Major links / lists of articles

Pentadact.com ← Tom's website/blog
PC Gamer - Articles by Tom Francis
Rock Paper Shotgun - What Works and Why ← A series by Tom, started 2018.01.30

Articles / Videos

2013.02.26 - Gunpoint Dev Log: A Broken Mess To A Working Game
- He set a goal to play through the game to get 'par' times for each level.
- He didn't expect how broken it would all be because of all of the new features he had added.
- It took 6-7 working days to fix everything.
- 3:24 - It was very draining to do all of this on top of his normal day job.
- 3:45 - He's been working 14 hours a day.
- 3:55 - It's mind-meltingly complex.
- 4:35 - It's also draining to have something that was supposed to take 1 day take 7 days instead.
- 7:00 - He gives an anecdote from the day before, where he was stuck on a particular bug for 2.5 hours, was very frustrated, decided to sleep on it, and when he woke up the next day and started trying to fix the bug he was able to fix it in 5 minutes. He thought, "Oh, it's probably that", and the bug was fixed.
- 8:05 - He had another task that he thought was going to be really hard which turned out to be really easy.
- 8:25 - After having two tasks take way longer than he estimated, he was feeling daunted and afraid at the amount of work left, but once he moved on and he had some other task take much less time than he had had estimated, he felt much more optimistic.
- 9:30 - He describes how he has assigned certain days to certain tasks.
2013.03.27 - Gunpoint Dev Log: I am now an independent game developer for three months
- 2:55 - Making the game wasn't taking too much of a toll on him until he tried to finish it, and realized how much work was left to do.
- 3:07 - "The amount of work I'm fine with, I've actually, I've sort of made my peace with the fact that games take a long time to finish. And I'm not at all bored of Gunpoint, I really like working on it. I'm really excited about finishing the thing. The only time it becomes at a negative thing is when I look at what I've got left to do and I can't see how it's going to cram into the time that...like, I don't have a hard deadline, but I always have a mental deadline. Like, 'It can't take another year...it can't take another YEAR.'"
- 3:40 - "I never actually feel like, 'Oh, god, this next thing I've got to do is just boring.'...There are boring tasks in it, but when the reason I want to finish Gunpoint is that's what I want to do (...) The really big thing pressing on me every day is, I want to know if I really can do this, as a job."
- 5:19 - "If I start to assume Gunpoint is going to do well, there's no good that can come from that. Because then it either meets my expectations or it doesn't."
- 8:00 - "The only unpleasant thing about doing Gunpoint is [that I don't get weekends off]."
2014.01.25 - Game design: the non-stick plan
- 0:40 - He does most of his design work in text files and Google docs, and Dropbox. He says 'They're a bit of a mess'.
- 1:00 - He does have a plan, and says a lot of indie developers say you shouldn't have a design doc, because they restrict your thinking.
- 1:27 - The more-traditional way of making games (used by big studios) is to have a very rigid plan to allow large groups of people to work together efficiently.
- 1:40 - His approach is half of each: "have a plan, and don't stick to it."
- 1:55 - "I'm not that fast at programming, and I'm struggling with a lot of stuff–it's all quite new to me–so I don't have time to make things that, if I just thought about it a bit more, I would have realized weren't going to be good design-wise, they weren't going to be fun. Although you can never truly know how something's going to work out just in your head."
- 2:50 - Usually a game starts with a particular idea or experience he wants to capture. Gunpoint started with Deus Ex, and his love of breaking into places that he shouldn't be.
- 3:55 - He goes on a tangent for a few minutes about how Heat Signature has some central mechanic that he isn't ready to show yet.
- 6:40 - When he has gaps in his design, he tries to figure out what the problem is, and then what the possible solutions are.
- He goes on for a few minutes about things he'd been thinking about when designing Heat Signature.
- 13:40 - It's preferable if the gaps in your plan have a variety of different ways that it can be tackled. So you can make that part of the design really complicated or really simple or anywhere in-between.
