Shift Dress Pattern


#1

This is a pattern based on a basic shift dress with gathering at the bust one of my employees draped on a dress form. We’re not sure of the exact size of the form as it was old and got recovered, but the measurements are 37-28-38 and we’re calling a size 12. What I hope you find interesting is the method I came up with - a way to code the formulas so that instead of drafting a pattern based on a system, it grades an existing pattern. How I accomplish this is to apply formulas at standard grade points based on the difference between the original draped size 12, and what size we want to grade to.

*I should note that since many of the drafting systems always deal with only 1/2 the pattern, the instructions will always refer measures like the “Chest Scale” which is 1/2 the Chest measurement… so I simply created custom measurements @ChestScale = bust_circ/2, @WaistScale = waist_circ/2, and @HipScale = hip_circ/2. That way it negates having to divide everything by 2 all the time. I simply use it here for continuity.

So here’s the formula for the point at the top of the side seam at the arm hole (B to B1): 9+((@ChestScale-18.5)*0.5)

To break it down … for the size 12 - the width of the back at the bust line is 9"… and based on the bust measurement of 37" the @ChestScale = 18.5". Normally if we were to grade a simple dress like this 2", each side seam will grade in/out 2" / 4 seams or 1/2" at the bust. Here’s the neat trick… if we subtract 18.5 from the @ChestScale for the size 12 we get 9 + 0 x .5 or 9" or no change, but if the bust_cir is 39" and we grade up 2" we get 9 + (19.5-18.5) x .5 = 9.5". If we grade 4" to a 41 bust we get 9 + 2 x .5 = 10"… or if we grade down to a 35", we get 9 + (-1 x.5) = 8.5". By knowing how much a grade point should move up/dn and in/out per 2" grade we can apply this technique at each grade point. With a little more thought one could make use of the nape_b_to_waist_b measure to customize the torso length instead of having to “lengthen or shorten here”,

Shift Dress.val (45.2 KB) Wolf_W_12.vit (4.2 KB) Wolf_W_18.vit (4.2 KB)

A final note on this pattern… in the process I came across a few more bugs in Valentina in regards to the seam allowances and notches when using 0" for a “cut on the fold”… there’s a glitch in the seam allowance at the neck line, and the side of the intersection notch at hem line on the front does not show.


#2

Your comments on grading are sounding like the text I read in a really old pattern book. :slight_smile:

Perhaps rules for “user specified rules for alteration” is something that a future versions of patterns could include? The thought here was that within reason in a costume shop application it would be reasonable to be able to take an old pattern and ‘regrade’ it for a new performer without having to redraft the whole pattern. Various old ‘manuals’ also gave advice on ‘pattern alterations’ which didn’t need a whole new pattern. You noted elsewhere a pattern which distorted if it got too long, This is something “user alteration rules” could note and account for?


#3

Personally, I figuring how to save these somewhere that I’ll find them when I need them. And I’m sure that others are thinking the same. Having them here on the forum means they’ll get lost in the sands of time. However, if one could have them on the tutorials page, under an appropriate heading, I think that will be brilliant :slight_smile:


#4

Valentina has an “if_then_else” operator in the form of ? : .

This could be chained to add, for example, ease dependant on size.

P.S. I think formulae work in tape as well.


#5

I like to add ease by using a formula for example (waist_circ * 1.1) would add 10 per cent of one’s waist measurement as ease. (yeah, that is probably too much, but my point is that using a percent based on the actual measurement will scale whether your model is a giant or a 12 inch fashion doll


#6

Hi @Douglas! Thanks for this post! Keep a list of the issues you find, take screenshots, make notes, post them here. Then when issues are up they can be added.


#7

Yes… I saw that Susan mentioned that awhile back. I’m trying to find places where it could be useful - like dynamically adjusting darts based on the difference between bust & waist.

Hi Susan… Been doing just that on paper. I’ll make a mental note to grab screen shots.


#8

I was thinking maybe this would be a good use for increments? Where you could add an ease increment and apply that to the formulas. By using an increment it’s built into the pattern file and is not dependent on the measurement file.


#9

I suppose you could do it with increments but that seems unnecessarily complex to me. By putting the formula into the pattern (e.g. substituting waist_circ * 1.1 for waist_circ + some number) it is not stored in the measurement file.


#10

It’s ease…y :slight_smile:

Open the Variable Tables via the measurements menu or (ctrl-T). Click the Plus sign… type in your name for the new increment… like #ease… fill in a value in formula. Voila! Now you just use it in your formula like waist_circ + #ease or if you prefer waist_circ * #ease. Now you can change the “variable” #ease once and it is reflected in every formula it’s used in. Increments are stored in the pattern file not the measurement file.

