Barton's BAT & Lot Workshop

[Raymond7cn 6,772]

Woww! Barton, It's wonderful! Do you remember that we have had a talk about art work with @mattb325 in a post by you? Yes with 3ds max, we can create more fancy models, but I think it makes no difference with art work, I knew Matt is also a good painter,  Now I believe so are you. It also reminds me of those Louvre statues, I was amazed at them when modelling, Couldn't believe they were made thousand years ago, shapes and colors and shades, when they are put together and you just squint at them, You will know that's real Art, Please keep it up, otherwise, We will lose another great artist in SC4's world.

Life is a journey, Keep learning is one of the most beautiful view in this journey, I'm so happy can learn from you everyday. @BartonThinks @mattb325 @Simmer2 @AP @sejr99999 @…

Yours,

-- Raymond

[BartonThinks 3,416]

That's very kind of you to say @Raymond7cn. Though I wouldn't call myself much of an artist. I can barely paint a wall, never mind a canvas.

Another couple of flora experiments/practice exercises. Some wild flowers and a hedge. Skipping the Gmax shots this time. Included yesterday's tulips for reference.

For the flowers, I used a flat plane and then added a scan of an actual flower with an opacity map to define the petals. Possible overkill for something this small, but I felt it added something in Zoom 6 when compared to a flat color/a simpler shape for the petals. Seeing them next to the tulips I made yesterday, it's clear the colors I chose for the tulips were a little too blue. Probably a good lesson in the importance of getting color balance correct.

The hedge definitely has room for improvement, but it was a helpful experiment for larger/leafier flora. Basically, the hedge is five boxes nested inside one another, with speckled opacity maps to give the impression of multiple layers of leaves. I imagine it's a technique that's been used by a lot of other creators for Gmax flora. Unfortunately, that seems to leave a black "halo" around the edges of the model when it's exported. I was able to reduce the halo effect by playing around with the opacity maps, so I'll need to experiment further to see if I can get rid of it completely (or near completely) for future models.

[mattb325 28,880]

One of the easiest ways to make a hedge is to simply make a box with a lot of segments and then apply a noise modifier.

It literally takes under 2 minutes.

This is using gmax with a 2x16x2m box with 30 length segments, 200 width segments and 30 height segments.

The noise modifier parameters are included in the screenshot, but you can play around with them as you see fit to get different results.

 

 

[BartonThinks 3,416]

9 hours ago, mattb325 said:

One of the easiest ways to make a hedge is to simply make a box with a lot of segments and then apply a noise modifier.

It literally takes under 2 minutes.

This is using gmax with a 2x16x2m box with 30 length segments, 200 width segments and 30 height segments.

The noise modifier parameters are included in the screenshot, but you can play around with them as you see fit to get different results.

 

 

Oh! I had read Jason's tutorial suggesting this, but I just automatically assumed this feature was only available in 3ds Max. Maybe not the smartest assumption to make!

[BartonThinks 3,416]

Another experiment, this time with lighting. Someone over on SC4D was having trouble lighting a lamppost. They were using an omni light, which was creating a shadow effect similar to the one below.

After a bunch of failed attempts, I found a setup using Free Spot lights ringed around the light that seems to solve the problem. 

The one catch is that this setup creates a bit of a starburst pattern due to the overlapping lights and shadows. However, the pattern can be reduced by increasing the number of lights in the ring.

In the image above, I used 16 lights around the lamp. In the image below, there are 32 lights, and the starburst pattern is much less noticeable.

I'm not sure if this is the easiest solution to the problem (it's certainly a very light-intensive way to light up a single lamppost), but it was a fun bit of problem solving and a good way to get a bit more familiar with how lights behave in Gmax.

If anyone's interested in a longer explanation, here's a link to the post on SC4D where I detailed the setup I used.

[BartonThinks 3,416]

More work on small flora. This time, it's a daylily prop (there's only the one prop on this lot -- I've clustered it at the sides).

The leaves around the bottom could probably use a little more variation to break up the surface, or possibly some simple texturing to give them a bit more definition. I originally had several more flowers and leaves modeled, but I was surprised at how much fuller the foliage appears once the model is exported compared to how it looks in Gmax and even in the preview renders (e.g. how much fuller the leaves are in-game vs. how they're modeled in the Gmax model below). Even the four bulbs in this version look a little crowded once the prop is in the game. It might be worth winnowing it down to three flowers when I model a final version.

Otherwise, I'm pretty  happy with how this turned out, though I'm (as always) open to notes and feedback.

