Murders & Mistresses Dev Log #3: Making the love interests
Hey everyone,
I hope you had a wonderful Holiday season and New Years. I spent most of my time off concepting a small game for fun… I can't really help myself oops.
Since my last dev log we' ve revealed the core love interests and released the Steam page including our two minute trailer, which I'm very proud of.
In this post I wanted to share the goliath effort it took to make those love interests look as good as they do. Because unfortunately, as is so often the case for us, nothing went quite as expected...
The Murders & Mistresses Prototype
Whenever I work on a new game, I try to get something playable to my team as quickly as possible. I find it very difficult to understand whether an idea in my head actually has any legs unless I get to play it myself and also get to see others' opinions on it. Prototyping for an AVN can be difficult because so much of the game comes from the polish. In my opinion (and I may be completely wrong as I've never worked on a platformer (games like Super Mario Bros. or Cuphead) before), but I would assume that you can get a feel for whether your platformer prototype is good, mainly based on the mechanics, you don't need a fancy UI or good looking characters yet.
For AVNs I think that's a bit more difficult. So much of the game comes from the visuals, the characters being attractive, the locations looking interesting - even things like the UI looking clean.
For a game like Murders & Mistresses, which takes itself very seriously, immersion is key, even for a first prototype. So outside of writing it, I had to make attractive love interests and a reasonably decent UI as well.
For the initial prototype, it took me about a week to make all the rough first character models, I had clear ideas for each of them so I could move relatively quickly. The UI was also more work than expected simply because the game had a lot of UI screens (Deductions, conclusion, map screens etc.).
Even with all that prep work, there were definitely some moments in the M&M prototype where a bad render with a weird facial expression of an unfinished and poorly lit Morgan completely ruined what was supposed to be a dramatic turning point in their marriage.
When I was finally able to give the first prototype to my team - the response however was overwhelmingly positive. To my relief, people enjoyed the core detective gameplay and really liked the writing & characters. There were some rough edges - mainly my very outdated Daz rendering (images) & animation skills as well some confusing navigation around crime scenes - but the game had something special about it. We all knew it.
Outgrowing the Prototype
I made the first M&M prototype in December 2024. By the time College Kings 2 Ep6 released in July, all of M&M had been written (although not yet fully edited), but outside of that, we hadn't made any progress on the game. The entire team had been busy with CK. When we looked back at the prototype it was clear that the characters were not good enough for an actual release. The male ones were all great, but we had such a high bar for the love interests - we felt very strongly that every single one of them needed to be both attractive, but also interesting looking in their own way, as well as fit the character visually.
First we tried using my prototype characters and working from there - but that did not yield any good results. After a week or two - I said, don't worry about it. I will sit down non stop 16 hour days for a weekend and remake them all from scratch.
So I did. But I still wasn't happy. A weekend turned into a week, turned into two, turned into a month.
By this point, Morgan specifically (your in-game wife), had turned into an equally funny and depressing running joke inside the team. At a certain point I started numbering images of her starting with "Morgan v.1Million.X". We had tried so many versions of her, everyone in the team had also tried their hand at making her. Nothing had worked.
Technical Issues Adding Up
We had eight core love interests to make - and without them we could barely render any of the scenes needed for the public demo, nor start making the trailer or Steam page art. I told the team it was gonna take me a week back in July, because that's how long the prototype characters had taken. But I didn't realize how much more difficult it would be to make the characters just 20% better. Just that final 20%...
Switching to a new version of character models (Daz Genesis 9) for significantly more detailed and realistic characters also didn't help the process. We couldn't use any of our existing Genesis 8 models, assets or tricks to make our M&M models look better. We even had serious discussion about switching back to Genesis 8, but the visual fidelity that Genesis 9 provided was too clearly superior. We all believed that it was the future of Daz, and not going with it would make our game look outdated a year from now.
But the improved realism caused all kinds of new problems, my original models had looked much worse under professional lighting (instead of the sophisticated flashlight in the face lighting method I had used for the prototype), and even small physical oddities (a slightly too large head, one eye being higher than the other, etc.) immediately made the model look uncomfortably off.
We were a month in and we still hadn't actually managed to complete a single love interest, nor really make any improvements towards the original prototype models, which so clearly weren't good enough to put into the actual game.
Getting it Done
I originally wanted to title this paragraph "The Breakthrough" cause I thought it sounded cooler, but the reality is there wasn't one specific "Eureka" insight that changed everything. We just kept at it. For months. Back and forth. Hundreds of iterations. Trying new base models, changing every single possible modifier to every single possible position. By the end we had bought pretty much every single female model in the daz store. Suffice to say, it cost us a lot of money.
During that time I was really struggling personally. Not because of anything else (although I was going through a breakup), but front of mind for me were these characters. I loved M&M so much as a game, I was so proud of what I'd written - and with every day that I looked at a character model and felt strongly that they weren't even nearly good enough, I started to lose hope. What if it doesn't matter how good the game is because the characters aren't attractive enough for anyone to wanna play it?
I would sit at my PC until 4 am, with a cider in hand, somehow trying to find the magical change that would make a love interest click. The more time went on I started doubting love interests we'd already completed. One of the worst parts of working on faces is that when you stare at one for hours, you stop being able to "see" it. As in, your brain gets so used to it that you can't look at it the same way that you normally would anymore. This means that every hour or two you'd have to stop what you're doing and step back for 30 minutes.
Healthier? Sure. But also incredibly frustrating when you feel like everyone else is waiting for you.
It took until the last week of September, two and a half months after the original mid July deadline that I had promised my team for me to finish Morgan, the final of the eight love interests. My bad.
Despite knocking our entire production schedule back by over two months and contemplating ripping out all of my hair multiple times – I have no regrets. I love how the character models turned out, I think they're beautiful, instantly recognizable and perfectly fit their personalities and I think they complete Murders & Mistresses visually.
Sometimes I find it difficult to know the difference between wanting a high quality result and obsessive perfectionism. I probably cross that line more often than I'd like to admit, but I also deeply believe that some things cannot be skimped on. The love interests are the core of the game and they needed to be of the same high quality standard that I had set for the writing and the gameplay. They needed to be incredible.
And in my opinion, that's exactly what they are now.
Outro
I hope you guys enjoyed this little insight into the sometimes grueling realities of trying to make something the best it can be without having any idea of how to get there.
Taking those characters and turning them into a banging trailer & Steam page - that was an entirely different experience (though some of it was equally painful). Let me know in the comments if you'd like to hear about that next time, or if there's anything else you'd like me to write about in these dev logs!
Thank you for all your continued support, M&M's Patreon Demo release is not that far off!
Much love,
Steve