Post Mortem ; Teamp flytrap’s Umibozu

We been working with the game concept Umibozu for now two months and much has happened during the development of it. We have both had ups and downs during this period but we have pushed through and managed to make a game. A playable one at that. I am happy and proud of this simple fact because this is my first game and the first game for the rest of my team aswell.

I have learnt a lot in a very short period of time. I have learnt to use photoshop more efficiently but I still have a long way to go before I master this software. I have also learnt to make animations, a UI, GUI, HUD, spritesheets and the process of iteration. Indeed, most of the assets had a starting state and went through a couple of changes before getting the final approval.

Most importantly, I have realised we should always playtest our game and question our concept and our mechanics, which I know we have not done enough in my group.
Our end result is playable but somehow mediocre and I think that is mainly due to the fact we didnt question the concept enough and try to fix or improve the mechanics that didnt work or didnt add anything to the game.

On top of that, I have realised that a game is only playable and enjoyable if it gives both sound and visual feedback to the player. Indeed, that is the only way the player learns about the challenges and actions offered by the game.

In general I am happy with the art style of our game because most testers have given us praises and positive feedback on it. What im generally disappointed with is that our art and our mechanice might not have fit together and we didnt manage to communicate well the challenges and actions to the players. In fact, many didnt understand how to win, what made them die or how the powerup works. These problems could have been avoided with a better communication within our team and more meetings.


My goal is to improve my productivity and the quality of my art aswell as my communication during the next project.



Supplementation on the end result :

Our game is an exploration type of game in which the player roams the foggy waters in order to find and save a fisherman that has disappeared at sea. The player fights off the creatures coming at her in the fog. We have got 3 different ennemies and the boss. They come towards the boat in different patterns and speed. The player can either try to avoid them or use his harpoon to kill them. We have 3 pick ups : one to repair the boat, one to increase the life on top of the boat and one to increase the speed. The 2 boost effects only last for a few seconds. We made the pick ups look like bright shining flowers because of the traditional chinese art in which one can often see lotus flowers.

The aesthetic goal “mystery” was supported by the story of the fisherman going at sea to discover the mythical creature. It was also supported mechanics wise by the ennemies being hidden in the fog until the light of the boat touches them. The same mechanic was used for the pick ups.  Finally, the mythical creature keeps a shadow form even when the boat comes across it and its attacks are underwater aswell as placed randomly so the player can not predict what will happen next.

On the top left corner the HUD in form of a parchemin with 5 boats drawn on top of it, lets the player know how many lives she has left. Being hit byt an ennemie or an obstacle results in losing one life.
The ending boss, mythical creature Umibozu, appears as a dark black shadow with red eyes, the music changes from a calm melody to an upbeat challenging one. The umibozu has two attacks, one of which is hitting with tentacles while the other one is whirlpools.
Both attacks have  visual effects before they happen so the player can anticipate them and avoid them.

In conclusion, I am proud to say that we have a playable game even if the balance on the pick ups and the ennemie spawning does not feel right. People have praised the black and white look of the game but mentioned a lack of visual feedback to understand the pickups effects. Going forward, I will keep in mind that balancing, visual information and feedback aswell as communication within the team are critical in game development.







First playtest :

The first playtest for the alpha was an important milestone for our group as we had been putting on hold design choises and art syle choice until we would get proper feedback.
When it comes to the art style, we have found that most people enjoyed the black, white and grey palette and found the idea of playing on the pages of a book interesting and that it set our game apart from the other umibozu groups.
From the point I then could take my old assets and color them with grey watercolor and move on to other new assets. From this feedback we also decided to keep the GUI black and white.

The other very important feedback we got was that players didnt realise the crates and barrels were pick ups/power ups and that they were not obstacles to be avoided. That came from the color of them and there non attractive design.

From that feedback I changed the power up to a new design, with flowers to keep the “ink traditional chinese” drawing and I added the icons of the effect of the power ups in the middle of the flower along with some color. It was the first color we used in the game and I decided to keep it yellowish to fit with the light of the boat and the lantern of the GUI.

Here is the change I made to the pick ups :journal

Second playtest :

The second playtest was helpful in a way that it gave us feedback and QA. Our art style remained appreciated by the players except for the fog shapes that seemed odd to many people in the way it moved and in the way it looked.

