Yes, manually: after spending considerable time thinking about different ways of trying to automate the process of converting user-posted lists into a coherent master list, I rejected them all because most of the copy/paste approaches are prone too much dirty data and require all the data to be in before you can really clean it all up. But more importantly it cannot be done automatically because even trained NLP systems still need a source of truth about what a car model is, and humans can’t even come to a consensus themselves.
Imagine three car fans walk into a giant car corral at a race event with all the model variants you can imagine. One is thrilled to see all different Porsches. One is thrilled to see all different Corvettes. One is thrilled to see all different Subies. Each one thinks the other two are crazy because to them the other cars all look the same, while their favorite cars are all different. Try telling F1 or NASCAR fans that all the cars are the same. To the game, they have to be different because of licensing. Try telling a Porsche fan that a 911 Carrera and GT3 and GT2 are all the same. To the game they have to be different because of PI. If an event is for mid-engine cars only, one Corvette’s going to make it in and the others will not. The American VIN system doesn’t even use model generation, and makes each model year a different code. Now add in all the racing variants. Then add in all the named concepts and custom builds that users have requested and the community still hasn’t named every one that has existed.
We have users who request over 1000 models (1400 is the most I’ve seen so far). Many users will include all the different trims is a single car model’s generation as different cars in their lists. I can’t tell you how many cars this game or the next will ultimately have in the game but I think everyone here can assess the reality of trying to match just that one list to a Forza game release. To count every model variant mentioned in a user’s list would result in bias because you’d see multiple votes for many mass production models, with fewer votes for models with only one trim level like the 1980s DeLorean or Ferrari 458.
So after settling just on the data collection method, I had to settle on some sort of classification system to come up with a coherent summary to present to the developers. I decided on Model Generation (including multiple facelifts), performance level (base/mid/high/track/race), and body style (all the roof and door/pillar variants). I came up with over 11,000 “models” across 780 makes. I then went through every single post from the thousands of users in the wish list thread and matched each of their requests to one of the 11,000 models line by line in a spreadsheet. The fastest I was able to process them was about 1 minute per 1 post (60 posts an hour). Then using a pivot table I could do a count of the master models and number of users - if a user asked for three trims of the current model F-150 that were not significantly different in performance, the pivot counted 1 vote for the current F-150 road trim. I then posted updates about the top 100 in the summary thread you see above. This was all separate from the in-game data I’ve shown on my manteomax website.
As you can imagine, companies want less labor-intensive methods, which is why we’re moving to the Suggestions Hub system of upvoting threads and then pulling the data from the forums. You will see a far more extensive Suggestions Hub soon. The current Cars subcategory with user lists will remain to allow users to show off their whole lists, but we will be moving to single model threads for upvoting.
You can read the thread linked above with numbers. To summarize, I have said this for years and will continue to say it, and the people who really need to process it won’t accept it but it’s still true:
- No one person or group speaks for what the entire community wants. “We all want x” is always false. There isn’t even a plurality consensus, much less a majority consensus. No one person knows the magic car roster that will satisfy everyone.
- Users will be disappointed by at least some of the choices the game makes and the fact that no game is ever perfectly executed at scale. You the reader are one of those users.
- It is not possible to satisfy everyone in a licensed game, fixed budget environment. User claims of what is possible are usually pretty far off the reality of game development. Game development is tedious and never as fast as the developers would like it to be.
- What was possible in a previous era or game is not necessarily possible in Forza given a number of factors, many of which are game development-related that users do not see.
- Turn 10 and Playground Games are particularly interested in following user feedback, incorporating user feedback, and demonstrating that the game is developed with that input, regardless of user claims to the contrary.
- Turn 10 and Playground Games are not trying to bleed anyone’s wallet, regardless of user claims to the contrary. On the other hand, without profitability there would be no game to improve.
- Cars are not the only asset or feature requiring game development for overall game success. Developers don’t just walk over from Character Customization Building 1 to help out with Car Development Building 4 when users believe the community wants cars instead of cosmetics.
- User feedback does not supersede licensing, time budget, financial budget, logistics, game mechanics, or game vision. “Most popular” does not guarantee the first or fastest addition to the game, or addition to the game at all.
- Relationships with manufacturers do not guarantee that all their models or the “obvious” models will appear in game. Every model is licensed specifically and separately for each game, or ultimately not licensed for various reasons - all logical - that users have not imagined.
- Car wishlisting is not a vending machine.
- User timelines and development timelines will never be in sync.
- What users perceive as intended bias is unavoidable.
- If there’s anything guaranteed in gaming, it’s that there will be bugs.
- Users will pick apart this list and try to counter it. Ask yourself who is more correct.
In short, user disappointment - in some way - is to be expected. On the flip side, Forza has enjoyed many years of success and players who are overall happy with the game they’re playing. Despite hundreds of cars in the roster, by and far most players only use a handful of cars in game after they’ve found the ones they like. The cars they like in game are not necessarily the cars they favored before coming into the game. Game purchase costs over the years have largely remained flat despite inflation and increased number or quality of cars and content (again, you will dispute this, but please reread the list), and Xbox has made it easier for more players to get into the game with Game Pass and offering the game on Steam.
I don’t know if this wall of text remotely satisfies the OP or other users on the forum. I’ll still be plugging away (including on weekends). Whether you believe it or not, whether you’re satisfied or not, we believe in community engagement and we are continuing to make an effort to mesh community interests with development realities to provide a successful, enjoyable experience for everyone.