I updated the Shift Dress pattern to add ease using an increment… Note that that total amount of ease added in this examle is divided by 4 per piece, but may differ depending on the pattern.

Shift Dress w_ease.val (45.3 KB)

Wolf_W_12.vit (4.2 KB)

Wolf_W_18.vit (4.2 KB)


#11

Thanks @Douglas. I misspoke when I said using increments was unnecessarily complex. I meant to say that using increments is not a good fit for the issue that I am trying to address.

I understand how to use increments the way you did, it is just not the way I would choose to do it. And note that if I use waist_circ * 1.1 I am adding 10 percent of WHATEVER the waist circumference is, independent on what set of measurements I choose. My point is that I want the formula to be scalable among the most diverse group possible. (and yes, I know that in many or most patterns one will deal with 1/4 of the waist circumference in any given pattern piece).

So, it is great that you have posted an example of how to use increments and it will work for the two examples you gave (Wolf_W_12 and Wolf_W_18).

Now, imagine for a moment that you want to use your pattern to make a garment for a fashion doll which has the same proportions as the Wolf_W_12 model and whose height is 12 inches. It is “easey” to see that adding 2 inches of wearing ease around a waist that small would not give the results one might expect.

So, what I do when I create a pattern, is attempt to make all of the adjustments (like wearing ease) using a formula that is proportional to the actual size of the model. If the model has a 30 inch waist, then multiplying by 1.1 will give 3 inches of wearing ease (adjusted measurement is 33"). If the model has a 24" waist, the formula gives 2.4 inches of ease, adjusted measurement is 26.4"). If the model has a 2 1/2 inch waist (visualize a barbie doll) the model will give 0.25 (a quarter inch) of wearing ease for an adjusted measurement of 2.75 inches.

I am trying to demonstrate that the only value in using #ease in the formula would be making the formula more readable to a human. It may be a good idea for that reason alone. I am also trying to illustrate that using a formula that ADDS ease will work for only a limited portion of the potential models that may be in the target audience. That is why I prefer to use a multiplier rather than an added value. In fact, I assert that if you want to ADD an ease constant, the increment table in the pattern file is NOT an ideal place to keep it. Because it is so dependent on the size of the target, it would be better to use a custom measurement and store it in the .vit file.


#12

@Douglas, I went back and looked at one of your earlier posts:

I believe that your suggestion of using the nape_b_to_waist_b to customize the torso length is right on. That is a good illustration of the concept that I am trying to illustrate when I suggest using a multiplier rather than a fixed increment to adjust a pattern.


#13

I got that… that’s why I said “if you prefer waist_circ * #ease”… hard coding the formulas with 1.1 makes it harder to change the amount of ease later on. Picture someone else using your pattern and they want to change the ease amount - you may know where every instance of 1.1 in the formulas is, they won’t. Not to mention that the fabric used will determine the amount of ease required… IE: knits may require a negative ease.

I also realize that one might want to put the ease in the measurements so that the ease remains with say a group of actors where different actors may need more or less ease depending on their size or what their role requires.

In practice I never use ease per se… we just build most costumes with 1" side & CB seams so we have a built in 3" of ease to let out.


#14

In practice I never use ease per se… we just build most costumes with 1" side & CB seams so we have a built in 3" of ease to let out.

It sounds like we have something in common. I do most of my patternmaking for costumes. I specialize in finding sources of “extra” in garments so that I can make them fit different actors


#15

Since the majority of our business is costume rentals, we have to think in terms of adjustability… more so than even theatres. Still a lot designers / theatres we do custom work for want extra seam allowance whether it’s 3/4 seams or adding extra to side seams. Probably the only exception in 30+ years was a recent job we did for the David Byrne - St Joan musical, where we had to make these long fitted coats out spacer knit - which is kinda like working with neoprene. We had to put them together like scuba suits… I was dealing with 1/8 seam allowances!


#16

The first big show I worked on at this theater was Joseph and the Amazing Technicolor Dreamcoat. We had 30 female chorus members wearing white cheesecloth robes with long neck bibs made of a “sandwich” of heavy cotton, really stiff interfacing, and gold lame. Other than being a nightmare to work with (especially sewing the cheesecloth) the costumes were beautiful. The costume designer was right about how good they would look under the lights but the seamstress who sewed most of the cheesecloth offers to become violent whenever anybody suggests doing something like that again