We decided to try making it lighter to fit in with the particle effects made in unity and to add a paper texture on it so it would look like it was within the background which is a paper tile and then it looked like this :

fog 1.png


But apparently from our programmers feedback it looked too bright and didnt fit in so we scraped the idea

We received feedback that our music was good and immersive.


Playtesting was important to me in order to know which asset and style to scrap and if the direction I was taking was the right one. In term of design we received a lot of feedback that helped us in taking decisions regarding ; our ennemie movement pattern, the goal of the game was now improved to not only discovering a creature but also saving someone from the said creature in order to give a purpose to the player ; the player movement seemed odd so we changed it.

One thing that has been brought up to us during feedback countless times and that we havent solved is that the enemies are very easy to differentiate from the pick ups and also there is no real reason why not to shoot all the appearing fogshapes because u can move on without the pick ups.
I would personnaly think that we need a longer cooldown on the harpoon so that the player can not spam shoot everything on the screen, I would also want to change fogshapes for simple patterns on the background tile so that the ennemies would look as if they were under the paper and they would be difficult to spot until the light was cast upon them.


Those are things I will bring up to our group but I think it might be too late into the process to make these changes.


Comment on Roberto’s animation post.

“Insight #4 – Lasers & Animation”

  1. hola,

    Interesting blogpost about the animation process of your behemoth. I have been lucky enough to witness your meetings and all the work that has been done on this game so Im glad to be commenting on your blog.
    Its interesting for me to read that people thought your laser looked like it was made of water because it was my first thought when I first saw it. On my part I thought the reason for that was the shape at the tip of the laser that made it look like water colliding onto something and splashing away. So i think I would have started with changing that tip shape into someting more square or angular as if the light of the laser was bouncing/refracting from a surface.
    However, the changes you made to the temperatures and the brightness in the middle part of the laser combined with the animation of the laser lighting up really made the trick and I think it looks great. There is a huge progression from the first laser you had in alpha compared to the one you have now!
    When it comes to the process of animating, most of us graphics students have encountered the same obstacles and difficulties. First with learning the theory and the ‘how to’ in photoshop and learning to communicate with the programmers and our groups about the process of animating. It is painfully time consuming when you want to achieve a certain standard.

    In the end, I think your laser and canon looks dramatically better now thanks to the color changes, the animation of the laser, the design change on the canon and the particles around it.

    Great job!


Sprint 6 : Game over screen and victory screen

This week I have been working on both the game over screen that will appear when the player loses and the victory screen that will appear when the player completes the game.

When it comes to the game over screen, it was quite simple for me to decide what should be on it because our concept document revolves around a mystical creature called -Umibozu. I decided that I would draw the sunken shipwreck being taken away by the creature.

The process for this image are the following :

  • Sketching the shipwreck and the umibozu outlines.
  • On another layer sketching background to give more depths to the image.
  • Try a color palette of greys.
  • Color the image on another layer, keeping the darkest for the foreground and umibozu and the lighter colors for the background. I am not sure that was the right way to do it, if i would recolor it now i might make the reverse decision.
  • I added some paper texture that I had used for the background tile of our game to keep the harmony with the game.
  • On top of that I used a sponge brush i downloaded online to create more darkness around the edges of the picture.

Here is the game overscreen :

umibozu ending game over

For the victory screen, I decided to draw the boat getting back safely to the harbor since the player starts the game by leaving it. I decided to go for a concept art and not keep the top view of the game. It gives a more ” narrative ” feelings to the game and also I did not want to reuse the harbor of the main menu because it would look lazy. In order to keep the “oriental ink on paper” style  I looked for references and decided to use the next image as a style guide :hand-drawn-japanese-landscape-background_23-2147560881

I first sketched the harbor to look a bit similar to the main menu one, I added mountains around it because there are a lot of cliffs in our game. I sketched the boat and the trees and vegetation.
Then I decided to keep the sketched lines and just erase the ones that were too messy and redraw some of them to look more polished.
I used a watercolor brush with different opacity to color the picture and finally I added paper texture to it so it would be similar to the other screen and the rest of the game.
Here is the result :

win screen

Sprint 5 – Scrum

All teams have been working within the scrum framework. It has been quite simple to implement thanks to Marcus’s template.

During our kick off meeting all of us worked on adding assets to the product backlog in the most organized way possible. We split the different assets into categories such as UI, Player character, Power ups and other pick ups, non interactive assets such as the background and decoration, Menu…etc. We then split them within our different minors : coding, art, design and pm.

