The marginal cost to fabricate the GPU die with the full ram allotment, and then disable a portion for 1) salvaging otherwise useless die's with partly bad ram for lower ram use cards; or 2) market segmentation; is tiny compared to the marginal cost of creating X different masks and fabricating X different die's, one specific to each market segment/ram size amount.

In addition, fabricating the same die for every SKU point in the product line means the maker does not have to guess how many version X vs. version Y vs. version Z dies to fabricate. The just have one fabricated. Then the dies become SKU version X, Y or Z as demand requires during the card assembly process.