Yield is calculated for logical tomography circuits shown in Fig. 2b,c for the error-suppressed (error supp.) scheme with feedforward (blue circles) versus post-selection (blue squares); a standard scheme is shown for reference (orange squares). All rates use datasets reported in Fig. 3 where error bars represent 1 sigma from bootstrapping. The shaded area of the graph shows the increase in yield for the error-suppressed scheme using feedforward (FF) compared with the post-selection (PS) scheme or the standard scheme. The optimal acceptance rate assuming no noise is 75% for the feedforward scheme, 37.5% for the post-selection scheme and 25% for the standard scheme. The observed acceptance rates are because of the additional detection of errors. We estimate the yield in the presence of noise in the section ‘ Estimates for magic-state yield ’. We observe a stark difference in yields between experiments conducted with the logical tomography circuit shown in Fig. 2b,c , shown to the left and right of the dashed line, respectively. We can attribute this to the depth of the logical tomography circuit, in which deeper circuits, such as those shown in Fig. 2c , are more likely to introduce detectable errors. This is discussed in the section ‘ Estimates for magic-state yield ’.
