Reading Old Threads May Help Shed Light on Why Things Haven't Happened In The Game Yet

I wasn't around the forums when TS4 first came out so I never had the opportunity to read this post when it was active. I stumbled upon this thread by accident while looking to see if we could build apartments in other worlds.

That thread is like 76 pages long but I've quoted some interesting questions that may help others to understand what goes into implementing certain things;

It sort of explains the following:
Cars
Horses
Freeing Babies
Apartments

Looking back on that thread 6 years later, there's actually quite a bit of good information explaining "why" something was done this way or that way or why they deleted/changed certain things and their reasoning behind those actions.

It talks about:
Open Worlds vs TS4 Style
Emotions
Multitasking
Why some animations were removed (such as dancing face to face)
Why some animations were not implemented (such as hand holding taking a walk with a loved one)
Why teens are the same size as YA and Adult

"SimGuruMax;c-12816742" wrote:

"ULIBABA66;12750172" wrote:
Is it hard to do horses and cars with animations for The Sims 4?

Horses and cars are very different animals. In fact, cars aren't even animals (see what I did there?).

Let's start with Cars because they are simpler. The main problem with cars is making room for them in the world - parking, driveways, garages, streets, routes. This is also a problem for horses, but we can re-use some amount of the solution for both. The main issues relate to the layout of the player's lot, the lot boundary with the street, and how to standardize parking spaces both on-lot and along the street so cars don't clip into each other or the ambient traffic. There is additionally some animation cost to animate Sims getting in and out of cars, but thats not too bad. We also need to animate the cars driving, but that is also not too bad since wheels turning and rotating is very simple.

Horses are much harder and get into why Pets is often the third or fourth expansion. Horses have an actual walk cycle. They need to plant their feet and turn realistically as they follow routes. If we just animated a horse doing a walkstyle and then lerped it (linear interpolation) across the path, it would look extremely bad going around even slight corners due to foot sliding and the horse rotating around a pivot point in the center. Also, it would clip like crazy through everything because our entire routing system is based on Sim size and turning radius. The routing system needs to be expanded to account for something large and/or oblong like a horse or a dog. The main sticking point is that we have a much higher standard for animation quality on the Sims than most games do - we take things like foot planting vs. sliding very seriously and we wouldn't want to ship the version of horses that most games ship with. Just look at the rampant foot sliding and glitchiness of the equines in a game like Red Dead Redemption if you want to know what I mean.

Edit to mention that some of the issues with sliding that plague horses would also be an issue with Cars. We have to be careful when authoring their route splines to make the corners shallow to avoid sliding. This also further complicates the boundary between the lot and the street as it might take weird curves to get the car joined in with the street waypoints.

"SimGuruMax;c-12642962" wrote:

"agent_bev;12641313" wrote:
Thank you for taking the time to humor us. I have a few questions:

Is it technically feasible to make babies real again? I use the word "real" because I cannot think of a better word to describe it right now, I just mean having babies needs be visible, allowing babies and older sims to build some kind of relationship, having babies that are not tied to the crib, etc.

Also, will it be possible for lot sizes to be bigger or are they 50x50 at the maximum because of engine limitations?

Finally, I know one of the gurus mentioned that it is not technically possible to have CASt in the game, but would it be possible to have an external tool where you could re-color and re-texture all the objects and clothing in game, and then upload it to the game?

I can say somewhat unequivocally that it is technically feasible to make babies "real" again, as you describe it. As for prioritization, that is really up to our production team. Babies not being tied to the crib is going to be a sticking-point, though. That decision was made for many reasons, not just time-constraints. We felt dissatisfied with babies in TS3 and were looking for ways to get more good things without having all the crazy complexity. One of the most complicated aspects of babies is that Sims could carry them around - babies aren't just another carryable: they cannot be holstered (Sims can magically disappear items in their hands temporarily to play animations that need the hand) due to believability issues and there end up being a huge number of weird rules about them not being able to be placed in different locations, etc, etc, etc. Carryable babies were such a huge headache on TS3 and we had to ask the question "Is it worth it?" I know everyone wants perfect, realistic babies, but the reality of making games is you often have to make choices because you are working with constraints on all sides. Complexity is our enemy when building The Sims, so we look for any opportunity to keep it down. Ultimately, we thought we would rather spend our time making the actual baby interactions fun and interesting and have more time for people like me to spend devoted to our new socialization system and multitasking - it would have occupied my time entirely for many months to build carryable babies. Keep in mind that everything is effectively zero-sum; adding complexity in one place necessitates reducing complexity in another as we don't have infinite resources.

One thing about babies I am displeased with is the fact that your Sims put them back down in-between actions. We have tech to solve this problem and somehow it just didn't end up happening. I would very much hope that we address that problem in the not-too-distant-future.

"SimGuruMarion;c-12736133" wrote:

"Jendowoz0612;12718039" wrote:
Hi Max,

Very interested to find out if the interactions with babies will be modified. I tend to queue several interactions when it comes to babies. They are babies for such a short time. If Sims are able to carry around other things and move about the Sims' World, can't this work with babies? As it is now the Sims put them down, pick them up, put them down, pick them up for the next interaction. Thanks.

Kind regards,
Jen :)

Hi Jen,

I agree that when you chain interactions with the baby, the sim shouldn't put it down each time. I think this is a bug. I will take a look at what we can do to fix that.

You are asking how babies differ from other carryable objects. That is a really good question!