2015.03.06 - GDC - Lessons Learned Making Gunpoint Quickly Without Going Mad
- Alternative link: http://www.gdcvault.com/play/1021949/Efficiency-for-Game-Designers-Lessons
- The fourth thing I learned: The best place to save time is in the concept.
- 37:00 - The fifth thing I learned: How to stop being a perfectionist, or "Fuck it, that'll do"
  - Each time you come back to work, ask: "Is this the most important thing left to do?"
  - "The main thing is, every time I come back to work, I just ask myself, 'Is this thing I'm working on the most important thing left to do?' 'If I can only do one more thing before it just gets cast out into the world without any further input from me, is this what I would be working on?' And 90% of the time the answer is, 'No, there's something much, much more important I should be working on.' And so I keep a to-do list that's in priority order–as best I can–I try to put the most-important things next. And I constantly reconsider it, and I also have to use mental tricks to get over the pain of having to cut features. Like, when you put things in priority order, then what you really need to be doing is, when adding a big new idea to your game, you should be cutting something out, right? You've only got a certain amount of time, you've probably planned it reasonably well so that it's not going to take forever, and if you're gonna add something major you should take something away. But, emotionally, I can't do that. I'm too attached to everything that's already there, and I don't want to not write down the great idea I just had. So instead I just try and put them in priority order. And it's kind of a trick I learned from writing about games: I would often be way over the word count, and I couldn't bring myself to cut out whole paragraphs, so I'd just go through it and indent the paragraphs that weren't quite as good as the other ones, and once you do that, and you realize–you look at the word count, and you realize, 'All those ones I indented equal the number of words I've gone over, so if I cut them all, right now, then I'm done; the whole thing just fits and it's fine. Suddenly it's much easier to make that call. Whereas if you're going through and committing to deleting it right then and there, it's a nightmare. So if you have your to-do list in priority order, if you try and put all of your ideas for games in order of, 'Which is the most important thing not yet done?', then the great thing about that is that you don't ever have to cut anything, you can just pretend you're going to do it all: 'It's all in the list, I'm going to get to it one day!' And then when you're halfway down that list and you've done the most important half of those things, it's really easy to cut the rest because it's been three fucking years! <laughter> You're just like, 'I need to fucking release this!' So that's what I did with Gunpoint; the to-do list for that is twice as long as the things I actually did, I just got to the point where I'm like, 'Agh, I just gotta get it out there, I don't care about this goddamn camera bug."
2015.07.20 - What it's like showing Heat Signature to the press
- 1:50 - Turn off the tips, because the players aren't going to actually read them. You're just going to explain stuff yourself.
- 2:50 - He finds that watching press play is different from watching non-press play.
- 3:05 - He takes his game to 'Rezzed' and 'EGX' (conferences?) just to watch people play. He takes notes on what they get stuck on, what doesn't work. But he doesn't think about what features it needs.
- 3:40 - When you're showing the game to the press, all that matters is, 'Did something cool happen? And if not, shouldn't it have?' And, 'How can I make it happen?'
- 4:15 - If he's showing a game to someone who has already seen it, and he shows a trailer that shows all of the new features he's added since he last showed it, then what will the press person get from playing it?
- 4:40 - He spent time talking with someone from RPS who he knew, and they were talking about whether there was any change to the game that was substantial enough to warrant an article.
- 6:00 - One of the ways you can improve a game is to make it feel nicer to play, but the press won't be able to write about that. They write about features. The readers won't care to hear that the game feels nicer. His point: "What previews well and what makes a good story aren't necessarily the same thing."
- 8:50 - It's sort of rude to ask a journalist if they like the game because it puts them in an awkward position.
- 9:30 - There was a feature he added (re: strategy) that he didn't feel like the journalists really needed to play through, because they could just get the idea from the trailer.
- 11:35 - What he wants to see is that when he is having someone demo the game and he stops them from playing, the person doesn't want to stop playing.
2015.09.03 - Reading Heat Signature alpha feedback
- 0:10 - With Gunpoint he was sending it to relative strangers within the first week to get feedback. It was 5 people at first and got up to ~2,000 by the end. He had 15,500 sign up to test it.