Looks like I'll be sticking with smaller plants for the next little while -- I seem to have gotten a bit hooked on them. I'm still trying to figure out how to create leafy plants (herbs, shrubs, etc.), though I'm thinking of working my way up with some other, leafier flower species. I'll also need to figure out how to handle seasonal modeling for these and other plant models.

[AP 3,588]

May I ask if you bent the leaves "yourself" using an editable poly or if you used a bend modifier?

[BartonThinks 3,416]

9 minutes ago, AP said:

May I ask if you bent the leaves "yourself" using an editable poly or if you used a bend modifier?

Simple bend modifier in this case, though I've played around with applying the bend modifier to isolated sections of an editable polygon and/or manually adjusting the vertices, particularly for the stems. That made it much harder to tweak the modified elements, though, so I've been avoiding that where possible.

In any case, the leaves are just an ellipse spline with a bend modifier (with some slight variation in the bend between leaves) and a UV map. I resized and adjusted the rotation of individual leaves to get a bit of variety.

[BartonThinks 3,416]

So this is a bit outside the scope of my current projects, but I felt compelled to write this down somewhere. Might be useful if I decide I'd like to create diagonal BATs...

In @Indiana Joe's thread, I mentioned a trick that CP used to control the growth of certain types of lots, where he would use zoning density to restrict growth. In CP's case, he had some low-density lots, some of which were more appropriate for the country, and others which made more sense in a "village setting." So he set the country lots so that they would only grow on low-density zones, and the village lots so that they would only grow on medium-density zones -- despite the lots themselves being Stage 1/2/3 lots. T Wrecks uses a similar trick for the IRM mod to control the areas where dirty and high-tech industry grow.

I got to thinking that this might offer an easier way to grow diagonal residential buildings. For example, if you set all of your 1x1 residential lots so that they only grow on low-density and medium-density zones, you could assign 1x1 diagonal lots so that they only grow on high-density zones. This would allow you to grow diagonal residential areas without the usual headaches of diagonal growables (having to bulldoze buildings that show up where they aren't supposed to, rampant use of the "make historical functions," elaborate sets of lot blockers, etc.). The same technique would also work for commercial areas, if you'd like to avoid plopping diagonals as much as possible.

The first catch is that you'd need to avoid 1x1 residential (and/or commercial) buildings for high-density growth stages. But 1x1 lots are relatively rare for these growth stages, so this seems like a small price to pay if you want to grow diagonal neighborhoods.

The second catch is that you'd need to stick to smaller lots. However, you could always use ploppable buildings for larger buildings, at least in commercial areas.

 

Edit: Tagging @The British Sausage to get his thoughts on whether this might work, given his experience with diagonal growables.

[sejr99999 8,343]

I like these ideas as I would like to see more diagonal houses   one simple trick is to create a res desc file and boost the occupant numbers to force it to stage 8 on a 1x1 lot

makes some serious traffic issues however

[BartonThinks 3,416]

1 minute ago, sejr99999 said:

I like these ideas as I would like to see more diagonal houses   one simple trick is to create a res desc file and boost the occupant numbers to force it to stage 8 on a 1x1 lot

makes some serious traffic issues however

Thankfully  you wouldn't actually need the lots to be high-density growth stages. You could keep the lots at growth stage 1/2/3 (or higher, if you're making diagonal townhomes) -- they just would only grow in high density areas.

[The British Sausage 5,449]

On 7/13/2021 at 10:12 PM, BartonThinks said:

So this is a bit outside the scope of my current projects, but I felt compelled to write this down somewhere. Might be useful if I decide I'd like to create diagonal BATs...

In @Indiana Joe's thread, I mentioned a trick that CP used to control the growth of certain types of lots, where he would use zoning density to restrict growth. In CP's case, he had some low-density lots, some of which were more appropriate for the country, and others which made more sense in a "village setting." So he set the country lots so that they would only grow on low-density zones, and the village lots so that they would only grow on medium-density zones -- despite the lots themselves being Stage 1/2/3 lots. T Wrecks uses a similar trick for the IRM mod to control the areas where dirty and high-tech industry grow.

I got to thinking that this might offer an easier way to grow diagonal residential buildings. For example, if you set all of your 1x1 residential lots so that they only grow on low-density and medium-density zones, you could assign 1x1 diagonal lots so that they only grow on high-density zones. This would allow you to grow diagonal residential areas without the usual headaches of diagonal growables (having to bulldoze buildings that show up where they aren't supposed to, rampant use of the "make historical functions," elaborate sets of lot blockers, etc.). The same technique would also work for commercial areas, if you'd like to avoid plopping diagonals as much as possible.

