The Big Post on Scale Compatibility

Please post your thoughts and reviews of our new line of Dwarvenite Game Tiles here.
User avatar
Minotaur Lord
Minotaur Lord
Posts: 3668
Joined: Mon Apr 02, 2012 8:32 pm

Re: The Big Post on Scale Compatibility

Post by Harneloot » Sun Mar 13, 2016 6:01 pm

Exhaustive and precise - thanks for collecting and analyzing all that data!

User avatar
Posts: 100
Joined: Mon Jan 18, 2016 6:24 am

The Big Post on Scale Compatibility

Post by Jonathan » Sun Mar 13, 2016 5:48 pm

Good day everyone,

The differences in exact piece sizes between various DF products have caused quite a stir. Some customers don’t mind, some consider it sloppy, some consider it thoughtless, and the response from Stefan and the team has seemed to many to be inadequate.

There’s a lot of speculation about what happened, what decisions were made, and what can be done. I think speculation is helpful for some things, but it can lead to a lot of misunderstanding when a speculation is repeated often enough to be taken as a fact.

I think it's important to be able to pass judgment fairly and based on facts. To this end, I have selected pieces from KS1, KS2, and KS3 to represent approximately 24”, and lined them all up. Here’s what I discovered using a yardstick. Disclosure – I only had 10 catacombs floors and 10 catacombs wall pieces, so those rows are supplemented with comparable resin pieces.

Actual measurements of 12 sets of 12 tiles:
  • Catacombs 1 2x2 floors: 23.50” (596.9mm)
    Catacombs 1 walls: 23.79” (604.3mm)
    KS3 narrow passages [long passage formation]: 24.07” (611.5mm)
    KS3 narrow passages [parallel passage formation]: 23.63” (600.1mm)
    KS1 walls: 23.75” (603.3mm)
    KS1 corners: 24.00” (609.6mm)
    KS1 floors: 23.55” (598.2mm)
    KS2 narrow passages [long passage formation]: 24.12” (612.7mm)
    KS2 narrow passages [parallel passage formation]: 23.64” (600.4mm)
    KS2 elevation blocks: 23.88” (606.4mm)
    KS3 building floors: 24.02” (610.2mm)
    KS2 floors/ground: 23.63” (600.1mm)
    KS3 crenellations: 24.05” (610.9mm)
    KS3 sidewalks: 24.00” (609.6mm)
IMG_20160313_105943.jpg (428.55 KiB) Viewed 1972 times
Other observations:
1. Not all units of each piece type are identical. There’s minor piece-to-piece variation in width, depth, and height.
2. Some pieces are not square. For example, the open floor quadrant of the KS1 corner wall piece is slightly sucked in, such that a long line of corner wall pieces forms an arch when pushed together. The KS1 walls and the KS3 crenelations also have this effect, although to a lesser extent.
3. Pieces within a KS year are not all the same size. The KS1 corner wall pieces are about 50.8mm (along the wall side, at least), but the KS1 floors are about 49.85mm.

Everything above is a fair reporting of what I found in the sample. Below are my thoughts based on these observations, some speculation, and a few more measurements further down.

My hypothesis regarding observations #2 and #3 above: Standard dwarvenite tends to shrink a bit, particularly on narrow/flat parts of a piece, such as floors. Where the floor meets the wall, the structure helps reduce shrinkage in that area. Areas further from the joint experience more shrinkage.

When I’ve done builds solely with KS1 pieces, I’ve found that things don’t always line up as well as I expected. At the time, I didn’t do measurements, I just dealt with it and moved on. However, now that I’ve done the measurements, I see the issue: some KS1 piece types have inherent production flaws that result in a slightly different size/shape.

Was there a scale change?
These sentiments have been expressed many times recently: “DF had a scale for KS1/KS2. DF is now using a different scale for KS3. It may or may not have been a switch from metric to inches, but either way it’s a scale/size change.”

