We're not going to see stuff like this in big budget, AAA titles with detective/crime angles simply because they are mainly designed as power fantasy outlets. The entire "detective vision" mechanic in the Arkham games is pretty much a microcosm of this.
This was admittedly one of the things that, as I recently played through the Arkham games again, I realised was actually kinda disappointing. For the most part you simply don't do any actual detective work - you just find all the items the game marks as 'evidence', and then follow the dots after Alfred, Oracle, or the Bat Computer has done all of the actual investigative work. Arkham Knight at least introduces various, timeline related puzzles, so at least finding the evidence involves a bit more work. The hologram sequence is just kinda egregious though, since it relies on information not even Batman is aware the computer has somehow already factored in.
I'm a big fan of the Game Makers Toolkit youtube channel but he definitly forgot Blade Runner.
It fits that category perfectly and the biggest advantage of Blade Runner is the non linearity. You can't make a wrong deduction. The story adapts to the choices you make.
It's one of the best games ever made.
Thinking some more, there are so, so many games that could have been mentioned. I guess we've been trying to solve this genre for a long while.
Take text adventure mysteries. I haven't played many, but for example, An Act of Murder puts 100% of the investigating and mystery-solving into the player's hands. That's the strength of a text parser, I suppose, though the total lack of direction is the consequence. That approach certainly takes us farther still from mass appeal, if that's to be a goal.
There might also be a place in the discussion for games like Umineko: When They Cry. Rather than a "detective game", they're essentially "mystery novels that ask you more directly to solve them". Still, the mystery novel has a tried-and-true tradition of making the reader feel like a detective, and when it works, it works fantastically, so I don't think it should be ignored.
In the particular case of Umineko, the extremely brief bits in which the game asks for input were more or less an afterthought for most players -- the main bulk of the investigating, reasoning, and being judged happened in the real world, as players of the four-year-long episodic series bashed their brains against one another in competition to find the truth (or the heart) of the tale. The game itself only polls the reader in the final, eighth episode, but the community had been so invested in scrutinizing one another's theories that the truth was relatively clear to most folks by the end of the fifth episode. It took until the seventh to convince just about everyone, but in the end, the theory that eventually "won" had been proposed at least as early as the third episode.
That debate could continue for so long, and yet the answer be found so early, is testament to how well-crafted the mystery was and how engaging mystery-solving can be even without game mechanics backing it up.
Another example with a similar idea, though a less meta implementation, is Banshee's Last Cry. It uses a standard visual novel format with branching paths, but at one key juncture early on, you're given the option to accuse someone of the murder by typing in their name. It's a simple trick, very reminiscent of certain orthodox mystery novels --
you have to type the name of the first victim, as it turns out that person was being impersonated by the killer and actually had died before the game began.
To solve it, the player must take into account little bits of information from across the game to come up with a theory in their own brain that none of the characters, the protagonist included, ever proposes. For that reason, its design philosophy has been stated as "a visual novel without flags" -- the idea there being, the game will not track which routes you've seen and what the main character therefore knows, so the player has to use their own memory and intuition to put the facts together.
I have high praise for the intensity of Nonstop Debate, but the actual bullets are nothing more than multiple-choice answers -- and it's made worse by the fact that each debate doesn't even allow you to choose from all your options, since only a few semi-relevant bullets are loaded each time. Essentially, it takes the Phoenix Wright cross-examination system, but trades off most of the actual thinking for more aesthetic spice. It feels great, but it's not particularly detective-y.
What do you find particularly notable about truth bullets in particular?
Unlike the rest of you i find this video disappointing, usually GMT deeply analizes game mechanics and tells why they work or why they don't, what are all the consequences of a design choice, why all games should have some feature and so on, this video on the other hand sounds more like a wishlist, it's nice to want things but there's a reason why big games don't leave everything in the hand of players, they need to appeal to a wide audience hence those games can't be too obscure and complex, honestly i was expecting from the video some kind of solution or compromise.
When i watch GMT videos i always learn something new and sometimes my mind is blown too, but nothing said in the video sounded to me particulary new or smart or whatever unlike all the other videos,
I think Mark did a decent job of articulating why certain mechanics work for him and others don't. Multiple-choice answers, leading questions, tiny lists telling you which items/facts/locations are relevant, etc. are said to take the act of proper deduction away from the player. When he then goes on to list examples of games with more open-ended response systems and less-obvious clue trails, that is his proposal of a "solution".
As already mentioned, this doesn't solve the problem of mainstream appeal, but I don't think that problem needs solving, necessarily. Some indies have demonstrated reasonable success already, after all.
I'm a big fan of the Game Makers Toolkit youtube channel but he definitly forgot Blade Runner.
It fits that category perfectly and the biggest advantage of Blade Runner is the non linearity. You can't make a wrong deduction. The story adapts to the choices you make.
It's one of the best games ever made.
