Experienced Forum User, Player, Published Author
(99)
Joined: 10/19/2009
Posts: 2527
Location: Italy
The thing is that it’s not unnecessary… it’s supposed to be required for the addresses to even work. I’m not sure how it’s working for you without pointers, but when I tried, it worked for 2 restarts and after the 3rd it didn’t, so I just started using pointers at that point. It doesn’t really require math or a calculator either, you just go from right to left and do simple 1-digit additions. You just need to know that hexadecimal goes from 0 to F.
RAM Search/Watch and memory viewing are usually (always?) separate options. Not sure how many emulators have both though… BizHawk and DeSmuME do.
I thought it could do that? I’m not sure if it’s on access but it can do breakpoints.
Experienced Forum User, Player, Published Author
(99)
Joined: 10/19/2009
Posts: 2527
Location: Italy
^To be honest, if there are only 3 instances of RNG, an RTA would just reset a few hundred times until it gets the perfect RNG, and then it’d be pretty similar to the TAS I take it? Basically in a TAS you have savestates, in RTA you have retries. Sure it could take a long time but people have speedrun games for thousands of attempts and resetting for RNG isn’t exactly anything new (hello Pokémon?).
Just saying.
Experienced Forum User, Player, Published Author
(99)
Joined: 10/19/2009
Posts: 2527
Location: Italy
It’s impressive compared to RTA play you doofus. Nobody wants to watch a TAS on easy difficulty in a game that people run on hard so that it’s possible to lose a WR run to a boss. And your argument literally makes 0 sense: you’re literally saying “oh it uses savestates so it’s not impressive”. Welcome to TASVideos??? Good argument, man.
moozooh wrote:
But if you don't mind posting a few TAS examples where extra boss fight time results in a non-repetitive, unpredictable fight scenario (where the TASer has—rather than chooses—to employ a varied strategy stemming from the inability to end the fight sooner due to more boss HP), it would definitely enrich the discussion and your argument in particular.
You’re missing the point. It’s not about the TAS. It’s about how it compares to RTA runs of the same game. If the RTA community runs a game on hard to make it more difficult for themselves, everyone is gonna hate on a TAS that runs the game on easy, regardless of whether or not it becomes more or less repetitive. Even if in the TAS the fight becomes repetitive, it could mean that the TASer was able to optimally manipulate the RNG in ways that RTA couldn’t possibly ever do, showcasing a run much more skillfull than any RTA runner could do, which makes it impressive.
A Youyou Kengeki Musou easy RTA will showcase boss fights that are pretty much equal to a TAS, since bosses die too quickly to RNG you. On hard, they’d be worlds apart. That’s why hard is far more impressive. And the only difference is boss HP.
Edit: You clearly have a wrongly biased view on this. I’m just saying that it should be case-by-case, not “more HP? nope not allowed”. I 100% agree that in many games more HP is just a bad idea. Doesn’t mean it’s for all games.
Experienced Forum User, Player, Published Author
(99)
Joined: 10/19/2009
Posts: 2527
Location: Italy
Lmao, how is using pointers complicated? I have no idea how you’re even making CE work without using pointers. Nice info about when the memory ends though.
Experienced Forum User, Player, Published Author
(99)
Joined: 10/19/2009
Posts: 2527
Location: Italy
I disagree about the thing that says “if harder difficulty only gives more HP to enemies/bosses then it just gets more repetitive and shouldn’t be used” or whatever. I’ve always found that thought pretty stupid, more HP means that it’ll take longer to beat enemies, and they’ll have more chances of hitting you and therefore kill you, making a TAS that for example does the whole thing damageless a lot more impressive. I hold the Hard RTA WR in an action-RPG game and have always found the runs on its easiest difficulty to be pretty worthless, even though the only change is “bosses have more HP”. More HP give more chances for different patterns to be shown and more chances for the enemy to kill you.
Experienced Forum User, Player, Published Author
(99)
Joined: 10/19/2009
Posts: 2527
Location: Italy
Emulator input files are very finicky. You should never expect them to work on different emulator versions. They can fail to work on the same version they were made on, even.
Experienced Forum User, Player, Published Author
(99)
Joined: 10/19/2009
Posts: 2527
Location: Italy
There’s a new guy playing the game and routing any% glitchless for Rangers (http://www.twitch.tv/blooiskoo) that gave me the idea that a TAS should actually use a RAcast rather than a FOnewm. You can find elemental weapons as ??? drops, and those are basically Techniques (what FOnewm uses) but have faster animations and I think deal more damage, and you can also use OHKO special attacks with Shadow weapons. Or, depending on how much RNG manipulation is feasable, you could go the route of finding 100% weapons and critting every single Heavy Attack to deal absurd amounts of damage. Not to mention with a cast you get traps to kill slimes much quicker in Caves, and with guns you can make Rappies run away instantly by shooting from far away.
