The silvicultural practice of pruning juvenile stems is a value-adding operation due to the formation of knot-free wood after the pruned branch stubs have healed. However it is not until after the log has been processed that the added value is realized. The motivation for this paper stems from wanting to extract as much of that added value as possible while minimizing scanning costs and data requirements. In this paper we test the hypothesis that increased value can be attained when the cutting strategy is based on a measure of the defect core (DC) size as opposed to a measure of log size. We consider both simplified and intensive methods for determining DC and make small changes in cutting strategies through selecting the cant size from one of five predetermined sizes, according to: 1) Small-end diameter: with the cant positioned to maximize lumber volume ('Best Volume'). 2) Defect core size: with the cant positioned to maximize lumber volume ('Best Core'). 3) Allocation of both cant size and position according to optimal lumber value ('Best Value'). The relationship with log small-end diameter and defect core size to value recovery is also demonstrated. Two geographically disparate samples of pruned Douglas fir logs were sourced: one sample, comprising 97 logs, from the US Pacific Northwest, and the other with 61 logs, from plantations within New Zealand. All logs were measured lengthwise and crosswise in 2 planes (elliptical cross-sections) and, for the NZ sample, deviations of each cross-section from a central string line was recorded (irregular log shapes). The US sample was processed through a sawmill into primarily structural lumber while the latter samples were cross-cut into discs at 20 cm intervals, clearwood cut away with an axe to expose the branch stubs, and the extent of the DC measured at each cross-section ('Intensive DC'). For the US sample, DC measurements were made at the log ends only ('Simplified DC') and were based on ring counts to year of pruning. Each log's external geometry and internal knotty core (generated by computer for the US sample) was reconstructed into digital formats and the digital logs sawn according to the three cutting strategies using the AUTOSAW sawing simulator. The resultant boards were graded into four categories: a) knot-free; b) knots on one face; c) knotty both faces; and d) knotty both faces with pith. Board value was calculated relative to that of the highest valued lumber using weights: 1.0 knot-free grade, 0.85 knotty one face, 0.30 knotty both faces, and 0.25 knotty with pith. Significant increases in yield were noted with the intensively measured DC, not only for the optimal value solution but also for the Best Core strategy. The yield increases averaged 11% and 5% respectively, equivalent to $24/MBF ($10/m3) and $11/MBF ($5/m3) based on a price of US$840/MBF (us$356/m3) for knot-free lumber. Based on our research findings of value yield we conclude that: 1) Small changes in sawpattern have a significant effect on lumber value. 2) Selection of the cant size and position is critical to optimizing value yield. 3) Volume maximization is not a good strategy when it comes to sawing pruned logs. 4) Increased value can be extracted from a pruned log when the DC is derived from intensive measures of the internal knotty core structure Thus a system that includes a non-destructive evaluation tool that scans the log for the internal knotty core, determines the DC, selects the appropriate cant size, and applies an optimization procedure to position the cant would yield significant benefits in terms of increased value. To realize the full potential of pruned logs and achieve optimal value, a complete knowledge of internal defects coupled with optimization capability is required ('Best Value'). However, simplified optimization procedures (aka heuristics) together with some knowledge of the internal defect core can enhance value extraction from pruned logs. This was demonstrated by the 'Best Core' strategy in association with the intensively measured DC. Although there was not a significant increase in value with the simplified DC, more work is required to establish whether or not this result holds in the presence of complex knotty cores. Because of the commercial importance of determining the minimal set of requirements for increasing value this will be addressed in future work.