Absolutely! That was a pretty huge omission, come to think of it. The video never comments on the entire sub-set of detective games in which the killer is not set from the beginning, really; while I'm not a great fan of most implementations, some games, like Blade Runner, manage it quite well, and it's worth further exploration.
Just watched this video yesterday, another top analysis from Mark Brown and glad to see him getting some love here.
For folk who like his stuff, I suggest the Youtube channel "Writing on Games"
Mark Brown really needs to play a few Tex Murphy games. I'm surprised they're not referenced considering he talks about Contradiction (another FMV game). I suppose they're too old for some people but few games do a better job with the private eye thing.
If you're interested, Under A Killing Moon and The Pandora Directive are both great. Overseer and Tesla Effect are also worth playing.
edit: I haven't watched the video yet. Too long to watch now. Basing my post on the list of games covered.
Mark Brown really needs to play a few Tex Murphy games. I'm surprised they're not referenced considering he talks about Contradiction (another FMV game). I suppose they're too old for some people but few games do a better job with the private eye thing.
If you're interested, Under A Killing Moon and The Pandora Directive are both great. Overseer and Tesla Effect are also worth playing.
He does talk about the Sherlock Holmes: Consulting Detective series, which were early 90s FMV games roughly contemporary with Tex Murphy. (So, age probably wasn't a factor for Mark, at least.) With so many great titles to talk about already, even more games had to be left out for lack of time, I guess.
Unlike the rest of you i find this video disappointing, usually GMT deeply analizes game mechanics and tells why they work or why they don't, what are all the consequences of a design choice, why all games should have some feature and so on, this video on the other hand sounds more like a wishlist, it's nice to want things but there's a reason why big games don't leave everything in the hand of players, they need to appeal to a wide audience hence those games can't be too obscure and complex, honestly i was expecting from the video some kind of solution or compromise.
Why does there need to be a compromise? A detective game that doesn't handhold, that touches on all the ideal tenets of being a detective, is already a super niche proposition, so why spend time considering compromises for what big games would have to do?
It's a wonderful video and it points out the glaring problems in most detective games really well, and it's something I've been thinking about lately as well.
The big problems in detective games are often related to how to simulate the deductive process of the player into effective gameplay elements. The more diverse and interactive the thought process is presented as, the more effective the detective gameplay should turn out to be. That's why text parsers and highly detailed and numerous interactive bits of information (like Papers Please) work really well.
Papers Please is an especially good example, because its universe is condensed in a single booth, with very tightly defined and strict rules, and pretty much every important bit of information is clickable/interactive. So as I mentioned before, the deductive process of the player, the thought process and linguistic elements are represented as clickable, interactive pieces, so similarly to Detective Grimoire (or a text parser, as a most extreme example), the player is basically writing down sentences, which is the most natural way of conveying your thoughts. Also the great thing about Papers Please is that the setting works really well with the gameplay mechanics, because you're doing the job of a customs officer, an office worker, and it kinda feels like doing a simplified version of a desk job, but in a fun way. Detective work is also mostly a desk job, and it's no surprise that Paper Please kinda feels like a detective game done well.
That's still a really tough problem to crack in games with an interactive narrative in general, and the advances in natural speech/writing recognition and AI bots will go a long way in improving this aspect of video games.
Another thing I think detective games also often miss to address is that detective stories revolve around there being exactly that - trained detectives (or at the very least highly intelligent characters with above average deductive abilities), which the players themselves often aren't. Following that, the "good" detective games also rely on the players having to remember a lot of things, and even if equipped with an ingame register of names and events, they still often require the player to either juggle them around in their head or use a real pen and paper to summarize things and organize information for easier viewing, which makes the whole thing even more tedious and difficult for a lot of people (which is why often games go to the other end and make it all trivial).
An important part of detective games should be the training, as a much more extended "tutorial" that, in the optimal case, should be non-intrusive and not obvious. Papers Please eases you into the mechanics day by day, and introduces increased complexities every few days, so it keeps the pressure on the player to not fumble and make mistakes, but also gives them time to learn and practice just by playing the game.
That kind of approach, combined with a very simplistic but functional information processing interface for marking and remembering any word or sentence ever spoken or written by anyone, and having the tools to write and combine your own virtual notes could be a nice match as a decent start for a good detective game.
The GMT video mentions games that present these clues and make you connect the dots in the real world by "doing the math" in your head (or a piece of paper), but I've never felt very comfortable with doing that, even as a kid playing Elite (making trading tables for different star systems) or drawing level layouts for large, multi screen adventures. It was kinda fun as a kid, but now it feels like the game is forcing me to use a pen and paper to write a letter when there's a perfectly working word processor on my PC, which I'm much more efficient and accustomed to at this point. I'm not saying every detective game needs word processors, but they do need more advanced UIs for organizing and interacting with all that data, because it would definitely make for a better experience.
So yeah, I very much agree with everything said in the video, I'd just add that games should teach us how to be better detectives and give us the tools to organize and use all of that data in a more efficient and fun way.