I don’t think a FOnewm would be faster for TAS…
Experienced Forum User, Player, Published Author
(99)
Joined: 10/19/2009
Posts: 2527
Location: Italy
This is cool. It’s hard to vote yes for entertainment, though…
One thing I find weird: you say that Ponyta is the least valuable teammate, but he’s the most used at least up to 27 minutes into the run. Or it seems like it.
Can you go more into details on how you manipulate the RNG?
Edit: For comparison, I think the RTA WR is 30 minutes slower (but the game is emulated faster than console, apparently): http://www.speedrun.com/pkmnstadium
Experienced Forum User, Player, Published Author
(99)
Joined: 10/19/2009
Posts: 2527
Location: Italy
Brian_pso wrote:
Now talking about the sample TAS you did, couldn't you hit De Rol Le while he was doing his laser attack? I remember hitting him with a handgun before, but I'm not totally sure. Either way, your gifoie will wreck him on the real TAS.
Yeah, you need a gun to hit him during that (or possibly higher level Techniques, not sure). With a FOnewm shooting at him would do barely any damage, so I’m not sure whether it’d actually change anything. I didn’t have a gun though so that’s why I didn’t use it.
I’m not planning on doing a serious TAS, this is one of the hardest games in the world to optimize because it’s so RNG-dependent. You’d need perfect drops and perfect maps, both of which are determined at the start of the game and are not manipulable except for waiting longer (well, drops can be manipulated a little during gameplay…). I did that TAS just as a comparison to RTA, I actually started it from my RTA save up to that point. For comparison, you can watch the RTA WR here: http://www.twitch.tv/verallix/v/40233062?t=39m10s
Experienced Forum User, Player, Published Author
(99)
Joined: 10/19/2009
Posts: 2527
Location: Italy
JosJuice wrote:
Based on the beginning of the video, it looks like you're using Dolphin 4.0. The development builds should work better, and versions as old as 4.0 are deprecated for submitting here anyway.
Would newer versions fix desyncs? I thought that was just a settings issue. From what I’ve seen they still don’t have encoding capabalities which is sad, but I could be wrong.
Experienced Forum User, Player, Published Author
(99)
Joined: 10/19/2009
Posts: 2527
Location: Italy
TAS’d any% glitchless De Rol Le with Foie (real TAS would have Gifoie by this point): https://www.youtube.com/watch?v=1KOlLnGt02s
First time using Dolphin to TAS and it was fucking tragic. Don’t even know how to encode with it. Kept desyncing, and the first time I spent 2h30′ TASing without it even recording the input. While redoing it I killed 1 attack faster though so I guess that’s fine…
Addresses to use with Cheat Engine:
Start pointer is value GPOE8P
+25F3E00 = RNG 4 bytes big endian
These are part of player data memory space:
+2DC5CF8 = max HP 2 bytes big endian
+2DC5CFA = max TP 2 bytes big endian
+2DC5D6C = current HP 2 bytes big endian
+2DC5D6F = current TP 2 bytes big endian
+2DC6798 = possessed Meseta 4 bytes big endian signed
Boss memory space:
+2DC991C = Vol Opt’s phase 2 HP 2 bytes big endian
+2DC9A7E = De Rol Le’s HP 2 bytes big endian
+2DC9A82 = De Rol Le’s mask’s HP 2 bytes big endian
+2DC9C4C = Dragon’s HP 2 bytes big endian
+2DC9EEC = Dark Falz’s HP 2 bytes big endian
+2DD0C1C = Vol Opt phase 1 HP 2 bytes big endian
Boss parts memory space?:
+33C6768 = one of many Vol Opt parts’ HP phase 1 (totems, screens, etc.) 2 bytes big endian
Edit: Actually a bunch of these addresses are wrong. Probably all the boss ones. I’d need to find a pointer for them but CE gives me an error when I do a pointer scan, eh.
Experienced Forum User, Player, Published Author
(99)
Joined: 10/19/2009
Posts: 2527
Location: Italy
Pasky13 wrote:
Cheat engine works, only problem is values inside of Dolphin are big endian. So looking at a 2/4/8/float/double value can be troublesome unless you write a custom scan script for Cheat Engine.
2 Byte Big Endian Custom Scan:
—cropped—
Been using this and it’s been great. Today though, for the first time Cheat Engine is reading a value incorrectly… it is 2 bytes big endian (0x0514 = 1300) but it shows as 276. This is a boss’s HP. After I drop it below “0”, it goes to ≈ 1023 then keeps dropping to 0, death. What’s going on here? http://i.imgur.com/EUmF9TY.jpg
Edit: It seems like it’s reading the value with a modulo 1024… so 1024 = 0 and 1025 = 1 for whatever reason…
Edit2: I used the custom type from a different topic and that one works.
Experienced Forum User, Player, Published Author
(99)
Joined: 10/19/2009
Posts: 2527
Location: Italy
xy2_ wrote:
The best way you can truthfully compare these would be to make one or multiple comparaisions with thee same segments of the same game; one on hardware, one without Advanced Bus-Level Timing, and one with. See what is more faithful to hardware.