Personaly,  I have found working in one week sprints efficient and it has kept me productive. I finish most of my assets on time for the review meetings and I am sometimes  able to add more assets into my sprint when things are going smoothly.

One of the difficulties was to meet on daily stand up meetings. We mostly are not meeting up for that and keep each other updated on slack or messenger. It has been difficult for our project manager to organize them because of the different schedules when every member is on campus and it is a lack of motivation to show up for a 5 minutes meeting that was problematic during the other days.

The part my team has struggled with the most is keeping the google sheets updated. Indeed, many of the assets have to be reworked and upgraded from the first week to the alpha and to the beta. It is not always clear for which milestone an asset is “implemented” or “ready to be tested”. I think the visualisation on google sheets isnt the best, it would be better with something more graphics with icons and symbol that would look like a huge board. Then it would be possible to grab one asset from one column to the other and everyone would be able to see the different steps the asset has been through.

However, another way to make it work is a strategy I have picked up from another team : They  to split the product backlog into 3 different product backlogs : alpha, beta, gold.



Sprint 4 – Polishing the harbor and player boat.

This week I have been tasked with polishing the habor for the starting page. Basically the idea behind it is that the harbor is the background for the main menu but also the starting location. Once the player hits play, he has control over the boat, gets indications on the controls and try the controls in the harbor before sailing away and starting his journey.

To the harbor I have added a city with vendors on the bottom so that it looks more aesthetically pleasing than only seeing the quay of the harbor. I also added a few boats because the harbor looked very empty.
I also tried different color palettes so my group can vote and we can ask other peoples opinion before making the final decision.

On top of the harbor, I had to redesign the player boat for different reasons. In total I have drawn 4 player boats since the first sprint planning and have wasted a lot of time. This was due to poor communication within the group. At first we had a engine power boat with very dark style, then we decided to go for a lighter no line art boat but still engine powered, finally we made a U-turn and our designer wanted the graphics to fit more with the idea of old chinese or japanese drawing, so that would mean drawing a woodden boat. I first drew a small one as i imagined the sailor to be a lonely man on his small rowing boat but it made it difficult to graphically fit an harpoon, a static light and a rotating light on the same boat.

Finally, I have come up with a bigger boat that ressembles an old woodden ship with a platform both on the back and on the front of it so that it will be easier to fit an harpoon with its animation and a rotating light with its animation.
I made the line art fit the harbor look and I used watercolor because we are planning on making the boat fade into a watercolor splash when he takes damages and when he dies from taking too many hits.

Here is a picture 4 boats i drew for the character player. The final one is the top right.

boats blogThis past sprint has made me painfully aware of how important it is to have good communication within the group and I would highly encourage anyone to write down design decisions regarding graphics, art style, brush tools, palettes etc in a document that everyone signs on so that artists don’t waste time drawing 4 times the same assets or colouring them 5 times until everyone agrees that looks best or change the concept of the game and everything has to be redrawn.

That’s a good lesson i learnt the hard way.


Sprint 3 – Menu and starting place.

This is currently the third sprint of our project. The Flytrap group is creating a game based on the concept document “Umibozu”. In this game, you are sailing dangerous waters in a foggy weather in order to find a mysterious entity. During the journey, you will have to avoid obstacles and enemies while collecting power ups.

On Monday we had our sprint planning meeting during which i decided I will be focusing on the menu background and the starting place of the game.

I am focusing on these two features at the same time because we have decided to use the same graphics for both. Indeed, when you first start the game, the main menu with the different game options/ start game/controls buttons will be displayed on the same background as when you start playing.

Since the player avatar is a boat the Menu and the start of the game are set in a harbor. I decided to draw different assets that make up a harbor seperately in different psd files and make them png files so i can later on add them in on same psd new file and place them the way i want.

Here is the document files with all my harbor assets as of today :

assets harbor blog

And this is the harbor I have put together as of today, under which I added a paper style background :

harbor blog post.png


I have yet to draw the boats that are in the harbor. All the assets are sketched and line art for now. The style we wish to go for is a game that plays on the pages of a book or on an old traditional paper and all the assets would look hand drawn on the pages of the book on which you are playing.
We think the idea is rather good and would differ from the other groups but we are painfully aware that the final look might not be what we are aiming for.
As of right now I have decided to stick to sketches and simple lines without colours so that all of my assets may be used after the alpha feedback. Wether we decide to continue with the style we first decided upon or to go for something different after the alpha, it will not affect the assets I have already produced as I can build upon them.