The first catch is that you'd need to avoid 1x1 residential (and/or commercial) buildings for high-density growth stages. But 1x1 lots are relatively rare for these growth stages, so this seems like a small price to pay if you want to grow diagonal neighborhoods.

The second catch is that you'd need to stick to smaller lots. However, you could always use ploppable buildings for larger buildings, at least in commercial areas.

 

Edit: Tagging @The British Sausage to get his thoughts on whether this might work, given his experience with diagonal growables.

The issue with using high-density 1x1 residential is that it will rapidly gobble up demand which means less buildings will grow. So unless you use a mod to set demand at infinite (which does mess up growing RCI if that is your method) then yes there will be far more diagonal residential buildings popping up but not many of them. This is because density is linked to growth stages and from what I recall high density includes stages 7 and stages 8- the highest possible growth stages.

There is another problem too: high density zones are really variable density zones based on city population and the proportion of residents/occupants in buildings from stage 1 to stage 8. This is why for small cities (less than 10000 people) high density zoning will nearly always produce small buildings for growth stages 1 to 3.

My trick to cut time is to zone the 1x1 parcels, let the simulator run, some of those parcels will produce a diagonal house- mark those "make historical". Then I plop in the diagonal houses on the squares without diagonal homes and job done. As long as you plop in a residential building on top of a parcel, of same size, which has a growable residential building, then the plopped residential building will not automatically become abandoned. Unless of course there is no demand for residential.

My 2 Cs!

[BartonThinks 3,416]

2 hours ago, The British Sausage said:

The issue with using high-density 1x1 residential is that it will rapidly gobble up demand which means less buildings will grow. So unless you use a mod to set demand at infinite (which does mess up growing RCI if that is your method) then yes there will be far more diagonal residential buildings popping up but not many of them. This is because density is linked to growth stages and from what I recall high density includes stages 7 and stages 8- the highest possible growth stages.

There is another problem too: high density zones are really variable density zones based on city population and the proportion of residents/occupants in buildings from stage 1 to stage 8. This is why for small cities (less than 10000 people) high density zoning will nearly always produce small buildings for growth stages 1 to 3.

Hmm, I'm not sure I follow. The demand should be controlled by the capacity satisfied of the lot, which wouldn't need to be high-density. That is, you could mod the diagonals with a capacity suitable for stage 1 or stage 2 buildings, which would keep the demand curve in line

Also, I believe the variable zoning is controlled by the lots itself, rather than the zones? That is, stage 1-3 lots have a property (LotConfigPropertyZoneTypes) that is set to allow them to grow on low-density, medium density, and high density residential zones (values 0x01, 0x02, 0x03). If you remove the high-density value (0x03) from all of the 1x1 ortho lots, then 1x1 high-density zones should no longer be variable.

Might be something I need to test out in game to see if it behaves the way I think it would. It also occurs to me that the biggest barrier is that you would need to create a mod/patch for all of the 1x1 residential ortho lots in any given user's plugins folder. So this would really only be feasible on a per-user basis. At that point, it might actually be less time intensive to just use the bulldozer + make historical and/or plopped residentials.

[mattb325 28,880]

Without mucking up demand, the easiest way to do this is to assign the houses to one tileset, and restrict the density to high-density only, and assign the lot to a lower stage, and to a corner lot in the road layout (a diagonal road is viewed by the game as a corner lot) and use the normal amount of residents assigned by the PIM-X. For example a small maxis R$$ family home tends to have 13-18 residents.

I have used this methodology for the cul-de-sacs by restricting them to low density/single tileset only and you get a lot of control.

Given that the HKABT team tend to have 1x1 pencil skyscrapers on the Euro tileset, then perhaps the chicago tileset might be an option. If there is nothing else in the plugins that's a high density 1x1, then the game will pick your houses (assuming other city building factors such as health, education, power, water etc are in place). Of course, you'll have to mark the houses historical once they grow to avoid them upgrading to higher densities/different tileset/wealth combinations.

I would caution against wholesale deletion of bunches of lot exemplars which runs the risk of corrupting city tiles. Some exemplars can be deleted, per CPs experiments, but by deleting other exemplars, the houses will grow, however once you save and try to re-enter the city tile, you may run into immediate CTD issues.

[sejr99999 8,343]

thanks to all for your input on diagonal lots  I learned a lot from this forum

[BartonThinks 3,416]

16 hours ago, mattb325 said:

Without mucking up demand, the easiest way to do this is to assign the houses to one tileset, and restrict the density to high-density only, and assign the lot to a lower stage, and to a corner lot in the road layout (a diagonal road is viewed by the game as a corner lot) and use the normal amount of residents assigned by the PIM-X. For example a small maxis R$$ family home tends to have 13-18 residents.

