Changing Scenery - what would you suggest?

[Ian MacMillan]

I use real dirt in HO scale, and did in N. However I have now switched to using Polyblend Sanded Grout for my base as it has a much more scale dirt texture.

[Ed Kapuscinski]

Ian, where did you get that?

[Philip H]

http://www.modeltrainstuff.com/product_p/330-510.htm

Anyone try this in N scale?

[Ed Kapuscinski]

Yeah, I've had it, I'd avoid it as the "base". It's a decent accent product, but it's not uniform enough, and isn't really fine enough to do the job.

Honestly, start looking around for real dirt that has the right "color" and scoop some up.

[DKS]

I have now switched to using Polyblend Sanded Grout for my base as it has a much more scale dirt texture.

Interesting. Since it's grout and thus will harden on its own, I wonder if it's enough to just wet the scenery surface, sprinkle it on, then vacuum or brush away the excess when it's set (in other words, no glue). Thoughts?

[Ian MacMillan]

Ian, where did you get that?

Home Depot

I have now switched to using Polyblend Sanded Grout for my base as it has a much more scale dirt texture.

Interesting. I wonder, since it's grout and thus will harden on its own, I wonder if it's enough to just wet the scenery surface, sprinkle it on, then vacuum or brush away the excess when it's set (in other words, no glue). Thoughts?

Thats pretty much what I do now. I put it down, spray on some rubbing alcohol and then mist it with water. In areas where there is already scenery I just wet it a tad and then pinch the grout on.

[Carolina Northern]

Great discussion. I've not an expert, but here's what I did.

I model North Carolina - lots of red clay. Dried clay can be powered. I dried some dirt from the area I am modeling, shifted it, and stuck it on a card with Matte Medium.

I took the card to the craft store (Michael's in this case), and matched it to the three closest colors of craft paint.

I brought those back and compared them under my layout lights and picked the best match.

I painted two coats of the craft paint on a piece of poster paper and carried to Lowes and got a gallon of piant mixed to that.

Now I have latex that I can sprinkle sifted dirt into and a readilty available craft paint for quick touch-ups and weathering washes. Latex is not realy good for the quick touchup and is terrible for washes.

My base is usually the latex with dirt sprinkled on or sculpamold with latex mixed in, then dirt on top.

When I recently expanded the Northern, all my colors were readily available.

Never baked my dirt - I use an old cookie sheet on the grill outside - avoids the smell issue in the house.

[MichaelWinicki]

Philip,

Here's a picture of one of my hills that was first covered with a layer of plaster cloth, then covered with thin layer of Sculptamold, then painted with "dirt" colored paint.

As you can see I started adding the ground foam.



I guess the point that I'm trying to make is that I think the coloring of the dirt is more important than the texture.  As you can see, even using a light layer of (fine) ground foam, one isn't going to see much in the way of texture, except for the small knobs and crags created by the plaster cloth/Sculptamold layers.  But, you can see some of the color of the dirt come through here & there. 

To me, the color is more important than the texture, except if you're doing something like a dirt road.

[Ed Kapuscinski]

I agree about color > texture, however my own experiment has shown a little texture is important, especially because, I believe, we tend to over ground-foam things, making them appear too saturated.

[wm3798]

That looks just like William Shatner's toupee!



(yes, I do have too much time on my hands, now that you ask...)

 

[MichaelWinicki]

I agree about color > texture, however my own experiment has shown a little texture is important, especially because, I believe, we tend to over ground-foam things, making them appear too saturated.

Absolutely Ed.

I agree 100%.

I found that the plaster cloth on its own gave too much texture.

The Sculptamold seems to yield a nicely textured surface... At least for us doing an east-coast kind of layout.  If I were doing something western oriented then I'd probably stick with the plaster cloth. 

[DKS]

I agree about color > texture, however my own experiment has shown a little texture is important, especially because, I believe, we tend to over ground-foam things, making them appear too saturated.

You're touching on some key points here that certainly deserve some further discussion. (And I feel a couple of blog posts coming on.)

Mostly I think it's about finding the level of texture that successfully communicates the correct information to the eye. When looking at a real scene, the mind knows it's real, and so automatically makes assumptions about certain features, without much in the way of detailed information. Dirt, for instance, is pretty much assumed without the need for any specifics such as texture.

