Design Discussion (1/2017): Lawnmower Investigations

[Ferdielance]

Unofficial Case Design Discussion of the Month: January 2017

What is the Unofficial Case Design Discussion of the Month?

It's a place to discuss specific challenges we face and decisions we make when we design cases.

Currently, to the extent that AAO members discuss design, it's limited to private chats, if it occurs at all. I propose to address one topic each month in a dedicated thread.

How does the discussion work?

Each month, I'll post a problem or challenge in the form of a short post followed by a series of starter questions. I won't give my own thoughts until a few people have replied. You don't have to answer the starter questions directly, though! They're just there to get discussion started!

Ground rules:

1: Do not cite fancases as examples of a bad design decision unless 1) the author(s) have given permission to open up their cases to critique in this thread, or you are the author yourself, AND 2) you can make a positive suggestion as to what that case could have done differently. It's easier to criticize than to solve problems, and we don't want this thread to descend into some kind of public shaming ritual.

2: Commercial cases (such as the AA games) don't require permission from the author for constructive criticism here. However, the rule that no problem should be cited without a positive suggestion still applies.

3: However, it is encouraged that you point out fancases that have handled a design problem particularly well, and set an example we can learn from. There is no need to ask permission for this.

4: Avoid gratuitous off-topic content. For my own part, I promise not to spam pigeons.

5: Of course, all of the usual forum rules apply, too!

-------

Fancases open to constructive criticism (i.e., pointing out problems and possible solutions:)

* Any case by FerdieLance alone, as well as Phantasmagoria of Betrayal (with Enthalpy)
* Enthalpy's Turnabout Proxy
* Reecer6's The Normal Turnabout
* Blackrune's cases (barring collabs with people not on here.)
* BP's cases (barring collabs with people not on here.)
* DWaM's cases (barring collabs with people not on here.)
* (If you're willing to be added to this list, please PM me and I'll add you.)

-----
EDIT: Argh! I accidentally edited this post when I meant to copy and paste the format for the March one! Well, the starter questions are gone now, but hopefully this discussion will still make sense.

[TheDoctor]

I don't really know. I always found investigation segments to be the most tedious in any Ace Attorney game. But then, when Dual Destinies came out, I hated how linear they made the investigation, so my thoughts on this aren't exactly consistent.

One example I can think where the investigation structure was problematic was in 1-5. There's a point where you need to examine a crumpled note on the floor; however, there's a character blocking your view of this note. Now, the game points it out to you the moment you walk into the room. The problem? Let's say you have to save and get off the game before you can examine the note. Once you get back into the game, the game no longer prompts you to look at the note, but you're blocked from progressing until you examine it. Once again, the character is blocking your view of this note, so it's very easy to forget it exists, and this character won't go anywhere else until the note is dealt with.

Possible solution? Just add the note to the dialogue options instead of requiring examination.

Another problem I see with AAO cases in general is poor examination feedback. In the official games, the examine button will pop up whenever your cursor hovers over anything noteworthy (and in DD and onward, it'll even tell you whether you've already examined it); however, in AAO, we don't know what is meant to be clicked on and what's just background.

Possible Solutions? Make an optional examination overlay that'll let the player know what items can be clicked on. Make clickable items incredibly obvious.

[kwando1313]

Sliiiiiiightly off topic here, but I know that I always wanted to do the whole make the examine button pop-up thing if you're hovering over an examinable object.

But yeah, unfortunately this is a big issue in pretty much AA game. Most investigations are just go pixel hunting and find the missing object. IMO, 1-5 is the biggest offender for this (and one of the big reasons I really don't like it as a case), since there are many many pixel hunts from what I remember of the case. Another example I can think of is 1-4, where you're (really obviously and stupidly) forced to present the DL-6 file to Manfred von Karma, and then he tazes you. Which is a really dumb thing to do, but is needed by the plot.

Similarly as to what TheDoctor said though, linearizing investigations is not a good or well-thought out solution either, since it eliminates the sense of control from the player. Instead of having an investigation, you're getting at best an interactive cutscene where you point out and click on various different things in order to make sure you're going to proceed in the scene.

Lawnmower design, however, is good at one thing - ensuring that players get everything before proceeding on to another event. This prevents the player from not seeing vital cutscenes/information that they'd need to solve the case. As a result, I get why it's done. It's an easy and efficient manner to force players to get all the flags done to make sure that they are at a point that they can solve the case.

So, to me, ultimately, it's what type of game does AA want to be, that matters most to me. And, since I don't think AA is meant to be a pure mystery game, the lawnmower design works well for its usage - it's a Visual Novel with mystery elements to it, rather than a mystery game in a visual novel format. As a result, the lawnmower design makes sense - you make sure the players get all clues so they are able to solve the mystery (and, tbh, most AA games are pretty easy to solve).

