Dev Blog

(This week’s blog post arrives a day late because Tuesday was a national holiday. Happy birthday to 99-years-old Finland!)

Last week the soundtrack for Riptale got finished. The challenge with the combination of modern day audio tools and pixel art is restraining yourself from all the thrilling possibilities a plethora of plug-ins presents. We wanted to do something retro-sounding, but not quite chiptune.

Stylistically, the music was very much inspired by the soundtracks of early Mega Man games and melodic heavy metal. We were trying to keep the chiptune feel where each instrument could only play a single note at any given time. When it comes to the sounds, there’s of course plenty of directions you could go in. If chiptune is too simplistic and harsh, one might try stepping up a generation or two with the retro game synthesizers. But that was not the sound we were looking for.

The early game synths have a distinct sound to them, but suffer quite a bit from the limitations in the technology of the era. Game music has since grown out of the restricting synths of those days but the basic sounds found in chiptune music are still used to an extent in modern pieces. We were looking for something harsh that would convey the energy of distorted guitars found in heavy metal. We also wanted to sound a little bit more modern while still retaining the feel of FM synthesis that chiptune is known of. It was an interesting experiment but we are pleased with the results.

Like most everything in game development, creating the music for Riptale was an iterative process. Only a third of the themes that were worked on made their way on the final soundtrack. We wanted to make sure that the music actually fit the stages they were planted in. Some of the themes might work in another game, but we felt that they had no place in this title. Could they be recycled? It’s possible… but unlikely.

Dracula is dead! Symphony of the Night can be challenging at times, but compared to something like Order of Ecclesia, it’s still pretty easy. You feel kinda cheap using the Shield Rod + Alucard Shield combo to defeat some of the harder bosses, though.

What is so interesting about Symphony of the Night? Why is it still such a great experience that it’s worth going through? The game really shines with its content. There are so many bosses, enemies, items, weapons, and locations that you’re just curious to see what’s around the next corner. Granted, you’ll never use even half of the items that are handed to you and half of the bosses are a bit of a joke, but if they weren’t there, the game wouldn’t have the same impact. The gameplay isn’t carefully crafted by any stretch of the imagination, but there’s enough of interesting content to keep you going for 11 hours.

So, does quantity over quality work? The content in Symphony of the Night isn’t really low quality. It’s just badly balanced. The art of the game is breathtaking and the enemies you encounter are often clever and interesting. The amount of secrets helps, too. When you know all the breakable walls, spell combos, combinations of items, and tactics to beat certain bosses, it’s a fun feeling.

What’s the takeaway for a modern roguelite project? You can’t compete with the quality of hand-crafted levels and challenges that game such as the first Castlevania or something like Super Mario Bros. 3 offers. One clever way to keep players hooked is to craft enough differing content and offer them challenges in form of secrets. Have the game offer more than is immediately obvious at a surface level. Hand out easy secrets to the players so that they’ll know there might be something else cool hidden in there. Don’t let the sense of intrigue die off. That can take some careful planning, too, but it’s really a tradeoff. You have to perfect something.

The idea with Kumo was always that he’d be wearing a long scarf that’s wrapped around his head and flows in the air current as he flies through the caverns. You know, like Rambo doing ribbon gymnastics. However, the problem with a visual element that’s this elaborate is that it’s not very high up on the priority list.

But how useless is a visual element like a flowing scarf to the end product? Think about it for a second. Visuals can sell games. It’s just that pretty much everything in the 2D end of things has been done already. Coming up with new ideas is tough. And sure, there are plenty of 2D cloth simulators out there, some that probably even look better than what we currently have in our game. The scarf is still a very striking visual element in every animation and screenshot. It might make people remember what the game is. That is not a small feat.

So, the scarf is now in the game. It took a lot of time to implement, but the character has finally got his destined identity. (That was a weird prophecy, anyway.)

Other than that, the team has redone the character attack animations, effects, and a good deal of blood spatters. We are trying to make the game look as presentable as possible for the Steam Greenlight campaign which means that adding new things is on the back-burner for a while as we concentrate on refining what’s already there.

We finally got to the flipped castle in Symphony of the Night. It’s nice to play some old classics and see what they did right. (And wrong.) Meanwhile, we’ve been reworking a whole lot of animations for the game.