Thing is, if we reduce the texture of typical dirt (plain dirt, as opposed to dirt/gravel mixtures) to N scale, the result is likely equivalent to the surface texture of flat latex paint. Yet, if we actually used just flat latex for our dirt, it might not always read as dirt. So, we need to coarsen the texture a little--just enough--for it to clearly say to the brain, "I'm dirt." If we go a little too far, though, we could then introduce ambiguity: is it dirt, or gravel, or some other thing I can't quite recognize? Or, at the extreme, we end up with something that's wholly unrealistic.

Ballast size is another example of how this works. Accurately-sized ballast sometimes reads as something akin to mud; it doesn't carry enough texture to say that it's ballast. So, we go just a little coarser to help the eye read what we want it to. It's also related to the "how big is a brick" thing that lets us get away with over-sized bricks.

There are a great many factors to figure into the process. We need props (trains, structures, figures, whatever) to help the eye gauge the scale. Texture versus landforms: we may have the texture nailed, but are the contours of the surface shapes believable? Then there's the problem of how some textures read better to the eye than they do to a camera, or vice-versa--which can be a function of how our stereoscopic vision is impacted in drastically different ways between looking at a model versus the real deal. Not to mention lighting, yet another complicating factor in how textures are perceived.

I think this whole topic can get really deep.

[up1950s]

Dirt is many colors , textures . Where the dirt is matters .

Construction dirt almost looks the same everywhere , though textures vary .
http://kandnexcavating.com/millcreek.html

Top soil and swamp will be damp and dark .

Land fill will be light , dry , and rock strewn .

[Ed Kapuscinski]

I agree about color > texture, however my own experiment has shown a little texture is important, especially because, I believe, we tend to over ground-foam things, making them appear too saturated.

You're touching on some key points here that certainly deserve some further discussion. (And I feel a couple of blog posts coming on.)

Mostly I think it's about finding the level of texture that successfully communicates the correct information to the eye. When looking at a real scene, the mind knows it's real, and so automatically makes assumptions about certain features, without much in the way of detailed information. Dirt, for instance, is pretty much assumed without the need for any specifics such as texture.

Thing is, if we reduce the texture of typical dirt (plain dirt, as opposed to dirt/gravel mixtures) to N scale, the result is likely equivalent to the surface texture of flat latex paint. Yet, if we actually used just flat latex for our dirt, it might not always read as dirt. So, we need to coarsen the texture a little--just enough--for it to clearly say to the brain, "I'm dirt." If we go a little too far, though, we could then introduce ambiguity: is it dirt, or gravel, or some other thing I can't quite recognize? Or, at the extreme, we end up with something that's wholly unrealistic.

Ballast size is another example of how this works. Accurately-sized ballast sometimes reads as something akin to mud; it doesn't carry enough texture to say that it's ballast. So, we go just a little coarser to help the eye read what we want it to. It's also related to the "how big is a brick" thing that lets us get away with over-sized bricks.

There are a great many factors to figure into the process. We need props (trains, structures, figures, whatever) to help the eye gauge the scale. Texture versus landforms: we may have the texture nailed, but are the contours of the surface shapes believable? Then there's the problem of how some textures read better to the eye than they do to a camera, or vice-versa--which can be a function of how our stereoscopic vision is impacted in drastically different ways between looking at a model versus the real deal. Not to mention lighting, yet another complicating factor in how textures are perceived.

I think this whole topic can get really deep.

Yes!!!!

Exactly. I've come to the same conclusion after trying to just use flat paint as a dirt substitute. Scale dirt doesn't work, but so much dirt ends up being too coarse. It's a really delicate act. That's why I like the textured paint in this regard.

I've got the same exact feelings about ballast, and is one of the reasons I'm not so averse to using the woodland scenics "fine" stuff as I was about a year or so ago.

[MichaelWinicki]

Great thread.

Very useful.

I understand what you're saying.  But how much texture do we want to give dirt?  I can see giving it "rolls" and "dips" and basically keeping it from being totally level due to the strata underneath, however I'm not sure dirt merits a texture... But maybe it comes down to where you live.  I know here in western NY, NW Pennsylvania, there just isn't much texture to it.