For an AAO-standpoint though, I think we've moved beyond that point. Our fancases... Aren't very AA-esque anymore. They either go to the more story-centric focus (where, I'd still think a lawnmower design works fine), or more towards a mystery-focus, where the lawnmower design... Well, just doesn't really fit nicely. Since in a mystery, I'd say that the finding of clues is more important, I'd say a more open-ended structure where you can do things in order and where time actually flows would be better. Clues would be missable! People would move around! A more sandboxed idea (I think like what Ferdie was trying to pull off in AtBaS) would be a really good way to mitigate this issue. However, this approach would be extremely difficult to pull off in the AAO engine as is, due to just the sheer complexity of variables of a "living" world. (I've tried to do this for something else. It's really hard to keep track of time with different examines and whatnot and slip in different events).

So. The tl;dr. Lawnmower design, in a pure-AA perspective, isn't necessarily bad, since it's doing what needs to get done in a easy to understand manner, however, for the way the AAO meta has moved onto now, I don't think that it's a great way to structure our investigations.

[TheDoctor]

The question of what type of game does the writer want it to be is actually pretty important. In a sound novel, I have no problem just click-click-clicking through the text, because that's the only choice I have. However, in the traditional Ace Attorney games, I much prefer the court segments and spotting the contradictions over the investigation scenes. The most enjoyable points of an investigation segment for me were using the Magatama, which isn't to say the investigations were all bad outside of that, it's just that the Magatama brought what I liked about the game outside the court.

One fangame that I thought did a great job with integrating the beloved courtroom gameplay into the investigation was The Virtual Turnabout. Much like AAI, you had to rebut someone's argument before ever getting to court. The difference here though was that you still ended up going to court, which for me, is part of the AA experience, and partly why AAI just isn't quite the same (even though AAI2 was really freaking good).

[kwando1313]

Yeah, I think that integrating court-like segments into investigations is a much better way to deal with just pure linearity in investigations.

Like, it feels like psyche-locks were an attempt to bring a court-like segment there, but it still acts, ultimately, like a chokepoint, since you need to have all the pieces of evidence to break the locks, it still falls under the idea of lawnmowering, in my opinion.

But still, it's a better attempt than nothing.

Maybe like a full "come up with a theory, we'll come up with a rebuttal" section during investigations would help? That way, it feels at least that the investigation isn't for the sake of having one, but more to come up with a case.

But, then again, it does go against the whole idea of hopelessness and the idea of a turnabout.

Hmm.

[Gamer2002]

For Firefox you do have examine button, sort of. When you hover over any portion of examined picture you see in left bottom window javascript code that will direct you to a specified frame. Just memorize the number for "nothing here" and you can identify all points of interest.

Anyway, currently I'm working on Part 3 of Confessed Turnabout, which is second day investigation (writing mostly done, just optional presents and examines left, though recently I don't have much time to work on it). My series is intended to be my take on AA game, so I do things like in main series - first trial against Payne, second trial with investigation against rival prosecutor and such. So, Lawnmower Investigations is also used, since as stated above it's efficient way of ensuring player seeing everything in intended order. Part 1 was it played straight as arrow - talk with people, present defendant's request to Gumshoe, examine outside of motel, talk with people, examine motel room, talk with people.

Still, that was Part 1, which was first investigation in my game. Part 3 is actually more spiced up. To not spoil too much, there is examination section that isn't about clicking stuff in the background, psyche-lock section without psyche-locks and a riddle. Funny thing is that the riddle wasn't originally planned. At first the player was supposed to go through "pick only one evidence out of two because reasons" and presenting it to a certain person for a lengthy exposition. But then I said "that's freaking lame" and I made a riddle that forces the player to figure out the core information of exposition's first half.

AA's problem with investigations isn't in its basic design, which is fine for its purpose. The problem is that they contrast the fast paced trials with tedious busywork. And no, Rebuttals aren't solution to investigations, they are trials' gameplay that take place in investigation's setting. Using them to "fix" investigations is escape, not solution. Trials/Rebuttals gameplay is "read dramatic argument -> solve problem posed in front of you -> repeat". Investigation's gameplay is "talk/present/examine -> figure out where the hell you are supposed to go next -> maybe try presenting everything on everybody -> have you clicked on the clock in Larry's mansion, though I don't cite any specific trial as example? -> repeat". The trials/rebuttals don't force the player to think "what's next", the prosecutor/investigation rival/witness/judge constantly slams your face with the next. But during investigation, you tend to grow impatient about the next.

Of course, it wouldn't be investigation if the player didn't wonder "what's next". But like there is a difference between good cross-examination and bad one, there is a difference between good investigation and a bad one. We have tutorials for the former, is there any for the latter?

