates/frame advance. For things like turning, I focus on one turn at a time (rather than focusing if it slows down several turns/corners later).
I'm aware this approach has flaws; for the former, any TAS-only tricks would be mostly infeasible even at slow-motion without frame precision/luck manip, and for turning, something fast for 1 turn may be slower overall. Also things that require constant changing directions mess it up. I want to improve upon this, but I'm having difficulty finding a better approach.
Tl;Dr - How do you folks optimize 3D game movement efficiently?
Cool. I got the RAM address to work in Bizhawk which is good news. I will try in my final submission for this category to match exactly what Dooty did in PCSX to save even more frames. Bearing in mind we did have a small improvement of Abe's Oddysee back in 2014 which was a 21 frame improvement.
http://tasvideos.org/4298S.html
Since you just learned of RAM Watch now, it would probably be wise to find x/y addresses and see if some minor frame savings can be gleamed from the run.
I got a suggestion: Try running the console verified runs on BizHawk and see if they sync. Most of them were done on FCEUX, but if for whatever reason it does not sync maybe something is wrong?
To avoid derailing, from: Post #436223Invariel wrote:
This is a way to expand your playerbase without alienating your old players. Similar techniques are used when designing traditional cross-platform games (PlayStation Something <-> X-Box Number <-> Nintendo Alphabetti-Spaghetti <-> PC), but in the case of developing for the PS3 and the PS4 specifically, the developers on these teams were already developing a game for the PS3 (a systemnotoriouslydifficultto write code for), whereas the PS4 "features an AMD Accelerated Processing Unit (APU) built upon the x86-64 architecture" which is a far more common architecture to work with. I'm not saying that the work required to develop the PS4 version out of the PS3 code is trivial, just that you don't need as specialized a team to do it, particularly if your PS3 and PS4 teams are talking together on a regular basis. Doing cross-platform development like this while the game is still being worked on is factored into budgets, and is a completely different animal than digging up a fossil and adding new code to it.
That made me curious on developing for Cell, but a google search along with wikipedia only conclusively states PS3 as using the Cell microprocessor. What other mainstream devices/consoles/things utilized Cell anyways? How about today? Is it still in use?
Yeah, uh, don't do that. That's clear abuse of the system. One more cancellation like this and you won't be able to submit anymore. Feel free to submit the final run, but warn us in advance and talk about your reasoning long before you choose to cancel it again.
EDIT: To everyone else, if you're immediately voting No on the run just because of the submitter, that's not cool either.
It's 2 no votes and 1 yes vote (at this post); I'm positive chances are people don't like this game/run more than having some conspiracy against him.
I have no idea how it happened, but when I loaded a tasproj file this suddenly appeared:
http://pastebin.com/jKV9gjAb
I cannot reproduce it yet, but I'll post again if it happens. This appeared for the dropbox version Feos linked.
Edit: Feos, was the seeking feature removed? Currently for some reason the latest dropbox link still has "Turbo seek" option, but it does not seem to do anything.
http://imgur.com/kJk179r
Edit2:
Anyone else experiencing this bug with feo's revision? In case no one knows what is going on, right clicking sometimes makes a different marker get selected.
This Movie is made with only turbo tools, re-recording tools and frame step tools. It is meant to show players how the game can be beaten in five minutes.
That jumping glitch sure makes the game more entertaining, and quite funny.
Here's an encode of that WIP:
https://www.youtube.com/watch?v=e_1ptk64uhM
This is only the first mystery, right? If so, I wouldn't mind seeing more levels glitched like this.
Thanks.
And yes, this is only the first stage. There's 5 in total, and forgoing the glitch doubles the length of the first stage. Since the other stages are even larger, it probably increases the time even more.
Edit: So.....while going back to the first stage to optimize 0x11, I found out I forgot about a trick I used earlier in room 0x10 that could be applied to multiple rooms.
Well, a few casuals got something we call "Where's the room Glitch" in one of the rooms in Impact Creator, which appears to be a very random glitch though.
The glitch get's you into a room without the map transitioning, so the game actually thinks you are in a different room.
With that we can get infinite speed on a wall and then use that in some way to skip exo.
( I don't know the exact details for which triggers you want to hit/avoid with IS to make that work, but andy should know the specifics )
If we can replicate that, then we pretty much have exo skip confirmed.
Another possibility would be to just find a secret world (oob) in one of the rooms in Phazon Core, which might be possible with the use of Fission Metroids, where we have ideas, but no positive results yet.
If it works on dolphin, would recording an input file then trying this glitch over and over until it works help with debugging it?
Another marker request: When you delete a frame in TAStudio with a marker also on the same frame, can it please not delete the marker and instead make the next frame be that marker (unless there's nothing left in front)?
This may be a bit odd, but is there any online MMOs with "Quests" similar to "RuneScape"? Example quest from said game. I tried searching for non-RuneScape, and it gives "amazing" quests like kill x, escort y, fetch z, etc. Is RuneScape the only one that isn't...shit?
I found a way that you can theoretically complete the game without really doing any of the lessons and just doing one single minigame about 200 times after the placement test (each time would take an inhumanely short amount of time, done within seconds). Does that count as "nontrivial routing"?
Can you please try and make a rough run of this route vs the intended route and record it? It's rather hard to judge when this is not exactly a mainstream game people are likely more familiar with.
That's because this isn't the optimal solution; iirc real time runs abuse a glitch that allows pretty much all battles to instant end. Early pokemon games are weird like that.
But what about games like Zelda 2, Kirby's Adventure or SoTN? Those games have that category going on. The "no memory corruption" branch looks weird in comparison to those other games.
GL with that; the GB pokemon games are far too broken for a meaningful somewhat glitch-less-ish sorta category.
http://pastebin.com/jKV9gjAb
I cannot reproduce it yet, but I'll post again if it happens. This appeared for the dropbox version Feos linked.
Edit: Feos, was the seeking feature removed? Currently for some reason the latest dropbox link still has "Turbo seek" option, but it does not seem to do anything.
http://imgur.com/kJk179r
Edit2:
Anyone else experiencing this bug with feo's revision? In case no one knows what is going on, right clicking sometimes makes a different marker get selected.
Using BizHawk 1.11.6 to make a project, then dragging it to that version have this.