The main character was already altered once after the Ludum Dare demo, but the project lead felt like he still looked too scrawny. So, throw everything to the bin and try again. The earlier this sort of problems can be identified, the cheaper it is. Sometimes it just happens. You work closely with the project for months and one day realize that you need to give the game a new identity. If it pays off, it’s worth it.

As a startup, we’re trying get by with what we can get our hands on. One of the animators has a really low desk. We decided it would improve the work ergonomics if we just stuck one of the monitor boxes underneath the tablet. And it does help immensely.

Steam Greenlight is getting closer, but we’re not yet completely prepared for it, yet. This means we have to keep pushing certain deadlines further away. A lot of it comes down to the lack of experience, but some of it is due to the chaotic nature of the project. We feel good about the state of the game and the next one will surely go down more smoothly. For now, however, we’re just trying to create an awesome vertical slice.

The last week has been us grinding towards the end product. The game still has a lot of unfinished pieces that need to be put together. It is, however, important to occasionally play some old classics  to see what has been done well in some highly valued titles.

Everything is still very much unfinished. The levels need more work, the game needs more enemies, and the gameplay is not on par with what we want to have in our demo. It is just very difficult to get anything done with such a small team. Things take longer than what one might expect and it is easy get frustrated with the seeming lack of progress.

The good news are that there have been a great deal of advancements on every department. The art team just released a concept picture of an upcoming boss. The programmers have been working on particle effects and adding in new attacks.  For better or for worse, the game is starting to take shape.

The past week we’ve been planning for our Greenlight demo. We need new levels but don’t want to make it too big for multitude of reasons. With a smaller demo it’s easier to make it more focused and concentrate on things that work and things that need improvement. Still, the mechanics that go into the demo need to be polished enough. We want the word of mouth to be positive.

New demo means new attacks, new enemies, and new levels. Mechanically a lot has been done already but tweaking everything to perfection is a time-consuming process. If there’s one thing we really want to nail, it’s the player controls. The attacks of the main character, Kumo, are our main focus. There’s not a lot of time, but we’ll do what we can.

The new levels depict a mushroom cavern filled with ruined castles and other structures. We are trying to make the levels more open and concentrate less on precision platforming. The game is about doing cool attack combos, not trying to execute tricky jumps to a tee.

Time is always against us. As game developers we want to make great things for people to enjoy but how much time can great things take? Can we afford to go overtime if the difference in end product is like between night and day?

The usual situation is that game designers have a metric ton of ideas for content and mechanics of the game and in general everyone in the team wants to see all of those things added in. The problem is that typically implementing each feature takes three times the amount of hours allocated to the task. And that is not a small problem. Optimism kills projects and even if you had the time to add in everything you wanted, the game might end up a mess as a result.

Assets are usually not an issue. If the artists draw something, there might be room for improvements, but you’ll at least have something you can use in the end product. The real time sink are bugs in the game code. Bugs are unpredictable. Bugs are puzzles that need to be solved for the project to go forward. A simple bug can take anywhere from ten minutes to ten days to fix, because the cause could be anything. Often you can’t leave them in because they can turn your customers into an angry mob. Sometimes it might be a valid strategy to just seek for an alternate implementation for whatever is currently bugging out.

Good games have been crafted carefully. Tweaking everything until it’s just right is almost always a very time-consuming process. But it’s necessary. So, at what point do you stop adding in features and start honing them? It can be tricky. You want to be proud of the end product. You want to have fun while playing your own game. You just need to concentrate on the right things while you’re developing it.

Oulu Game Day was a success. Not that it had any right to be given the amount of time we had to allocate to testing. We got lucky. There were a couple a game-breaking bugs, but fortunately they didn’t occur too often. The game did freeze a couple of times, which is worrisome, since a bug like that can be a nightmare to track down in GameMaker: Studio.

It was great to see how people played the game. Many issues regarding the game balance surfaced during the event. Some of them were due to the lack of preparation time. Some were genuine problems we now have to evaluate and react to in one way or the other.

The game is coming along and having a deadline really made a difference. Sure, you can make imaginary deadlines to your scrum boards and really try your hardest to not fail. However, when you have to go present your game to other human beings, the idea of failing starts to become more terrifying (or if you have pride in your work, just plain unacceptable). Realizing this, we started to look into another events that would give us an actual deadline to work towards. And we did find one.