There are ways that can turn "what's next" from source of pointless frustration into a proper question.
* Both in Part 1 and Part 3 of Confessed Turnabout once Angela does what is needed in certain location, she states that there is noting left to do there anymore. Also in Investigations, when Edgeworth examines something closely (zoomed in, like in normal AA), he says "I must look in every nook and cranny" when he still has to find something and "I may take another look" when he doesn't. That way the player is told he can move elsewhere and there is no point in wasting time here.
* In my unreleased Mystery of Mason, investigation portion ends when Phoenix goes with Kristoph on a meeting. But when you talk about the meeting with Kristoph, Wright lists all the things he still has to do and says the meeting must wait. When you do something from that list, Phoenix doesn't mention it anymore. Only after you are done with everything you can decide if you already want to go on the meeting. Also, for Part 3 of Confessed Turnabout, I plan your assistant to update his "what to do" conversation, to give current directions to the player. Regarding what Doc mentioned about the paper in 1-5, it wouldn't be such a problem if unless you picked it up Ema was reminding you about it in her "what to do" conversation. You don't have to explicitly tell the player everything, but stuff like "remember the paper I want you to pick up" or "I don't think we checked everything in Gant's office" or even "let's confront von Karma about the latter! WHAT CAN POSSIBLY GO WRONG, NICK!?!" gets ride of problems you don't want the player to have.
* If the player is supposed to examine, make the place with stuff to examine. There is a reason why there is some funny dialog when you investigate in AA, even if you click something less relevant you get something that reinforces the sense of stuff being here to examine. So don't make the entire Larry's mansion "nothing here" with a tiny section containing a clock, though I'm not citing any specific trial as example. If I read "nothing here" 99 times I expect nothing for the 100th time.

Of course, there is only a single rule to creativity - there are no rules, only tried and tested methods. In Part 1 of Confessed Turnabout, after the player investigates motel's outside and talks in detention center, I don't make Angela tell the player any directions and let him wander around between locations. She literally says "I'm done with everything and I don't know what to do next". Because, at that point of story, she hits the wall in her investigation without accomplishing anything. Next she was supposed to return home after walking aimlessly and I wanted the player to have that aimless walking feel. And for mentioned earlier clock in Larry's mansion, while I'm not citing any specific trial as example, some specific trial could have it to be cruelly frustrating by design.

Anyway, there are also other ways to improve investigations. Good writing. Have some drama. A lot of AAO investigations, especially on the first day, are "let's talk with our client, go to the crime scene, talk with obnoxious people (often this happens already with our client), gather some evidences and call it a day". When AA first time used this plot structure? 1-3, the "filler" case where Phoenix does lawyering as his regular job instead of dealing with something personal. And 1-2, the first investigation in the game? Mia dies, you meet Maya, you examine your office, you meet Gumshoe, Gumshoe immediately arrests Maya, you trick Gumshoe and get stuff from him, you learn about Edgeworth, you agree to talk with Grossberg for Maya, Grossberg disappoints you, you become Maya's lawyer, you steal things from April May's room. Stuff happened.

And spice things up with additional minigames. Logic, Logic Chess, Little Thief, looking for prints, Psyche-Locks, examining videos... Some of those are recurring character's abilities that make the character stand out, some of those are cases' gimmicks that make their own cases' investigations stand out.

[Kroki]

Super interesting topic.
Gamer2002's suggestion of having an evolutive "What to do?" convo reminds me of a feature I was working to implement in the someday on AAO project :

Spoiler : :

The button with a face lets the player interact with the current person that's accompanying the MC (=assistant) and talk about the location, case, strategy... Depending on the current location and witness.

Because assistants can be more than furikake!!!

[Reecer6]

woah, someday on aao indeed.

Another problem I see with AAO cases in general is poor examination feedback. In the official games, the examine button will pop up whenever your cursor hovers over anything noteworthy (and in DD and onward, it'll even tell you whether you've already examined it); however, in AAO, we don't know what is meant to be clicked on and what's just background.

Possible Solutions? Make an optional examination overlay that'll let the player know what items can be clicked on. Make clickable items incredibly obvious.

Thiiiiis is exactly why I never put in a "Nothing here" dialogue. If you have no way of knowing what items can elicit special dialogue out of an entire scene, the next best thing is to be allowed to check each item for said dialogue as fast as physically possible. i.e., spam-clicking.

Back on topic (well, to some extent): From a shallowly analytical point of view, in the Ace Attorney games, the investigations were probably conceived by Takumi and friends as a device to break up the tense, drama-filled court scenes while still building the narrative - a very necessary thing, because otherwise, the story gets flat-out tiring. I don't actually think they're particularly any more information-filled than the court segments, which is kind of interesting to me. In investigations, obviously you're talking to people and picking up clues, but in the court segments, there's surprise evidence and tons of witness testimony you've explicitly never heard before.

The main differences between the two, then, is simply the tone and gameplay. Which makes sense, because, in video games, breaking up traditional gameplay with something else is also very important, hence why most action-orientated games have tiny mini-games thrown in every now and then. The two special aspects of investigations work in tandem to allow the player a break, which is the exact reason lawnmower-type gameplay is so mechanically important and useful (This is the thesis! I've buried this a bit because I've been writing this very stream-of-conscience). The player can sit back and click buttons and read for the most part, without HAVING to even rack their brain and come to any conclusions, and the story will still proceed as Takumi intended.