If you watch sims carrying non-baby objects, you will notice that there is a standard way in which they hold the object - it is in their hand, and they hold it in front of the body, a little away from their torso. This not entirely natural pose was chosen as a compromise between how people actually hold stuff, and our need to play other animations while the sim is holding the object. It allows the sim to make gestures, bend their body forward and backwards and so on without the object getting in the way, or the other hand going through the object, or the object disappearing inside the sim's body. This generally works well, but there still are animations where we cannot accommodate carrying something (for example if the sim is clapping, we need both hands free). For those animations we make the object temporarily disappear (we call this "holstering" the object).

The baby has several properties that make this standard solution not work well:

For one, we *really*do not want to holster the baby, we believe it would be very distressing and inappropriate to see the little ones pop out of existence and back in constantly. Not holstering means that for every case where we would usually holster an object, we have to create an alternative solution for doing that thing while holding the baby. Regrettably, all the alternatives are problematic. The best solution would be for each incompatible animation to find or create a different animation to play that works with the baby; but this is very time consuming, and not all cases can be fixed that way (holding a baby while eating cake for example). We could make the sim put down the baby, but this is disruptive, as she has to find a spot for the baby, and route there and back. We also can't use this solution too often, otherwise the baby will always end up on the floor after a few sim minutes of being carried. We could also simply disallow certain actions while a sim is carrying a baby - but this can creates strange and often hard to understand limitations on what sims can do, and so it is not really a desirable solution. There really isn't a catch all solution, so each of these cases needs to be dealt with individually in a way that makes sense for it.

The other big difference is the carry pose itself. You can hold a book in one hand at arm's length in front of your body, but holding a baby that way just won't work :) Babies need to be cradled close to the sim's body. The problem with this is that as soon as the adult sim's torso starts moving, the baby starts clipping with it. If the adult moves a lot, half the baby may disappear inside the adult at times. For the animations that have this problem, we would again either have to change the animation (making it less expressive for all situations), or make the sim put the baby down first, or disallow the interaction.
If you scrutinize the Sims 2 and Sims 3 baby carry poses, you can see the compromises we had to make there to make this work at all - the baby is precariously balanced on the sim's forearm and contact with the adult's body is kept to a minimum. Sims have very strong biceps! In spite of this, you can still find fairly severe clipping issues in both those games with the baby.

Another issue is that the baby is much bigger than the other carryable objects. The only other object that even comes close is the pizza box, and we severely restrict what sims can do while holding that. The reason is simple - the bigger the object a sim is holding, the worse the clipping issues that you run into. Socialization is particularly problematic - sims gesture a lot while talking, and when we did tests with carrying babies while socializing, we saw a lot of hands reaching into the baby or moving though it.

Finally, our normal pick up and put down mechanism doesn't work with the baby - with all other objects, we have the sim do a swiping gesture and the object jumps between their hand and the world. With living creatures, we don't want to do this, we think it would look uncaring and weird for the baby to pop to the floor. This means that we have to implement a whole different way of doing pick ups and put downs, with much more stringent placement and registration rules (and all the odd edge cases and routing issues that come with that).

All of this doesn't mean that it is technically impossible to carry babies, of course. But it is time consuming and difficult to do right. So, in short, having other carryable objects in the game provides a starting point, but it isn't the whole solution by a long shot.

That was long. :-o Hope this answers your question?

"SimGuruEugi;c-15219604" wrote:

"simgirl1010;15217688" wrote:
@SimGuruEugi, are you able to shed any light on the decision to confine apartments to San Myshunu and not allow the building of apartments from scratch? I think it would go far in helping players understand why these restrictions exist. Thanks! :)

You can think of an apartment building as a neighborhood. They are authored by our world creation team as part of the world we ship for the EP. (We actually can't patch worlds without invalidating saves).

If we had allowed the creation of apartment buildings from Build/Buy, we would not have been able to achieve the look we were going for. For instance, B/B limits construction to five floors, but you want buildings to look like they have many more than that, so you can look over the city skyline.

There are also performance implications. You might want a building to look like it has, say, hundreds of windows and small items of decor. If those were actual B/B objects, they would all count against our performance metrics.

We also allow multiple households to inhabit the same building. So we'd have to add some element that would have allowed players to define the boundaries of an apartment. If it wasn't properly authored (e.g. lacking front doors), some game functionality might not work.

So we reached what we think is a good compromise. We modeled "shells," which are apartment buildings that look great but are technically one object, and then we allowed B/B operations inside it.

I can answer more specific tech questions if you have any.

"SimGuruMax;c-12646236" wrote:

"Tarô;12646199" wrote:
Hi SimGuruMax
The thing that I dislike the most in the game actually is the Teens... Why are they exactly the same height and look exactly the same as young adults ?

My question is : Do you consider, and would it be possible technically to modify the teens at this stage of the game (after release) to make them shorter and possibly a bit different than the young adults ?
Thanks

Animation time is our enemy and we take our animation quality very seriously. Having a third height would have required a *huge* number of new animations. To save some time, we could probably have ported some of our existing adult animations onto the smaller skeleton. We have partially-automated processes to port animations in various ways already, so we could cut some time there too. Problem is, someone still needs to touch all of them by hand and some are going to require considerable rework to get right. And this is prior to getting into the animations that just don't make any sense at all on the smaller frame and must be done for the Teen from scratch, of which there would be many.

Much of what I end up talking about is tradeoffs. Its the brutal reality of building games, or really anything honestly. So we have on the right teens of adult height but substantially more animation time to devote to both teens AND adults, and on the left we have shorter teens and far fewer animations for both teens and adults. We made the right call.

I wouldn't hold my breath on teens getting shorter.

I'm sure there's lots of other interesting information but I just wanted to highlight some of the frequently asked for things. All of these things were from a post dated in 2014.

In case you would like to read through all of the tons of pages too, I found this information here:
https://forums.thesims.com/en_US/discussion/782859/simulation-tech-talk/p1