Cloud computing could definitely be used for some things to help free up some bandwidth, but for ai it’d probably have too much latency.
As these games have gotten more sophisticated, having good and especially believable ai has proven to be pretty difficult. Its a balance between what you need to do and what can realistically be done within the processing limits.
With this game they focused on the speed of the ai without the use of rubberbanding. The downside it seems is they have less overall awareness. In one on one situations, they seem to be fine and react like they should. But if theres a few more cars around at the same time, they fall apart.
Many developers are working towards making and utilizing machine learning systems to help make things better. But for the foreseeable future, ai opponents are still going to have similar deficiencies that they have now.
Sophy was a great addition to gt7, but it still exhibits some bad habits. They are aggressive, they will dive bomb and if they dont “see” you, they will make contact. But their biggest strength is they are pretty predictable in their actions for the most part. You can trust they will do what their supposed to do in most wheel to wheel situations.
This is what forza motorsports ai lack, they are unpredictable and usually cant be trusted to do the right thing in most situations. In a year and a half, imo little has improved in these areas. Fixes are often described as the reduction of certain flaws instead of actually eliminating them.
After seeing the implementation of ai in this game and the numerous “fixes” that dont fix anything post launch, i have little faith things will get better in the future. Certainly not in this game, but unfortunately i dont see it improving in the future either.
Im pretty sure he said 9. It almost seems as though when they were promoting this game, they added a lot of ideas that werent quite finished. The problem is, does the update now add these 9 lines, or were the 9 lines we have now not working properly or at all.
Im still troubled by their focus on speed when the ai’s issues overshadow that one positive, literally one, as only one drivatar is actually fast out of 23. Im a pretty quick player and while its nice the ai are faster, if they were actually functional, i could just handicap myself to have a good race. I guess we’ll find out soon enough what theyve cooled up.
You know the whole concept of the “vertical slice” that demos are based around? Best case scenario, I think they had it sort of nailed down for a certain vertical slice, and that’s what they showed to everybody. As long as you ran this car on that track on this equipment…it looked and played amazing. The problems didn’t come until they tried to make what worked with that Corvette on Maple Valley on a $5000 ultra-high-end PC work with a Lamborghini at Homestead on a $500 Series X.
Again, given everything else regarding this game though, I’m skeptical of that best-case scenario being the case, and I wouldn’t be surprised in the slightest if the build the reporters got to see was a dressed-up version of whatever proof-of-concept build was made to justify the investment to Microsoft studio heads.
Tbh, i dont trust reporters or most reviewers who try and play racing games as they usually aren’t good and certainly dont know anything about what makes them good. We didnt see the ai until they released the demo to youtubers instead of everyone, watching those videos the ai we received were the same. The game simply wasnt ready at launch.
I can certainly understand that, but they are still reporters, and the trend amongst all of those reporters’ previews was that there was a bill of goods sold that very obviously didn’t match the final product. The YouTubers who did speak out when able is sort of beside that point.
That’s why I said that I don’t think a single truth about this game has been spoken by the production team from the word go. The more we learn about various things that were promised and compare them to the product that was delivered, the harder it becomes to take anything that has been said or will be said by them for granted.
I think the reporters and reviewers helped sell this unfinished product, regurgitating the scripts they were given. Instead of questioning or highlighting any issues, the hundreds of bugs, they gave it high review scores and a game of the year award.
Post launch, youtubers had their say, and even then barely a peep out of any professional media sites about the game. They simply dont do their job and cant be trusted when they act like they do.
Maybe this is just the way things are now when it comes to these things. Labeling your game a service, overpromising and ultimately undelivering.
Use the success of the Forza brand to sucker people into engaging in whatever this is supposed to be with all kinds of marketing PR tricks. The latest one being this “new” AI drop a week before the big update to give the illusion that they’ve been hard at work. If I’m not mistaken they also claimed they were working on Fujimi Kaido “since launch” to have ready for the 20th anniversary. 20 percent substance and 80 percent marketing.
Hopefully the brand makes a comeback, but things need to change for that to happen. Disappointment is trending with this game, and it’s only going to get worse unless T10 makes some hard decisions.
I have my reservations about that too with GT7 and FM being console exclusives. If they optimize it perfectly for console then that pushes FM out and I doubt Microsoft want that to happen.
We’ll just have to wait and see, but if Kunos can optimize the game then that would make it way more popular, and there wouldn’t be a need to play Forza. Unless someone likes the variety of cars Forza offers (which doesn’t matter if they all sound bad and understeer heavily imo).
My expectations are low though. ACEvo will probably just be a teaser for the real experience on a rig like its predecessor. If they build off of ACC then great. With online games now-a-days I’ll wait until the reviews come out until I decide to waste my time and money.