The only real problem with this school of design is in terms with how evenly the two segments, the yin and yang of Ace Attorney, Investigation and Court, are mixed. In the first game, you had three-day trials, which, while allowing plenty of breaks in between lawyering, was very disruptive pacing-wise. Every couple of facts that went by, the game totally switched up on you, and suddenly you're in an entirely different mood. Meanwhile, you have the opposite end of the spectrum, Spirit of Justice, with its one-day trials where every piece of information comes very swiftly and smoothly in one steady stream, but it's such a long stream that it begins to feel like a marathon, or maybe the first leg of a triathlon. You start feeling a bit beleaguered, wondering just when the action will finally change pace and you can stop, or start up, depending on just what half of the system you're on.

So, what I'm saying is the Investigations system of investigating is way better than the main series system, and we all need to start writing like that.

[Bad Player]

I don't really know. I always found investigation segments to be the most tedious in any Ace Attorney game. But then, when Dual Destinies came out, I hated how linear they made the investigation, so my thoughts on this aren't exactly consistent. One example I can think where the investigation structure was problematic was in 1-5.

This seems like a separate issue to me. Lawnmower investigations to me sounds like an issue where investigations are boring, whereas you are discussing investigations that are obtuse. You can have an investigation where it's always clear what you need to do but it's really boring (lawnmower), but you can also have an investigation where it's exciting and stuff is happening but it's frustrating because it's always difficult to figure out what to do next.

They either go to the more story-centric focus (where, I'd still think a lawnmower design works fine), or more towards a mystery-focus, where the lawnmower design... Well, just doesn't really fit nicely. Since in a mystery, I'd say that the finding of clues is more important, I'd say a more open-ended structure where you can do things in order and where time actually flows would be better. Clues would be missable! People would move around! A more sandboxed idea (I think like what Ferdie was trying to pull off in AtBaS) would be a really good way to mitigate this issue.

I disagree and think it's actually the exact opposite. Like you say in the part that I snipped off, how you do the investigation probably depends more on what type of game you want to be and what you're trying to do, but in general, I think an open-ended investigation would fit much better into a story-focused game than a mystery-focused game. If you want to focus on the story, then you can still focus on the story if the player misses something--it just means the story is a bit different than it would have been. But if you're focusing on the mystery, it'd feel really really bad if you can't beat the game because you can't accuse the true killer because you don't have that one piece of evidence you needed because you didn't talk to that specific person at that specific place at that specific time 7 hours of gameplay ago.

One fangame that I thought did a great job with integrating the beloved courtroom gameplay into the investigation was The Virtual Turnabout. Much like AAI, you had to rebut someone's argument before ever getting to court. The difference here though was that you still ended up going to court, which for me, is part of the AA experience, and partly why AAI just isn't quite the same (even though AAI2 was really freaking good).

I agree with Gamer that I generally don't think just sticking trial gameplay into investigations is a very good solution. The traditional paradigm is that investigations are for gathering information, and then trials is where you use that information and make deductions. My concern is that putting trial-like gameplay into investigations means you're making deductions, which you could probably have done in the trial segment, which you obviously can't do anymore--in other words, you're taking trial segment material and putting it into the investigation, which makes the investigation more exciting, but makes the trial less exciting because there's less stuff. So it'd be okay if it's about stuff that definitely couldn't come into the trial--which is the case in Virtual.

Yeah, I think that integrating court-like segments into investigations is a much better way to deal with just pure linearity in investigations. Like, it feels like psyche-locks were an attempt to bring a court-like segment there, but it still acts, ultimately, like a chokepoint, since you need to have all the pieces of evidence to break the locks, it still falls under the idea of lawnmowering, in my opinion.

It's not like trial-like segments in investigations really fixes that. After all, you need to make sure the player has all the evidence necessary to complete the trial-like segment. So it actually becomes a stricter chokepoint than a psyche-lock, because the restriction is on when you can begin, rather than complete, the segment. Adding a trial-like segment in the investigation doesn't really make it less lawnmowery imo; it just splits the investigation into two lawnmowing segments with a mini-trial in the middle.

* In my unreleased Mystery of Mason, investigation portion ends when Phoenix goes with Kristoph on a meeting. But when you talk about the meeting with Kristoph, Wright lists all the things he still has to do and says the meeting must wait. When you do something from that list, Phoenix doesn't mention it anymore. Only after you are done with everything you can decide if you already want to go on the meeting. Also, for Part 3 of Confessed Turnabout, I plan your assistant to update his "what to do" conversation, to give current directions to the player. Regarding what Doc mentioned about the paper in 1-5, it wouldn't be such a problem if unless you picked it up Ema was reminding you about it in her "what to do" conversation. You don't have to explicitly tell the player everything, but stuff like "remember the paper I want you to pick up" or "I don't think we checked everything in Gant's office" or even "let's confront von Karma about the latter! WHAT CAN POSSIBLY GO WRONG, NICK!?!" gets ride of problems you don't want the player to have.