- 0:30 - With Heat Signature he's just done it now, 18 months into development. He had a two week alpha for people who bought the 'exclusive edition' of Gunpoint.
- 0:50 - He set up a Google form where he asked the testers questions. He did the same thing with Gunpoint.
- 1:20 - What he's been doing with Heat Signature has been taking it to events. With Gunpoint he didn't have the time or money to go to those events. This is great for seeing usability problems / what people don't understand.
- 1:55 - The problem with getting feedback in-person is that people may not be as critical as they would be if they were submitting the feedback anonymously. People are going to 'underexpress' what they really think.
- 2:50 - He doesn't ask people for a list of problems; he asks people: "What is the biggest problem with the game?" His reasoning is that if you ask enough people, each of the biggest problems will be at least one tester's biggest problem.
- 3:20 - He's trying to get the testers to do what he used to do as a games journalist: figure out why certain parts of the game did or didn't work.
- 3:30 - He asks the testers to give him a score, as if they were game journalists.
- 3:50 - He read bug reports so that he could try to fix the alpha as it went, but he hasn't read people's opinions yet because he's scared to.
- 4:00 - He admits he is not thick-skinned.
- 4:10 - He's reading it on camera because it is activating the social part of his brain, which changes how he'll react to it.
- 4:30 - With Gunpoint, nobody could agree with what was wrong with it. Every aspect of the game had someone who thought it was the biggest problem, and it made him feel like he'd done everything wrong. He spent three days in a depression where he couldn't do anything.
- 6:00 - Another question he asks is, "If you haven't already covered it above, tell me what you think of the game in general."
- 6:10 - The lowest review said "It's a 3.14 out of 10 so far". It ends up being a very complimentary review.
- 7:10 - His approach to working is to do whatever the game most needs next.
- 8:50 - 250 people responded, 200 scored it. He has 1 '3' score, 2 '4' scores. 14 '10' scores
- 10:40 - One of the '4' scores says "This looks like a great project and I'm interested to see how it progresses". So it wasn't at all brutal.
- 12:15 - He finally reads an answer to the question "What's the biggest problem": "It quickly gets repetitive and boarding feels too easy. Everything feels too easy. (...) Guards react too quickly."
- 14:00 - He explains that he hasn't yet implemented a way to scale the difficulty at an appropriate pace.
- 15:50 - His policy with receiving criticism for Gunpoint was that he would act on it if more than one person said it, or if one person said it and he agreed.
- 17:35 - He said to testers that if they gave him feedback they would get access to future tests.
- 18:00 - The average review is 7.7/10
- 18:30 - He has a tendency to think that the game just needs 'this one more thing' to make it 'twice as good'.
- 19:10 - He got feedback from another 4/10 who said "it feels more like a simulation than a game", which is feedback he's heard before. Tom thinks they're referring to having a particular story. Tom agrees the game might feel like a "sandbox".
- 21:20 - To explain his vision he draws an analogy to a mod for Skyrim, where it randomly assigns you to a particular person in a big open world.
- 23:30 - He reads another negative review that says the game feels like it's missing content.
- 23:50 - The Google form also has a 'Suggestions' box / question.
- 24:00 - The negative reviewer above said he wished the game had characters and a central storyline, more interactions with space stations, more kinds of upgrades for the ship, more weapons / gadgets for the player, the ability to fight with bare hands, guards being less like floating circles that wander around aimlessly, more visual distinction between the different factions' ships' exteriors, space police
- 26:35 - He reads some complimentary feedback.
- 26:50 - The reasoning for having a limited alpha is that he doesn't want the people who are most excited by the game to be sick of it by the time it's done. He's relying on the people who are most excited about it to be the ones who spread the word.
