Comments on "Analog layout using ALAS!" [and reply]

For the original article see ibid., vol. 31, no. 2, p. 271-4 (1996). In the aforementioned paper the authors consider the advantages of laying out matched pairs of transistors, resistors, capacitors, etc., with interdigitated geometries and common centroid. Then they show six examples of "matched devices". However, the commenters point out that not one has a common centroid. This leads to the conclusion that the computer program ALAS! seems to have a remarkable consistency in making layouts where the centroids are never in common. In reply, the authors thanks R.A. Pease for his comments on their paper and respond to his concerns, especially clarifying the issues regarding common-centroid versus interdigitated layout.