I “mixed” this up this morning:

I’m not sure I like the addition of the green; the value difference between the green and yellow makes a pretty visible value difference in the center of the “shawl”.  But if I “time” the green so it coincides with the lightest part of the turquoise,  perhaps that will help.

Regardless, I’ve learned a few things:

• Strong color contrasts (opposite ends of the color wheel) will add a lot of “vibration”.  This looks ok to good if used as pattern, but is very distracting when used as background.  An interesting “solution” to this problem might be to make the background on the reverse side dominant in one of the two colors, e.g. a 3/1 twill rather than a 2/2 twill (which is how it is currently set up).  I’ll have to try that with a different draft.
• Because of this, the overall appearance looks best if the weft gradient colors are (on the color wheel) located to either side of the warp that it weaves with.  In this case fuchsia/blue is weaving with turquoise, green/orange/yellow/red is weaving with yellow.  These are analogous colors, so look pleasing without too much “vibration” or dulling-out of colors.
• Differences in value are also very visible and will produce a “stripe” effect even if the gradients are smooth.  The results that are most pleasing to my eye (at least in Photoshop) are those where there is only one gradient across the length of the entire shawl.  It might be possible to get a pretty result with lots of gradients across the entire shawl, but I think you’d have to make the pattern extremely simple, otherwise it would simply be too “busy”.

I’ve also decided that learning is easier if you’re only turning one set of knobs at a time.  If I were trying to understand the color interactions while simultaneously messing around with structure, it would be much harder to reach conclusions.  I’d never know for sure whether the color interactions were different because of the weave structure or because of the colors themselves.  As it stands, I’ve learned a lot from studying color in just one weave structure, and moreover a weave structure that weaves identically (only with reverse colors) front and back.

Now that I’ve a better idea of what happens with colors, I think I may move on to considering what can be done structurally to differentiate front and back – using the same profile draft, but “filling in” the draft with different  structures on front and back.

After that, if I still have time, I’ll play around with different profile drafts, to give wholly new patterns, though I don’t think that is strictly necessary.  I think I can get enough interesting variations with a single profile draft to occupy me for three shawls, and I don’t think that doing more profile drafts will teach me much more than fiddling with a single draft.  I already understand the process for layering structures using Photoshop, so that would just be a repeat of what I already know.

I’ve begun to calculate what I’ll need for this project, in preparation for dyeing yarn.  Because I need such small amounts of any given color, a small skein goes a loooong way.  (For weft yarn, I think I only need 3 grams of each color per weft, if there are 30 colors!)  So I am thinking I will either dye 30 gram skeins in a quart jar, or 50 gram skeins in a half-gallon jar.

The trade-off here is time: I can fit four half-gallons or 7-8 quart jars into my big dyepot.  This means that, to get 60 colors for a two-gradient shawl, I’d have to do fifteen batches of 50 grams, vs. only 8-9 batches if I use smaller skeins.  That’s 3 weeks’ worth of dyeing vs. about a week and a half, assuming I can get  one dyebath per day on most days.

The advantage of dyeing bigger skeins is that, as it is nearly impossible to match colors perfectly, it results in less waste – I can use a single batch of colors for a long time rather than having dye another batch and throw the leftover colors away.  (Or, rather, reserve them for undetermined “other uses”.)  The downside is that it takes a lot more time, and requires bigger upfront investment.  (90 colors, which is what it would take to do three color gradients, times 50 grams apiece = 4500 g or about 10 lbs of silk yarn!  Even purchasing cheaply from India, that’s $300-350 worth of yarn.)

So I’m still mulling that over.

All is (relatively) quiet on the wedding front.  Pretty much everything is settled at this point, so it’s just a matter of greeting the out-of-town guests as they arrive.

My fingers are twitching to get into the dyepots.  Here’s what today’s Photoshop simulations turned up:

Here the blue/purple background is transitioning gradually from fuchsia at the edges to turquoise in the center, both along the warp and weft.  The other layer (orange/yellows) has a solid yellow warp, and a weft that changes colors gradually from red to orange and back again.

Here’s the back view (please click for the larger version, the resizing function blends out the colors unattractively in the smaller view):

I like this pair quite a bit.

Then I said to myself, “Too much is not enough – what if I did FOUR gradients?”

And of course I had to answer my question:

Ooh, pwetty!  I MUST try it!

Of course, first I need to dye another 60-odd colors, so I can get those smooth color transitions in fuchsia/blue and red/orange…and finish weaving this warp so I can start up a double color gradient warp…

…and I can’t do any dyeing until AFTER the wedding (T minus 6 days!)…

…so instead, I’m sitting on my hands and saying “MUST….NOT…DYE…YARN!!  MUST…NOT…DYE…YARN!!”

I’m sure it would be amusing, if I weren’t so eager to get into the pots.

Meanwhile, the ceremony is finalized, the music selected, vows written, my little speech to the groom sent off to the officiant…pretty much everything is ready.  So I’ve been burning off my nervous energy by cleaning up the house.  It’s not neat by any means, but it has less stuff cluttering every surface than it usually does.