Damien,
The length of the connecting rods becomes a question of piston projection (protrusion).
The Molnar rods are intentionally about .004" shorter than the most common length VW connecting rod. There are a few rod lengths that are shorter, but not many. Those OEM blocks built with 'short' rods would require decking in order to get proper protrusion of the piston. Usually, when we figure piston projection with the Molnar rods, we most often use a 1-hole head gasket. The length of the OEM connecting rod can vary .040" when .004" makes a difference. We are not going to make a variety of different length Molnar rods. So, with our Molnar rods, there are cases that the block will require decking.
Lengths of OEM rods for the older AHU and ALH engines do vary to adjust for crank to block height, and the engines can vary a lot. The misconception is that a rod marked '65', which is very common; that all of the same numbered rods are the same weight and length. This is not true. One time, just to prove the point, we matched rods marked '54', '57', 63' and '65' in a set, both for weight and length. We have an engine running with that 'matched set'. the only way to know is measure each VW rod by length and weight. So, in our opinion, the only way to know is to weigh and measure each component. In spite of claims, we have taken apart virgin engines with the average imbalance between reciprocating sets at 8 grams and between reciprocatings sets, 6 grams. The Molnar connecting rods have a variation of not more than 1gr +/-. It definitely makes a difference to balance the rods and pistons, and get projection accurate as the engine will run noticeably smoother. From the AHU to the ALH, some factory improvement in weight and length occurred. The PD motors are much more accurate, one to another. We don't have enough information to determine how the Common Rail engines run, but I think VW continued an upward trend.
All that said, as a rule we find that the block needs decked simply to remove the 'valley' down the middle of the block. At each of the webs between the cylinders, there is a propensity for the sides of the block to become higher than the line down the center of the block. In other words, the bolts 'pull' the head up on the sides. When heated, the cylinder head tends to warp lengthwise, high in the middle; the block tends to valley the opposite direction; low in the middle, for the length of the block.
When measuring the block projection correctly down the centerline of the block, the fact the block at the bolt holes is high compared to the centerline cylinder webs will not be seen. What is worse, it is difficult to measure. The only thing we can offer 1) Strike the top of the block with a fine, single cut bastard file, or better, 2) lay a machinists block side-to-side on the head bolt holes in line with the cylinder webs between 1-2, 2-3, and 3-4, then shine a bright light from behind the machinist's block. Either method, you can witness the gap. Although you can pinch a .001" feeler gauge in the gap under a straight edge, on average, we remove .002"-.005" to cleanly flatten the block corner to corner, while just touching the center webs.
The thought is if you are going to make the cylinder head flat, so also should you make the block flat for the best mating surface. In this manner, there is very often, NO CHANGE in piston to deck height (protrusion) and the same head gasket is used as before machining the block's deck.
The only other point is that the bottom of the block and the head gasket surface are not on the same plane. The block is decked at the factory using the main journals. In order to match their results, the block should be milled from the main journals. Instead of building that jig, our method is to either shim from the block's base so each corner of the block touches the cutter. Otherwise, flip the block upside down and deck the bottom first, so both top and base are on the same plane. Yes, I know... it's a PITA. But if you don't correctly deck the block, the result is usually the #1 piston will protrude .004" more than the #4 piston. .004" additional protrusion = one size difference in head gasket thicknesses and a .5 compression ratio difference. Decking the block properly adds time and trouble to the build, but removes an avoidable weakness between the head and block along with a compression error consequence.
I hope that clears things up...