If the intent for pre-KS3 pieces was to be metric, I’d say an ideal result would be an average of 50.0mm, with very little variation from that spec. If the intent was inches, I’d say an ideal result would be 50.8mm, with very little variation from that spec.

The KS1 and KS2 pieces I sampled varied in average size from 49.8mm to 51.0mm, with an overall average of 50.37mm and a standard deviation of 0.5mm. The median was 50.27mm.
Based on the comparison of shrinkage patterns (particularly with the KS1 floors and corners) and the variation in actual piece sizes, it appears that the intent was to make 2” pieces, but something caused various pieces to shrink in different amounts and in different shapes.

I’ve found that KS3 pieces (with the exception of the narrow passages) conform much more closely to the 2” (50.8mm) specification. In my sample, the average KS3 piece (excluding narrow passages) was 50.85mm, with ten times less variation (0.05mm).

In summary: There is no KS1 /KS2 uniform scale. Shapes from KS1 and KS2 appear to have been intended to be 2”, but individual KS1/KS2 pieces (floor, wall, corner, elevation, etc) all had different sizes from each other due to shrinkage that occurred in flat sections of pieces. KS3 pieces don’t have as much size variation from the specification. The more shrinkage a given KS1/KS2 shape had, the more noticeable it will be when used with the non-shrinking pieces from KS3. Some pieces will vary as much as 0.23” per 12.00” (KS1 floor), and others will have slightly shrunken bits (the open floor quadrant of the KS1 corner wall piece) but otherwise be exactly 12” every 12”.

I believe my review of this issue is consistent with Elye’s 3/12/16 feedback: “The DF scale has always been, and will always be, on 1" grid. Changes in manufacturing for the newer pieces (starting with KS3) have eliminated the slight shrinkage that occurred in earlier sets in the process of converting original sculpts to metal molds. This was a change that occurred at the factory, not some sort of executive decision by DF; DF has no control over the manufacturing technology, which has advanced to allow greater precision. Going forward, all sets will be made in this way; KS4 will match KS3 exactly.”

Based on DF feedback on this issue and my observations on the shapes of different pieces, it sounds like there are a number of factors that contribute to shrinkage, including weather conditions around the factory, production technique and technology used at the factory, the material used for the piece, and the shape of the piece in question, among other things. It sounds like some of these things are controllable, but some are not.

I suspect that both the factory and DF have both been trying to find ways to eliminate shrinkage and warping ever since 2013, in an effort to get the final product to come out as close to the intended design as possible. As time goes on, things have gotten better - the factory is eliminating the shrinkage issue, and DF is going to use gorgonite for larger pieces in KS4

Was DF right to call KS1 pieces compatible with CBS and Sewers at the time of KS3?
Compatibility is subjective. Compared to a perfect 50.4mm piece, would 50.401mm be compatible? 50.35mm? 50.5mm? 51mm? 61mm? I think we'd all say 50.401mm is compatible and 61mm isn't, but we may not all answer the same way for some of those middle options.

Every person decides how different is too different to be considered compatible. For example, KS1 pieces can vary from KS1 pieces by up to 1mm (e.g. floor pieces and corner wall pieces), but by my own subjective evaluation, I’d call these two pieces compatible.

I suspect that DF staff largely feel the same way about the 1mm difference in the KS1 floor and KS1 corner wall pieces (or any of the other pieces, all with between 0mm and 1mm variance) – they’re not perfect, but they are certainly usable together even if not identically sized.

I went back and looked at the KS3 photos and videos, particularly the KS1+Sewers photos, and the KS3 build where Nate used KS1 floors as sidewalks between the streets. The photos and the videos showed the size mismatches. The piece placement shows minor gaps in the layouts. I don’t think this issue was hidden; just considered within the normal tolerance for piece size variation.

Given that the difference between KS3 and KS1 pieces was about the same as the differences that some KS1 pieces have among themselves, and given that the spacing out of the KS1 floor pieces was visible in the promo video, I am satisfied with DF’s choice to advertise the pieces as compatible, at least for horizontal builds.