I think this is also conflating the issues of trials being boring and being frustrating. You could have Maya there telling Nick how many unmowed patches of grass are left and which quadrant of the lawn they're in, and that'd make the game less frustrating, but it'd still be boring.

Another example I can think of is 1-4, where you're (really obviously and stupidly) forced to present the DL-6 file to Manfred von Karma, and then he tazes you. Which is a really dumb thing to do, but is needed by the plot.

I feel like this is more a problem with the plot/scenario, and doesn't have much to do with the investigation at all. They could've confronted Manfred in the courthouse lobby in the trial segment and had the same thing happen, and not much would've changed.


will post actual thoughts (not just responses) later.
people can bash my trials all they want (although technically silence is the only one that I did alone but idk)

[Gamer2002]

I think boring/frustrating fells into same category when we talk about game design. The difference boils down to whatever the player becomes apathetic or angry over unnecessary problems we didn't actually intended them to have.

I took situation described by Fernie more as "you examine, present and talk between locations until the plot moves forward" than "there is ton of boring things you have to do and you have to do them again every time the story advances". I don't think there are AA games where you have to completely examine from top to bottom entire location multiple times. During first visit, yes, but next visits you examine anything only if something changes in the location. And it's not always the case, it mostly happens only in core locations of investigations, like crime scenes. You do have to use every talk topic every time you meet a character, but here writing is colossal factor when it comes to things being boring.

It's true that investigations are supposed to move on slower peace than trials. But the feel of slower pace comes already from ability to move, talk and examine at your own pace, instead of "read dramatic argument -> solve next problem". But then again, sometimes you have to make investigation more intense. Part 3 of Confessed Turnabout is long, and if things were always calm and without revelations, it would be completely boring. After all, there is giving time to breath and there is making things unnecessarily drag (I'm looking at you, 1-3 with your fetch quest nonsense).

And design is all about avoiding unnecessary and unintended problems.

[Blackrune]

Ideas to reduce lawnmowing or make it less boring:
1. Good optional stuff. If the player isn't forced to see it, it can feel more rewarding to find it. If the game acknowledges a lot of strange things the player may want to do, exploration can feel like a privilege instead of a burden.

2. Choices. Just choosing how to reply sometimes might help the player to stay engaged. Choosing between multiple ways to get a certain piece of information also works, as the player will want to see if they decision works out. (of course, usually all decisions will lead back to the same result, but it still works.)

3. Complications. AA already has these sometimes, like witnesses refusing to talk to you until you show them certain things. (and... psyche-locks) But it could be taken further, like small escape-the-room puzzles or anything. (bonus points if the stuff you need to gather to escape is also important evidence later - one would be sure to remember them quite well.) This one shouldn't be overdosed though. If done right it will feel rewarding, if done wrong the player will be frustrated that every little thing requires him to do extra stuff.

4. Plot twists. You're not supposed to do much deducing during the investigation, but if the information that is gathered is already very interesting on its own that can go a long way. Difficult, of course, because the biggest twists should be saved for the trial, but when someone you interrogated a few minutes ago is suddenly found dead, that's pretty cool stuff.

5. Actually making it optional to find everything. I'm not sure how exactly this would work, but presumably you just get the choice to 'end' the investigation when you've done a lot but not everything, and the detective or someone fills you in on the couple of things that you missed. Could probably also be done wrong and just be an infodump, but if the player already considering the lawnmowing a boring infodump, you might as well.

6. Evolving locations. There's a lot of this already, usually to let you know when there's something new to find in a place. Frequent use of these will make the player feel more like progress is made, since it gives a sense of time passing. I guess these might not actually reduce lawnmowing (they might even increase it, if new grass is growing in old places), but they do reduce the chance of it feeling boring.

7. Just have no investigation. If absolutely nothing interesting can be made to happen in it, it might just not be worth it. Information can be spread throughout the trial, the pace can be slowed down with a recess instead.

---

I don't think lawnmower-design is very problematic for small to medium investigations. Of course there should always be something to keep the player engaged from the get-go. But I think with the right writing and interesting setup, it can be made pretty likely that the player will want to gather more information, and the lawnmowing usually won't feel that bad. Of course course if looked at without the writing, it can look like pretty dull, basic, gameplay, and then it probably comes down to what the individual player values slightly more. Though I don't think most AA players go in expecting the investigation to even have as much gameplay as, say, an adventure game. They know a trial is coming up, usually.

Permission granted to constructively bash all my concoctions, for the greater good.

