The game has been released since August 7, so it’s time to share the steam stats and the analysis!
As you can see the number of units sold is quite low compared to the wishlist. I also have a high number of returned copies mainly due to the difficulty of the game and a bug present at the launch.
At the launch of the game, I had a health problem that required hospitalization so the press coverage of the game is non-existent except for Adventure Gamers.
The first review I had is negative so even if I have fixed the bug most people will not want to buy the game. Some people liked the game but did not leave a review and I don’t want to give the game for free in exchange for reviews.
If the game is too difficult, people can leave negative reviews I am open to it but I explicitly said the game is difficult in the description. If the game is bad or poorly made and it’s unfixable, in this case, negative reviews make sense.
Maybe the price of the game is too high $9.99, the average time of the playthrough is about 7 hours so it doesn’t seem so high to me. Other point-and-click adventure games are also in this price range between $5 and $15. On steam spy, the average price is $8.63.
For the time being, I am focusing my effort on my next game. For the next game, I will use Kickstarter to test the game viability before launch.
The game is still in beta testing, I have fixed a couple of important bugs crashing the game. During the beta-testing lot of testers find the game difficult because most of the time there are very little indications and little feedback from the game. Since the game is inspired by the game Myst (the first opus) this is no surprise. But it can be quite discouraging, so I have added a hint system to help players who want to progress in the game.
The first version of the hint system was not very useful because the hints were too general, I have tried to remake this but it can take a surprising amount of time. Because each time I have to make new triggers in the level to track where the player is and decide what the next step. The game is in an open world so most of the time there are multiple courses of action, but I tried to help the player find the most obvious one first.
A lot of testers quit the testing after the new year break, so I am mostly finishing the hint system and the game will probably be released shortly after that.
It’s been a while since I did not give any news. Hopefully, the game is not dead yet but the release date has been reported. I tried to playtest the game with six persons but most of them did not even finish the first level.
I have fixed a lot of bugs since then but I feel like the game still needs more playtesting to be complete.
For that, I have created a discord server maybe it will help to get feedback.
If you like point-and-click games and have time to playtest do not hesitate to ask for a steam key!
For the moment the game is only available on Steam.
I have finished the content of the game, the dialogs can still change a little bit but the art and the animation are done at this point.
I have also implemented localization for the game, for now, the game can be played in English or French. This took a little bit of time for the menu because everything was prerendered and saves as images.
The trailer and the marketing content is complete, you can see that on the steam page of the game.
I have also added game controller support for the game, it’s complete but there are still some bugs here and there.
The credits for the multiple endings is also done, now I am working hard on testing the game, I have made some AutoHotkey scripts to help me test and find out bugs.
I don’t have a release date yet because I feel like the game is still in beta and I still have a low number of people adding the game to their steam wishlist.
The development of the game is still going, I have redone some maps.
For example, before I had some maps with placeholder photographies :
Now I have drawn the trees to be more consistent with the overall style :
I have done more animations and added more content for some levels, the content is mostly done now maybe little things to tweak here and there and some dialogs to change.
On the programming side, the format used for the images are dxt1, dxt5 for the compressed format and tiff for the uncompressed format. It improved the loading and the memory consumption of the game by a lot!
I am working on a trailer and some marketing images for steam.
A lot of time have passed since the last update, I have done a lot of bugs fixing these past months, a lot of things at the beginning were broken, or some maps or objects where missing objects or textures.
The progress has been slower since I have less time to work on the game with a full-time job, I have maybe 3h/4h each day after work and more time on the weekend.
This month I have been working on new animations, one animation takes like one week to draw and maybe one week to scan, post-process and combine.
I would like to finish the game this year, but the game will be released when it’s done. I need to work on the marketing, I have so many things to do, each thing I do is less time for other things.
That’s why I have been focusing solely on polishing the game these past months.
The updates are less frequents because I am moving in another city, each night with the day job and the transport I have maybe two hours of free time. So the game is progressing very slowly these months. I have fixed some bugs and redone the water animation, synchronized the music. I need to retest the whole game because adding and modifying things can break some other parts.
In my new home I will have a lot more time, but no internet yet. I don’t know how much time I will have to wait to have the internet, maybe weeks. During this time I will work on the game as much as possible.
It’s becoming harder and harder to work on the game since I have a day job and it is hot every day. I have maybe three hours each day to work on the game and I am tired but I try to fix smalls bugs in the last level.
The other news is Steam Direct, I can publish the game directly on Steam after paying the fee but without a minimum of marketing, the game will sell very little. Steam will probably become like the google play store, so many games very cheap it becomes very difficult to make a living. It’s better that way, every developer has more freedom and the customer can choose between a lot of games.
I still don’t know how much time finishing the game will take, my first priority is to find the energy to work more on the game at night and find a flat near the job.
These last two week the third level has been completed, now I am working on the big level 4. A lot of bugs have been fixed, some of the art has been completed too. I have modified some dialogs and some collisions.
Every time I make some progress there is always more to do, it looks like the game will never be finished. I think I need to add a lot more dialogs to have a game not too empty.
I need to buy the music but for the moment there are still too many bugs the game is not polished enough to be fully playable.
Steam greenlight will disappear soon and maybe I will not have money to put the game on steam, in this case, I will release the game on itch.io first.
I post update less often because for maybe an entire week I have only fixed bugs. I can now sum up all the interesting bugs.
One interesting bug I have is:
OpenGL makes a border around sprites in a sprite sheet when the sprite is rescaled.
And the things is, this artifact is not a bug but depends on which filter algorithm is used by OpenGL. There are multiple but the two without mipmapping are GL_NEAREST and GL_LINEAR. Using nearest-neighbor interpolation produce an ugly result, so I need to use a linear interpolation for the filter.
The problem with linear interpolation is it compute the value of a point with the adjacents points, and sometimes with a certain size, position and ratio a border of one pixel appear around the sprite. This is when the linear interpolation algorithm needs to take the color of a transparent point. When I generate the sprite sheet with openCV the transparent color used is black (RGBA) : (0, 0, 0, 0). So the line around the sprite will be darker than the rest of the sprite.
The solution to this problem is simple once everything is explained, make a border of one pixel around the sprites with the good colors. I can do that easily by reducing the size of the sprite by 2 pixels.
Another classic bug with vector is the invalidated iterator after addition (when a reallocation is needed) or the deletion of an item.
In my case, the objects in the vector were stored on the stack, and I was using the address of the item as a pointer. Everything is fine if the vector is not modified, but I am inserting and deleting things while using pointer => pointer are invalidated => accessing invalid space in memory => game crash. The solution is simply to allocate thing on the heap to obtain a pointer that does not point on the vector and continue to be valid after modifying the vector.
There are a lot of others small bugs but when testing now I am using this to check for heap corruption (for visual studio only):