- 28:25 - One of the 9/10s said "Generally I like the gameplay, although it feels a bit shallow at the moment"
- 28:50 - He switches to another negative review (a 5/10). A negative reviewer felt like he wasn't progressing towards anything, and "once I got my first gun the enemy ceased to be a threat". "The Heat Signature concept should be improved upon, right now it doesn't impact the game much at all." "Stations and missions should be more diverse" "I didn't feel like there was a lot of emergent behavior between factions, like having them attack each other"
- He concludes by saying that making it a video really helped him to avoid getting depressed by the criticisms.
2016.12.07 - Heat Signature dev log: rambling about 3 things
- 0:35 He has a document called "Game inspiration snippets" which is filled with ideas that he's gotten from various games around the internet.
- 1:00 - One of his ideas in that document is "a game that is a framework into which I can put a story", "a game that serves as a vehicle for writing"
  - 1:30 - He says Gunpoint is kind of an example of this, because the story was always told between levels and had minimal impact on the levels.
  - 2:15 - Gunpoint wasn't ideal for this, though, because it was a series of handcrafted puzzles, and you could run out of puzzles before you run out of story.
  - 2:30 - One of the reasons Gunpoint is the length it is is that he didn't have any other ideas.
  - 3:00 - Heat Signature isn't great for that either because there's a lot of randomly-generated stuff.
  - 3:50 - He thinks he may try to make his next game support this.
  - 4:40 - So his idea is to be able to pass in a bunch of parameters to some wizard and get a level back.
  - 5:15 - He thinks he started thinking about this because he was trying to figure out how much story there should be in Heat Signature.
- 5:40 - Another idea of his is a game that's a vehicle for mechanics ideas.
  - In Gunpoint the various mechanics are generally just upgrades to the CrossLink or perks.
  - 6:30 - If he was designing Gunpoint again he would specify that certain items / perks could only be unlocked at certain times. The issue he faced when creating levels was that he didn't know what items the player would have.
  - 7:45 - He says he's not interested in hand-crafting puzzles ever again.
  - He spends a few minutes talking about how adding a bunch of different weapons / gadgets to Heat Signature really helped it feel more fun to play.
  - 11:45 - The game is slow on his laptop if he has a really big ship. But he's not too worried about that, since most players will have powerful computers.
  - 12:30 - He doesn't like to work sitting down, so he works from a standing desk with his laptop.
- 13:00 - What he's working on right now is mission generation. He spends a few minutes talking about how the missions work and various different ways of generating them that he's thinking about.
  - 14:30 - 5 minutes is a long time in Heat Signature. (ie the game is fast-paced)
- 22:00 - He thinks random generation is really good for supporting different playstyles.
2017.02.05 - Heat Signature dev log: new country, new office, new team
- 17:00 - end - He explains that he's just recently realized that the 'interior game' (ie what happens when you're in the spaceships) is really the core of the game, and that the spaceflight is really just a framing device for that core gameplay. He says he spent 2.5 years working on the framing device (the space stuff), and now has to focus on the gameplay for when you're in the spaceships.
- "at the end of the day I don't think it matters if I fuck that [the personal/final mission] up, or if I just do a lackluster job. I think that's a part of the game that can afford to be only OK. And the interior game cannot–it must be brilliant. And that's kind of like a revelation I've come to over...this far into Heat Signature, like way, way, way too late I've discovered that this whole idea of going inside spaceships is not the core of a game, it doesn't–by itself–sort of, form a good basis for a game, like, that's not what you're gonna...the actual moment you dock, it's cool, but that's not the core of what you do, right? It has to be what happens after you dock, or something else; it could have been the space game, or whatever. But the way it ended up is that the interior game is where all of my interesting ideas ended up being. And so what I've really made is a really fancy framing device for a top-down stealth game with, sort of, emergent elements and hopefully a very-wide toolset and interesting challenges. And that's cool; it's good news, 'cuz I can make that game, but I haven't made that game yet. I mean, I kind of have now, like since Stugen...Stugen 'til now I've been working on that game. But for the other two-and-a-half years I wasn't working on that game. I was working on the framing device for that game. So, it's a useful piece of information to have. And that's what [unintelligible]."

Browser not supported