What about vertical builds?
To review this, I stacked four pieces from seven different sources.
  • KS3 Sewers: 8.00” (203.2mm)
    KS3 CBS: 8.12” (206.3mm)
    Catacombs corners: 8.27” (210.1mm) [at tallest point when stood up straight; these bowed out and were uneven]
    KS1 corners, KS2 elevation+floors, KS 2 narrows, KS3 narrows: 7.81” (198mm) [the difference between these four piece types was not perceivable]
IMG_20160313_142729.jpg (320.22 KiB) Viewed 1972 times
My thoughts, based on this sample – KS1/KS2 vertical pieces are all about 0.05” shorter than KS3 sewers, which are about 0.03” shorter than CBS levels. Multi-level configurations will be imbalanced if relying on multiple piece types for support, and structures completely composed of one sort of piece or another will reach different heights from each other.

My personal experience with KS1 pieces is that they really don’t perform well in vertical builds. The above issues notwithstanding, I think most scenarios where I’m personally going to use vertical building, any given structure level is going to be composed of one kind of piece or another, thus eliminating the instability. If two adjacent structures come to different heights, that’s fine too, since the fantasy world structures I am simulating would frequently have not been precisely the same number of inches either. I’ve lived in houses where one part of the house was a few inches higher than the rest of it. It was an old house. It happens.

However, if I’m building sewers/dungeons underneath streets, I can see how it would be irritating to try to combine KS1/KS2 and Sewer pieces under the streets in such a way that it doesn’t cause the streets to be uneven, even if they are only slightly uneven.

Of course, we’re stuck with it now. I cut a bunch of cardboard scraps to 4”x4”, and those should help me fix my vertical builds to help eliminate some of the issues with the heights.
So, are KS1, KS2, Sewers, and CBS compatible for vertical builds? As with any question of compatiblity, it's different for different people, but for me, I'd say that the pieces can all be used in vertical builds, but they are not as compatible as they are horizontally.

Other questions:
“Why didn’t they look at the average KS1/KS2 piece size (i.e. 50.4mm), and scale all the new KS3/KS4 sculpts to be that size?"

Let’s go ahead and assume they anticipated the factory’s newly improved precision in 2015 (which may be an unfair assumption, but let's go with it for now), and could have scaled down the KS3 pieces to be a 50.4mm (100.8mm) scale to reflect the average KS1/KS2 piece size, rather than using the new precision to finally reliably produce at 50.8mm (101.6mm) scale.

The problem is that a good number of resin pieces and KS1/KS2 pieces really are actually 50.8mm (2.00”), which means that if you had future offerings that are 50.4mm scale, the KS1 pieces that are 50.8mm wouldn’t actually be able to fit between them, and the KS1 pieces that are 49.8mm would still be too small, and have exactly the same problems they have with KS3 being 50.8mm.

I feel that, even if they had known about the new precision, using it to make perfect 2" pieces was the right decision.

Final Thoughts
Rumor and assumption are friends to neither the critic nor the supporter. A recent comment on the KS pointed out that the disjointed responses about the issue ("it was an accident", "it was intentional", "it's the factory getting better", "it's not a big problem") come across as dismissive and defensive, which reflects poorly on DF. In many cases the remarks don't come from staff, but from fans who want to help, but instead cause harm.

I'm not a DF insider, just a customer. I hope I've been clear in which statements were my speculation and which ones were actual observations.

One of my pet peeves is criticism of someone else's creation or decision without adequately researching the circumstances and nuances of the situation. Ultimately, I feel that I've examined the issue thoroughly enough that I can now form a pretty good opinion about it. I hope that this review will help others consider the issue of the varying piece sizes with clarity and thoughtfulness.

[edited to fix a sentence fragment and slightly adjusted some phrasing]
Last edited by Jonathan on Sun Mar 13, 2016 6:25 pm, edited 1 time in total.

Post Reply

Return to “Kickstarter Game Tiles Feedback”

Who is online

Users browsing this forum: No registered users and 3 guests