From the Anderson & Sheppard blog, The Notebook.
"One of the most important aspects when cutting a checked cloth is to match it on both the top side and underside. This might seem simple, however more often than not the top layer of cloth has to be shrunk as the checks are not matching exactly as they should be."
There is something in the blog post linked to by tailleuse that puzzles me, even after pondering on it for some time.
In the first photograph one sees the cloth laid out and the front pattern (rather boldly) chalked out. The second photograph shows the right front cut out, with some overwidth it seems, lying on top of the second layer in order to match both layers accurately. The blogger, Matthew Borkowski, describes the process: the layers are matched, on the seamlines one assumes, then pinned together and the resulting top layer's bumps and fullnesses are shrunk away so that both layers are flat and perfectly matched.
All fine, but what I don't understand is this: shrinking away fullness necessarily changes the weave. The pattern has already been marked on the cloth, so that means the pattern may change as well during this operation and the possibly distorted pattern then forms the basis for the cut.
One may argue that the amount of distortion is likely to be small, say 3 mm or so. Possibly, but another location on the same piece may also need a 3 mm shift and that may well be additive. Besides, I've handled double layers that were quite a bit more out than 3 mm.
Suppose the possible distortions in the pattern are dealt with at the fitting stage, although those distortions are unpredictable from a pattern drafting point of view. Necessary alterations during fitting would then, in general, be recorded to be kept with the customer's data or used to modify the pattern directly. But as the pattern has been unpredictably changed from the paper original, those notes or changes would also be random and therefore worthless. Discarding the notes however would mean extra work for the next pair.
I can hardly imagine Anderson & Sheppard sacrificing reproducibility in this way, so there must be something I'm missing. With no opportunity to ask them and so many Professionals, Senior Professionals and Super Professionals contributing to this thread, I venture the question: what is it that I'm overlooking here that ensures pattern accuracy as well as fabric congruity? Enlighten me, please.