[Bad Player]

I think boring/frustrating fells into same category when we talk about game design. The difference boils down to whatever the player becomes apathetic or angry over unnecessary problems we didn't actually intended them to have.

I think the difference is that if it's boring, the player doesn't want to or doesn't care about continuing, whereas if it's frustrating, the player wants to continue but isn't able to. Which seems quite different to me; it isn't just about whether the player feels angry or apathetic, but whether they want to play your game at all.

[Acid Rain]

Cool topic - I always find these types of discussions interesting.

The first thing I want to talk about is how investigations fit into typical AA story structure. Personally, I think investigations can be just as good as trial segments if done correctly, in fact, my all-time favourite segment out of all the canon AA games is the second investigation in 2-4. Obviously, that particular investigation is a little different from your usual investigation. Setting aside the fact that this investigation actually holds 2-4's biggest twist, this part does a number of other things that AA cases don't typically do. First is this investigation's use of Matt Engarde's Psyche Locks (which, IIRC, are first shown early on and not solved until much later) to build mystery and suspense rather than block the player's progress. In the same vein, there's a point where Nick and Gumshoe have this slow-burning investigation into things that are not immediately related to the murder (camera/stuffed bear/etc, which are basically never seen in the following trial segment). Finally, the investigation ends with Phoenix and Edgeworth breaking into Engarde's mansion (in what is a very high-energy scene for an investigation), when they could have gotten the information across in any number of more efficient ways, such as a phone call, which they did in many other instances throughout 2-4.
In summary, this investigation:
-Introduces mysteries for suspense as well as plot
-Involves the player investigating the case in a non-direct manner.
-Progresses the plot using scenes that are interesting and dynamic.

The caveat to this is that the plot of the case needs to be developed such that these things can be included. AA protagonists don't often do stuff like the mansion raid because most AA plots don't need them to - the investigations are primarily there to introduce new material to explore in the next trial segment. I would argue that the plot of 2-4 is atypical of canon AA, which is why it has room to include this cool stuff in its story. Your "typical" fancase tries to imitate AA structure, so naturally they are also constructed to supplement trial segments. This can end up being pretty boring, since the investigation is basically setting up a bunch of stuff to happen in the next part, (which probably isn't released lol) and isn't satisfying by itself.

One other thing, which has been brought up already, is that most investigations don't incentivize the player to investigate other than to forward the plot. This is a bit of a let-down since investigations focus on exploration and are inherently more open than trial segments. As an example of what to do in the case of player incentive, I'm going to reference my own work, R E M, even though I'm a little hesitant to do so because it's not AA in the slightest (and also, hubris). If one were to play R E M and advance straight to when puzzles start popping up and then immediately solve every puzzle (lol) to advance to the labyrinth and then get through that to beat end the game, it would probably take them, like, 30 minutes. Instead, it can take an hour just to find any puzzles simply because there is so much stuff to examine and things to look at. It also really makes a difference whether a game uses custom content or not (at least for me). I think reusing the AA locations in the editor for new locations is extremely dull and is a huge turn-off for me. It's far more interesting when I have something new to look at, even if it's just something from those BG packages. Same goes for reusing canon sprites and evidence that doesn't quite fit the description. I think finishing touches like these go a long way in making an investigation more enjoyable.

Now, there's a very big and obvious problem with everything I've brought up here: this crap is difficult to pull off, and even harder to pull off well. It's not easy coming up with contradictions and whatnot for an AA case, and it's even tougher trying to screw around with AA story structure while keeping gameplay intact - they have a formula for a reason. Obviously, quality graphics are hard to come by, and good luck finding something that works with anything specific you had in mind without moderate-to-heavy editing. And it's hard to be witty enough to fill an entire game with Examine conversations that don't suck.

Basically, making a good investigation is just way, way harder than making a good trial.

[Ferdielance]

Volunteers for constructive critique added to top post. Thanks to all!

---

I don't really know. I always found investigation segments to be the most tedious in any Ace Attorney game. But then, when Dual Destinies came out, I hated how linear they made the investigation, so my thoughts on this aren't exactly consistent.

I second this - and I think AA6 struck a better balance than AA5 in this regard. I had more of a sense that I could poke at objects that were obviously irrelevant and get an entertaining response, and I thought that there was just a tad more (illusion of) freedom. That said, it was still pretty linear!

[...] n AAO, we don't know what is meant to be clicked on and what's just background.

Possible Solutions? Make an optional examination overlay that'll let the player know what items can be clicked on. Make clickable items incredibly obvious.

I think this overlay could be done without too much difficulty using objects!

So, to me, ultimately, it's what type of game does AA want to be, that matters most to me. And, since I don't think AA is meant to be a pure mystery game, the lawnmower design works well for its usage - it's a Visual Novel with mystery elements to it, rather than a mystery game in a visual novel format. As a result, the lawnmower design makes sense - you make sure the players get all clues so they are able to solve the mystery (and, tbh, most AA games are pretty easy to solve).

