Filled Up
Ah, the hell wth it. I can't keep apologizing for not writing.
We've got people coming this weekend, and I've gotten the place ready. In between moving things into their proper places, cleaning, fighting dust mastodons and vacuuming, I've been working on my jacket.
Well, the fill stitching is done. 1287.5 hours, 1410.5 yards of thread, a new callus on my middle finger that will stop bullets, and it's done.
Now I start on the GST.
So, taking a break from that, I thought I'd post with an update. This is the first item I've made with fill stitches more complicated than a simple speckle, and it's been... boring. Okay? It gets totally mind-numbingly boring, doing the same stitches over and over and OVER again. By the time I was on the last piece, I wanted to scream every time I worked on the lily or the pomegranate, since I decided I hated those fills the most.
The fills looked like this:
I spent some time at the V&A (while pinkleader and I were staying in London) looking at and making notations on the fill stitches used on the coifs displayed. One of the big problems we have reproducing them is that the thread count of most linens simply can't compare to the quality of 16th century linen. I've counted threads on various coifs, and I've found a thread count anywhere from 75 to 150 threads per inch. The nice quality linen I'm using? 50 tpi or so. And I did quite a bit of sample shopping before I bought my linen, shopping for a variety of things, from thread count, slub factor, density (I have some gorgeous 2oz linen that is airy and delicate and perfect for drawn work and making collars, but it's way too loosely woven for any kind of embroidery), and general feel. I bought pricier linen than I would buy for shifts.
And still, I simply can't get a really high thread count, which means that the more complicated fill stitches that rely on using two or three threads to pass over simply won't work in scale. When you pass over only one thread, you fight a constant battle with the embroidery thread wanting to slip under the weave. Make the design complicated, and you may as well ball everything up and throw it into the corner at the start, since you'll be doing just that sooner or later.
So, the thread count limits the types of fills used. That still leaves a lot of options - really, there isn't a design you can think up that hasn't been thought of before (except maybe if you do a fill that's little spaceships and aliens, but hey). Knowing this, I felt pretty confident about designing various fill stitches. Still, you do eventually run into a kind of embroiderer's block, where finding yet another fill for yet another motif seems terribly tiring.
Fortunately, small variations on a theme produce very different looks.
Such as this one:
This is a basic two thread passing fill, using squares that are then joined at each corner with an X. It's really simple, and looks nice. But, you don't want to use that stitch for everything, so:
Cross the X.
Or:
Put lines in between the X and the square.
It's that simple. And the designs really look different.
While you're designing and working on the project, I really recommend keeping a workbook; in fact, I can't stress that strongly enough. Your workbook isn't just a notebook, it's a reference, an inspiration, and a great way to show people what you're doing (plus: Documentation). I sketched all my possible fills on graph paper, then put all those designs into my workbook (a three-ring binder). Whenever I found another fill stitch I liked, I noted it - and then, when I got bored of the same old designs on the minor leaves, I changed them up. I kept the major motifs consistent, but there's no reason to keep rigidly to one design for the little stuff - I've seen plenty of coifs with varying designs (some even changed in the middle of a single motif - someone got bored, I'm guessing). The overall effect is not changed at all, and everything looks good.
Well, not everything - be sure not to skimp and make your fill too thin. I started on one leaf with just a series of X shapes, and it looked... crappy. Then I simply outlined each row with a running stitch, this is what I got:
Awesome looking, and it made the leaf look like a leaf.
The only way to really know whether your design is good is to work it up - keep a sample cloth handy (or, if you're lazy like me, work the tests on a part of the garment that won't be seen as much. But I really recommend a sampler). The design you draw on graph paper will not look like the finished embroidery. The fabric bends and softens the edges, and what looks crisp and clean on graph paper can look like a hot mess on fabric.
For instance:
While it looked pretty good on paper, it looked just terrible on the fabric. Fortunately, I tested it on the edge of a piece, where it will be under the seam (or covered by gold). This, by the way, is why I recommend a sampler - picking the threads out sucks, and leaves marks on the linen.
My takeaway from this is that counting is tedious - but it does look nice once it's all done. The overall effect is part of the look, as well - do you want it all the same level of intensity, or do you want to create a more varied look? - so making a planning sheet is a good idea, especially if you're going to work in multiple threads or colours - it helps you work out how much thread you'll need.
Here's a partial pic of my work sheet (I'm not showing the whole thing, because you can buy this pattern from Reconstructing History ["Alice"], and sadly, with embroidery patterns, I've found that people will take them from my pictures without asking me if I put up the whole pattern):
Using this sheet, I planned not only the general layout of the fill stitches (with the larger motifs referenced in my workbook), but also the sable GST, and the gold thread (plus spangles, but that's more so I can get a feel for the look, rather than an exact placing reference). I love using a worksheet - I never completely memorized which stitch went with which leaf, and each time I switched to a new pattern piece, I had to remember all over again. With the worksheet and the choices of fill stitches in my workbook, I had a handy reference giude and a lot less stress in my life.
A few more pics of the major motifs:
This has been an amazing project - and I'm still really only half-way. I'm slightly less than 22 months in, calendar-wise, and 2031 hours time-wise. For reference, my first jacket took 1947 hours total. I still have all the GST and the gold work (plus spangles) to do, then I have to partially stitch it together, work over the seams in gold, and then learn how to make lace so I can trim the whole jacket out with lace. I'm not doing the collar and cuffs, because it's a low-necked jacket, so no collar is needed, and I have linen drawn-work cuffs to go over the wrists, but I still need lots and lots of lace to go around all the edges.
It's fun, but it's a seriously long-term project.
Extreme costuming, natch.