Well, I had a go with combining what I did with the original tiled house. It's not great - mainly because the map files are quite different and the combination was a crude mask that erased some of the snowy version and its map file to reveal the original house underneath it. The map file for the tiled version is very steep. The map file for the snowy version is very shallow - so that the snow doesn't bleach out to being completely white all over. There are a few strange patches you can see on the darker parts of the roof to the south where the resulted blended map file is neither one thing or another.
Let me know what you think.
@JulianDracos - I live in an area with plenty of thatched buildings, and since thatch is one of the greenest roofing materials available that isn't likely to change very much. It is also about 10 times more insulating than tiles unless you have excellent insulation in the attic space of a tiled house. So the snow does tend to stick and stay much better on thatched cottages than it does on tiled houses. It's therefore not inconceivable that you would get some wind erosion of the snow on tiled rooftops and/or melt off. The tiles are warmer because they don't keep the heat in as well as thatch does.
The other question I have is: does the snow melt into tidy sharp-edged patches, or is there a blend like I've created above where the snow shades out gradually into a melted patch?
Here in the U.S. some locations have very slanted roofs because they get snow every winter. But other places may get snow only every 5 or so years. Those roofs aren't slanted much. Some places may get snow every 5 to 10 years. so they don't worry aobut it. They do get lots of rain, particularly in hurricane zones.
I searched online for "time lapse snow melting" videos and watched a few with visible rooftops. Interestingly, the snow patches on rooftops tended toward hard-edged, while the ground cover tended toward more diffuse edges. I suspect that there are enough factors that affect the shape of the melting snow that any consistent pattern would be plausible.
I certainly would expect the shape of melting from a single moderate snowfall on an inhabited home's roof during a warm snap to be different from the massive ice blocks forming from a winter's set of snowstorms that vary from wet to dry (warm storms to cold storms) and fluctuating temperatures (icicles and ice dams require temperature cycling).
Thank you, both :)
@jslayton That's very interesting to know, and partly what I suspected. I have vague memories of this being the case on tiled properties around me the last time the snow settled a couple of decades ago, but they were only vague memories. I also remember that the snow didn't stick in the cracks between the tiles, but in the middle, which is why I think I remembered it in the first place - an odd effect probably caused by heat escaping between the tiles and making the cracks between then slightly warmer than the tiles. But it also makes this problem a lot easier to solve. I can use a much narrower fade edge between the two map files, so I should be able to get rid of that blotchiness.
Hi Sue! Not sure if this will help... We just had 12 inches of snow 4 days ago. Here are a couple of pictures of melted roofs looking out of my kitchen window. There are no tile roofs in these but I can tell you that the snow melt is the same on tile. You can which have insulation and those that don't.
Thank you very much, Shessar! That's really useful :D
Having lived only in places with tile or slate roofs, I can confirm Shessar's photos are indeed entirely accurate for both those materials. Slide, wind effects and thin ice/snow melt basically works from the top down, and outer roof edges facing more nearly into the current wind, so those will all tend to clear of snow first. This can be enhanced around objects sticking out of the roof like chimneys and stove pipes, especially where those are in use. There may also be some smoke discoloration of the snow near chimneys that are in heavy, regular use as well (albeit that also tends to mean snow there will melt faster as well - introducing foreign particulates to the ice/snow helps it melt faster generally, like applying salt to road and path ice).
Roof patches do tend to be harder edged than you've illustrated so far, and with a tendency to remain in the hollows a little longer than the ridges on shaped tiles, as you've already noted. There are a lot of variables however, and commonly, once the snow's started to melt on such roofs, it will tend to clear fairly quickly thereafter, unless there's fresh snowfall heavy enough to fill-in the cleared gaps.
Thanks Wyvern :)
I've also just noticed from Shessar's shots that the snow may be staying longer either between rafters, or on the rafters. There's certainly a vertical striping element to the general pattern.
Since these rooftops aren't modern ones I think we can safely say there's no insulation, so the bare ridges and top levels seem to be the thing to do, though I won't make them too different to the thatch cottages or they won't look like they are part of the same set. So just a little bit of ridge and roof bared.
The vertical striping could be due to the shape of the roof, with the underlying rafters pushing out those parts directly beneath them, so leaving a hollow away from that (the difference is likely to be very small, so wouldn't show as variations in the roof's appearance otherwise, except maybe under very low-angle lighting). I agree the rafters will be likely the defining point though.
Keeping the look fairly straightforward sounds like a good plan!
Here are some pictures from seven years ago in mid March in an area which doesn't have snowfall that melts right away. The reddish house in the first image has a tin roof with snow brakes installed. The other houses have shingles.The buildings in the second two images have tin roofs. As this is mid March we were entering warmer weather so the snow is starting to shed from the tin roofs. This can be seen in the bottom image where the edge of the roof is visible and you see the snow piles on the ground next to the building. The second image is to show how the buildings blend into the snow on the ground. If these were taken directly overhead it would be harder to make out the building. Hope these help.