I disagree and think it's actually the exact opposite. Like you say in the part that I snipped off, how you do the investigation probably depends more on what type of game you want to be and what you're trying to do, but in general, I think an open-ended investigation would fit much better into a story-focused game than a mystery-focused game. If you want to focus on the story, then you can still focus on the story if the player misses something--it just means the story is a bit different than it would have been. But if you're focusing on the mystery, it'd feel really really bad if you can't beat the game because you can't accuse the true killer because you don't have that one piece of evidence you needed because you didn't talk to that specific person at that specific place at that specific time 7 hours of gameplay ago.

This is interesting. Let me see if this diagram helps capture part of the issue here. Graham Nelson said that an adventure game is a "crossword at war with a narrative." But there's also a tension between "what the author tells the player" and "what the player gets to decide for themselves:"



This isn't a strict binary on either axis. Maybe I should have made this a graph with axes instead of a table. But regardless...

BP seems to be saying that the upper-left quadrant is safer than the lower-left quadrant, and the lower-left quadrant poses a huge risk of missed information. kwando seems to be saying that the lower-left quadrant feels more like "solving a mystery" in an engaging way. I don't think these views are contradictory.

And no, Rebuttals aren't solution to investigations, they are trials' gameplay that take place in investigation's setting. Using them to "fix" investigations is escape, not solution.

This is a good point. I think that if a designer says "The point of investigations is exploring and discovering clues, and I want to make the exploration and discovery fun," then trying to add tension by throwing in confrontations evades the point. At the same time, Psyche-Locks demonstrate that some confrontation and "working out of problems" can be put into investigations without diluting them.

But when you talk about the meeting with Kristoph, Wright lists all the things he still has to do and says the meeting must wait. When you do something from that list, Phoenix doesn't mention it anymore.

Yes, we need more of this. My plans for the ridiculously delayed/abortive All the Brooks and Soldiers include a to-do list. There's no point in making a mystery of something that really is kind of mechanical - doing tasks off a checklist.

You could have Maya there telling Nick how many unmowed patches of grass are left and which quadrant of the lawn they're in, and that'd make the game less frustrating, but it'd still be boring.

...And this is what I'm worried about with a to-do list - that it might serve to cover up bad design or writing.

The button with a face lets the player interact with the current person that's accompanying the MC (=assistant) and talk about the location, case, strategy... Depending on the current location and witness.

This is neat. I think many players are reluctant to seek "hints," or wouldn't necessarily think the hint would be helpful. But if they were frustrated enough to click it and get an actually useful piece of advice, that would be a very pleasant surprise!

in video games, breaking up traditional gameplay with something else is also very important, hence why most action-orientated games have tiny mini-games thrown in every now and then.

Hmm. Minigame-based design (in the strictest sense of 'a sudden big change in game mechanics') is often seen as a bit antiquated. One of the major advances Half-Life made was in doing away with minigames and cutscenes, and folding in all of the "set pieces" and changes of gameplay smoothly into the style of the main game.

I wonder if there's a way to do something similar in AA? That is, to make investigation and trial "integrated" in the sense that the same tools are used for both? AAI seemed like an attempt at this, and a mostly successful one, but I think we could go further.

I have some ideas for that, but they're outside the scope of this thread. Or maybe not... I might write them up later.

And design is all about avoiding unnecessary and unintended problems.

It can also be about charging at them head-on, waving a battleaxe and screaming! I'd cite some of the ICOM adventure games, such as Deja Vu, which not only had red herrings but put great effort into filling every scene with red herrings that could be picked up. Or some adventure games with restrictive time systems that make it very easy to get stuck on an unwinnable path, but also add tension.

Or, outside of VNs and adventure games, Dwarf Fortress.

Of course, once one has decided to go after all the hairy problems and tackle them, one must still manage complexity even then. So the statement holds true even in the most ambitious case, to some extent...

1. Good optional stuff. If the player isn't forced to see it, it can feel more rewarding to find it. If the game acknowledges a lot of strange things the player may want to do, exploration can feel like a privilege instead of a burden.

2. Choices. Just choosing how to reply sometimes might help the player to stay engaged. Choosing between multiple ways to get a certain piece of information also works, as the player will want to see if they decision works out. (of course, usually all decisions will lead back to the same result, but it still works.)

3. Complications. AA already has these sometimes, like witnesses refusing to talk to you until you show them certain things. (and... psyche-locks) But it could be taken further, like small escape-the-room puzzles or anything. (bonus points if the stuff you need to gather to escape is also important evidence later - one would be sure to remember them quite well.) This one shouldn't be overdosed though. If done right it will feel rewarding, if done wrong the player will be frustrated that every little thing requires him to do extra stuff.

4. Plot twists. You're not supposed to do much deducing during the investigation, but if the information that is gathered is already very interesting on its own that can go a long way. Difficult, of course, because the biggest twists should be saved for the trial, but when someone you interrogated a few minutes ago is suddenly found dead, that's pretty cool stuff.

