Mr Football beta week 2: Gold, green and titanium

The highlight of this week has been moving some Phaser scenes into gold status: specifically, the List and Livery screens. I can’t tell you how many iterations I have gone through with these pages, I have started at them and tweaked them on a near-weekly basis for literally years now. Lighting them up on my Roadmap spreadsheet was highly satisfying.

Not for the first time, I have added a column which I probably should have included from the start. Firebase paths are roughly equivalent to relational MySQL tables, a logical way to organise data albeit without the ability to join. Adding this column introduces a fair bit more blue-shaded alpha entries, but that represents most of the work still in front of me before the soft release scheduled for December.

There are also a lot of gold entries which are single-underlined, indicating that maybe they weren’t quite gold yet as I am still iterating them. I confess that I was too keen to upgrade the status of some elements, as they were perfectly functional prior to recent meddling but lacked features to make them “complete”. That is the sort of stage that really should be called RC. Gold just looks better than green in my system though, so that’s just reflective of my eagerness.

It was important to me to maintain the maximum level of authenticity in club livery, both in the shield and jumper designs. Each of the above designs is used across Australia in urban and regional leagues. There are a few obscure ones that I haven’t included yet, like a diamond design and one example I saw of a double ribbon. Complicated motifs have proliferated in recent years, with widespread mergers between player-starved clubs necessitating hybrid jumpers. Some of the above nomenclature is taken from heraldry and flags, and is not strictly native to footy. Hooks is an Australian military term for multiple chevrons. Internet rabbit holes everywhere when researching that subject!

I spent most of the week cleaning up Phaser bugs and adding long-planned features to existing screens. I am just about finished that process now: the turds are all a-gleaming! I would be procrastinating if I don’t turn my keyboard to the next mini-project: PvE modes with properly balanced opponents. Campaign is functional at the moment with progression restrictions turned off. The game is supposed to start with division 3 in all regions unlocked, but higher divisions locked until you earn enough stars by beating lower opponents. Monday marks the point where I put aside the sexy Phaser stuff and turn back to the boring LAMP systems where my PvE JSON files are generated.

I trust that all of this work iterating on the client will improve quality-of-life for users, a very important issue for the Football Manager genre where navigating the UI to manage resources is a large part of the game. I haven’t spent much time on making the game fun as yet… maybe I am wrong to not have that as a primary focus, but I think slickness of interface for a title like this is also of high importance.

After some feedback on the onboarding process, I made the decision to buff Starter packs to include cards with some Fitness levels and items equipped, to make the opening matches more winnable against early opponents in Campaign. The RPG convention that your party should blow through initial opponents with easy wins to give quick dopamine hits exists for good reason, I am told. New coaches might get discouraged if they get belted in their coaching box debuts. This subject needs more dialogue, and maybe some A/B testing… though such experimentation has likely been tried and failed many times before, leading to inevitable convergence towards the only successful strategy.

One day this week was also spent on drafting legal texts, with Terms of Service, Privacy Policy and End User Licence Agreements all generated. It is possible to spend a lot of money on lawyers to get the intricacies just so. It is also defensible to write prose in plain conversational English, like Mojang does for Minecraft. I chose to use a series of generic legalese-sounding boilerplates to keep things as simple as possible, at least for now. American Web sites which do that sort of thing want to inject lots of falderol specific to American states. I suspect European requirements would be more stringent, as the EU tend to be more organised on data laws. In Australia, we mostly lag well behind the global curve which emfoggens the writing of terms that might stand up in court for a test case. Box ticked for now, with boilerplate enough until I have to deal with the big stores.

After a long fallow period I picked up the Midjourney cudgel this week to generate art assets for tokens. The most searched-video on this was very helpful to assemble each coin in Photoshop, but the real breakthrough came when I tried “bas-relief” as a keyword in Midjourney. The results are just so striking, easily transferrable to a coin context and take so little effort on my part. I mean, just look at that texture! You could hardly tell the difference from a photo of a real coin – apart from the blue shading I have given it, which is not really a natural metal colour unless you make it look explicitly like titanium. Hmm, now that I have looked that up, there is a lot of potential to mimic real metals and alloys used in coins, which would fit in with the photoreal art style I am going for enabled by Midjourney v5. The uncanny bas-relief valley is another big AI image generation plus from where I’m standing.

I have been telling everyone that December was the plan for the soft launch with in-app purchases. As to when I open up the beta from its current closed status (about two dozen invitees), that is probably going to be when I get all match and journey modes operational. I don’t know how long that is going to take. Some time in November, I guess.