I also suspect sun exposure is a primary factor in terms of snow melt on rooftops.
So depending on the direction the house is facing, and the surrounding temperature factors... The snow might be getting sunlight but the morning might still be really cold, and in the afternoon the sun might have moved to another side while the day has reached its maximum temperature.
I guess I'm saying it might be fairly arbitrary how the snow melts when you take out directional facings.
Thank you very much, Jeff :)
Once again the ridges are nearly exposed, but I think that in order to keep the tiled houses interesting I will have to assume that the snow of Winter Village is not so dry, and expose a little more than that.
DoubleDouble - this is very true, but unfortunately I can't predict which way around any of the houses will be pasted, so I will keep the coverage about the same on both sides.
Yes the ridges are exposed in those images if I were to take some images today they would not be. You are looking to show something that is not a blob / bump in the ground snow so drawing them that way is understandable as you want to do a single set of buildings not two different styles showing heavy snow and light snow as was previously discussed. Glad I could help. Looking forward to the March release.
What we need is a sheet effect - melt snow. That way, based on the global position of the sun, the roofs will show the appropriate melt coverage.
Sue, the vertical striping that you see over the rafters is because the wood of the rafters acts like an insulator from interior heat so the snow melts slower there.
I just want to add that I love what you are doing with this. We really needed a winter city set. Happy, happy!
I don't remember where, but I have seen no snow on the South side of a small hillock, and some snow in shadows under bushes on the North side of it.
edit: Very likely in the Spring.
@Shessar - I think you are probably right about the rafters, and thank you :) I think I remember a map you once did where you were trying to create a snowy scene not a lot unlike this one. Its been in the back of my mind for a while now ;)
@JimP - yes, that's definitely more of a spring theme than a winter one, but I know what you mean. Towards the end of the last inch of snow we had here it hung around in the shade for at least 2 weeks longer than it did in the sunny places.
On the progress side (or lack of it today) I've had quite a struggle getting the gradient shading of the map file correct so that the hollows around the dormer look filled with snow, and everything the right size and shape to be the same as the original but with a layer of snow on it. Not sure how all this is going to work when I start combining them, but we'll see in the morning now.
My only suggestion (which you are free to ignore because it is rather nit picking) is that the capstone of the chimney would be much too warm to be fully covered with snow. But honestly, I think that it looks amazing. 😍
Sun in a big factor, but with minimal or no insulation, so is the heat generated in the house itself. In a place that gets snow (I grew up in Nebraska), the house in literally uninhabitable without heat, and heat rises.
@Shessar - I was umming and ahing over that as well. I tried it both ways, since combining half of each or bits of each proved to be impractical. The original chimney colours are a bit yellow compared to the snow, and shout really badly if I use them instead. Maybe I can find a solution today now that I think I've got the dormer right.
Something a bit like this, then... Or is this too melted to go with the thatch?
Oh shoot! I forgot to cut the chimney extent out of the map file, so its a lot brighter than it should be, but you get the general idea.
I vote for more snow cover.
I like them as is. I like the single tile covered in snow. You can probably get away with a little more snow, but I think if you add much more, then you are going to have all of the tiles covered.
Two different views...
Since the same thing is happening on FB where I asked the same question, then at a guess I've got it about average, which is what I wanted. Some of the buildings might have more or less snow, just like in the photos we've seen above.
The snow still lying in the dormer valleys and clinging to individual tiles...perfection!
Having a mixture of both styles would be great so long as you have time to do them both.
Thank you, Shessar :)
It's all one set (or intended to be one set). There are 10 thatch cottages and 14 tiled buildings. Additionally there are stone buildings that I may or may not get time to do, even though they would be relatively easy since they don't have map files for the most part.
I've done the trees, but I also have all the connecting wall and hedge symbols yet to do. And then there are the additional things that would be nice to have, like sleighs and a couple of other things.
I might have time to bare a bit around the chimneys on the thatch, but I don't know just yet.
Having fun with different kinds of dormer. This is the same basic building...
Not sure about the "concrete pipe" dormers, mostly as they don't match well with the tiled roofs here, but that tiled "house extension" version looks particularly interesting, as an option to set up "L"-shaped buildings. Might need to be a little larger, so as to run down from the main house ridge, while still protruding a good way from the main house wall, perhaps?
LOL! They're meant to be lead arch dormers, but maybe the colour has gone off somehow - too yellow against all the blue snow? I will check the original buildings.
Bits of house would be the only way I could think of allowing people to make their own snowy building, since it's going to be impossible to make a roof fill that is snowy in all the right places to be used in the House tool, but we will have to see if there is time.