I have used this methodology for the cul-de-sacs by restricting them to low density/single tileset only and you get a lot of control.

Given that the HKABT team tend to have 1x1 pencil skyscrapers on the Euro tileset, then perhaps the chicago tileset might be an option. If there is nothing else in the plugins that's a high density 1x1, then the game will pick your houses (assuming other city building factors such as health, education, power, water etc are in place). Of course, you'll have to mark the houses historical once they grow to avoid them upgrading to higher densities/different tileset/wealth combinations.

I would caution against wholesale deletion of bunches of lot exemplars which runs the risk of corrupting city tiles. Some exemplars can be deleted, per CPs experiments, but by deleting other exemplars, the houses will grow, however once you save and try to re-enter the city tile, you may run into immediate CTD issues.

I didn't realize that about the HKABT team -- that's helpful to know. Given that I stick almost exclusively to North American style BATs and lots, I should be a bit more careful with my assumptions about how common certain types of lots/BATs are.

Ideally, this wouldn't involve deleting any lot exemplars -- it would just mean avoiding 1x1 high-density ortho lots in the first place and/or bulldozing any existing ones to keep things compatible. That would need to happen across all tilesets if you don't want to use the "make historical" function, since switching to a different tileset with 1x1 high density lots could cause those lots to replace the diagonal buildings. To me, the loss of 1x1 high density lits seems like a decent trade-off for a more convenient method of growing diagonals. It's probably not ideal for anyone, but given the number of mods that require some type of tradeoff or another, it seemed like a good idea for users willing to sacrifice those types of lots.

That said, it occurs to me now there might be a problem if you need to override all of your 1x1 low-density and medium-density lots to remove the high density value from the LotConfigPropertyZoneTypes property on these lots. That could cause a conflict with existing 1x1 low-density and medium-density lots in a city. It seems especially problematic if any of those lots exist in an area that's zoned for high-density. In that case, this might be helpful for personal use but not so much for wider release.

[mattb325 28,880]

It is a tricky one, for which there is no easy one-size-fits-all solution: it is also pertinent to note that many of the french/german/polish/UK BSC terrace-house lots also grow on 1x1 medium and high density corners. Again, these are largely confined to the Euro tileset.....but not always.

I think if you only set yours to grow on high density, one tileset and a corner lot, there will be a fair chance that yours will grow first given that the simulator will see a whole road zoned with 1x1s and will select from the potential growables to fill available demand increasing the likelihood that yours will be selected each time a diagonal road is zoned.

[BartonThinks 3,416]

Had to take a couple of weeks off of BATing, so not a ton of progress, but I think this is a big improvement over the last attempt at the day lily prop. Biggest change was switching over to a new script in Gmax that exports proper alpha maps, but I also made some key changes to the model and textures once I got the alpha maps to work.

I'm still playing around with the materials, as well as the size and spacing of the blooms. Not sure this is quite where it should be yet, but it's a lot closer, and fixing the alpha maps has been a huge help. I really felt like I was fighting against the export on the previous attempts.

---

Edit: I made a couple of additional color variations and re-lotted the props against a dirt base texture for improved contrast. Note that these are all the same basic model, just recolored. The clusters at the back of the lot each contain four props.

[BartonThinks 3,416]

First attempt at seasonal flora...

I don't love how the spring and fall versions turned out. There's a "divot" effect that I can't seem to avoid at the center of the leaves, which looks a little jarring and obvious when the plants are clumped together.

Edit: Not going to solve this tonight, but I'm getting slightly better results by making separate clusters of leaves for each flower, instead of a single cluster with a shared center.

[Oidaas 1,650]

Thanks for this interesting subject, inspired me to group my plugins into eras. Made a limekilm for you and every one else.

For Barthon.zip

[BartonThinks 3,416]

On 5/8/2022 at 1:21 AM, Oidaas said:

Thanks for this interesting subject, inspired me to group my plugins into eras. Made a limekilm for you and every one else.

For Barthon.zip

Glad to hear my project offered a spark in some way. I've always looked forward to seeing your work and the way you approach city-building, so that's great to hear.

Unfortunately, I've been swamped with work for the better part of the past year, and after upgrading my computer, I haven't had the time to get SC4 up-and-running again (which is a shame, since the new machine is much better-suited to running SC4). I'll have to bookmark this for when I have a chance to come back to the game! 

Edit: At least got PIM-X back up and running to see how this looks -- fine bit of work!