5. Actually making it optional to find everything. I'm not sure how exactly this would work, but presumably you just get the choice to 'end' the investigation when you've done a lot but not everything, and the detective or someone fills you in on the couple of things that you missed. Could probably also be done wrong and just be an infodump, but if the player already considering the lawnmowing a boring infodump, you might as well.

6. Evolving locations. There's a lot of this already, usually to let you know when there's something new to find in a place. Frequent use of these will make the player feel more like progress is made, since it gives a sense of time passing. I guess these might not actually reduce lawnmowing (they might even increase it, if new grass is growing in old places), but they do reduce the chance of it feeling boring.

7. Just have no investigation. If absolutely nothing interesting can be made to happen in it, it might just not be worth it. Information can be spread throughout the trial, the pace can be slowed down with a recess instead.

I might want to work with this list, add, and amend it. If we get enough suggestions and examples, we can sticky it at the top of the post.

[Enthalpy]

Starting out with a hit parade...

I don't really know. I always found investigation segments to be the most tedious in any Ace Attorney game. But then, when Dual Destinies came out, I hated how linear they made the investigation, so my thoughts on this aren't exactly consistent.

It's perfectly consistent to say that Takumi's style has a bad trait, and that Yamazaki's style has a different bad style. What you're describing is instead a dilemma where you haven't decided which option is worse. There may well be a third option here.

One example I can think where the investigation structure was problematic was in 1-5. There's a point where you need to examine a crumpled note on the floor; however, there's a character blocking your view of this note. Now, the game points it out to you the moment you walk into the room. The problem? Let's say you have to save and get off the game before you can examine the note. Once you get back into the game, the game no longer prompts you to look at the note, but you're blocked from progressing until you examine it. Once again, the character is blocking your view of this note, so it's very easy to forget it exists, and this character won't go anywhere else until the note is dealt with.

Possible solution? Just add the note to the dialogue options instead of requiring examination.

Another problem I see with AAO cases in general is poor examination feedback. In the official games, the examine button will pop up whenever your cursor hovers over anything noteworthy (and in DD and onward, it'll even tell you whether you've already examined it); however, in AAO, we don't know what is meant to be clicked on and what's just background.

Possible Solutions? Make an optional examination overlay that'll let the player know what items can be clicked on. Make clickable items incredibly obvious.

This is a problem, true, but echoing kwando, I'd rather see this solved at the programming level.

But yeah, unfortunately this is a big issue in pretty much AA game. Most investigations are just go pixel hunting and find the missing object. IMO, 1-5 is the biggest offender for this (and one of the big reasons I really don't like it as a case), since there are many many pixel hunts from what I remember of the case. Another example I can think of is 1-4, where you're (really obviously and stupidly) forced to present the DL-6 file to Manfred von Karma, and then he tazes you. Which is a really dumb thing to do, but is needed by the plot.

Can you explain how the DL-6 file ties in to the rest? (I'm not sure about your "moved beyond that point" argument, but I'll save that until I understand you on this point.) I think you're equating lawnmower with "chokepoints that aren't what the player will naturally do," but I want to make sure.

[L]inearizing investigations is not a good or well-thought out solution either, since it eliminates the sense of control from the player. Instead of having an investigation, you're getting at best an interactive cutscene where you point out and click on various different things in order to make sure you're going to proceed in the scene.

But is there a real sense of control in the lawnmower model, where you also "point out and click on various different things in order to make sure you're going to proceed in the scene"?

AA's problem with investigations isn't in its basic design, which is fine for its purpose. The problem is that they contrast the fast paced trials with tedious busywork.

Would tedious busywork be fine without this contrast?

But like there is a difference between good cross-examination and bad one, there is a difference between good investigation and a bad one. We have tutorials for the former, is there any for the latter?

Yes, actually.

Gamer2002's suggestion of having an evolutive "What to do?" convo reminds me of a feature I was working to implement in the someday on AAO project :

Good idea! Want me to add it to the list on the Open Source project?

This seems like a separate issue to me. Lawnmower investigations to me sounds like an issue where investigations are boring, whereas you are discussing investigations that are obtuse. You can have an investigation where it's always clear what you need to do but it's really boring (lawnmower), but you can also have an investigation where it's exciting and stuff is happening but it's frustrating because it's always difficult to figure out what to do next.

Agreed with this. I think that many of the posts here have conflated the two. While both "I'm bored" and "I'm not sure how to progress" are problems...

I think boring/frustrating fells into same category when we talk about game design. The difference boils down to whatever the player becomes apathetic or angry over unnecessary problems we didn't actually intended them to have.

...the causes of the two are different, which means the remedies are entirely different.

I'm generally agreed with Rune's list.

I have some more thoughts on this, but I want to check a few things before posting.