Riptale being our first game is the chance for us to show what we’ve got. When we have really established ourselves, we can probably ease with the deadlines a little bit, but now is the time to work hard and put the name of our studio to the map. In the simplest of terms, a successful game will allow us to make another one.

Next Friday we’re attending a local event called Oulu Game Day. Granted, Oulu isn’t a big city, but a change to get any feedback on our game is one that we don’t want to miss. You can always have people on the Internet tell you what’s good and what’s bad, but if you want to get a some nonverbal feedback, you have to see people as they are playing your game. The only question is whether or not we can get the demo in a good enough shape to be able to receive any useful feedback. I.e. you don’t need people telling you things you already know. The really valuable part is observing what things need to be balanced out more.

So, what is still so broken that it can’t be fixed for Friday? Our whiteboard says it all.

The one big hurdle so far has been the level generation. Due to the unholy union of Tiled and GameMaker: Studio, we’ve had to simplify our designs time and again to be able to get content out fast enough. Here’s the nitty gritty:

GameMaker doesn’t natively support Tiled TMX format, but it does support binary files. Hence, Mr. Mikael “Mick” Norrgård at GamePhase wrote a TMX to binary converter to make it possible for these two pieces of software to be used in harmony. Much obliged. The problem is that the converter doesn’t save all of the level data into the output file. It does plenty, but even so, we’ve had to come up with some interesting solutions to get GameMaker to generate the kind of dungeons we want for our game. Is it a huge problem? Not really, if we weren’t in a hurry.

There’s no doubt that we’ll be able to get all the functionality in the game by Friday. The scary part is that we won’t have time to test it properly. The best thing we can do in preparation is to dust off our humble hats and get ready to apologize a lot. Then again, it might go swellingly.

New sprint is upon us and we’ve tried to slalom around painters and construction workers who are working hard to make the building look as pretty as humanly possible. In the meanwhile, we’ve been doing our own construction work, trying to get the level generation working properly. The autumn mornings might be getting cold, but the drive to get Riptale finished is only getting stronger.

Not surprisingly, a lot of time has vanished into this site called HacknPlan. It’s a project management tool that offers the same functionality as something like Trello, but with a higher focus on game architecture and all the things that are required to have a successful game project. We’re impressed with how much this site has evolved during the past year. Still, even with all the best tools, dividing the project into task cards takes a good amount of time.

Scrumming is a lot of work. Meetings, planning, tasks, and assessments can take a lot of time. However, if that makes the completion of the project 10% more likely, it’s worth it. Game projects have a bad tendency to not get finished. Having a board of tasks in front of you also makes you more efficient because you always know what you should be doing.

Other than that, one of the programmers threw all of the GameMaker files to Visual Studio. With C++ syntax highlighting the setup is remarkably workable. The only thing that would have made it better (that we’ll know the next time around) is if all the game code was in script files and just called from where needed. The object files can be a bit messy with all the XML interrupting the otherwise clear and concise syntax.

Super God as a company is quite young but we’re already going through a good amount of changes. First off, we’ve recently switched our primary game concept from a relatively large platformer to a more manageable one. Secondly, we moved from our hand-crafted C++ engine to GameMaker: Studio. Finally, our programming department went through some drastic changes and the new people have been familiarizing themselves with the new software.

Changes are stressful and that’s not a joke. Some people work well under stress but that hardly applies to everyone. If things are constantly changing it can signal to people that what they’re doing doesn’t matter. Now, it would appear that we have finally found a path that we can lock into and we’re feeling positive about our current project: Riptale.

While the programming department is still finding their feet with GameMaker, we also found people who can help us out with the graphics and the social media side of things. We do believe that we’re making a great product, which of course is essential no matter the field. People want to know what you’re doing and you want to tell them. This type of communication just eats up a lot of man-hours. A dedicated social media person can make all the difference.

For some reason offices with a lot of furniture seem to be more comfortable for creative people. Posters, furniture, and other random stuff lying around everywhere can inspire already artistic personalities to rise to even greater heights. Then again, if you are a programmer, an office like that might be your nightmare.

This is our office. It’s a bloody mess, but things will find their places… eventually. At least we can work here. A lot of concept art is getting done. Here’s a picture of a few weapon doodles.