This counts how many cells the NRTS processor has discarded specifically from multicast circuits due to crossing the discard threshold. A rising count here, read together with cardNrtsMcastDiscardThresh, tells the admin whether the current threshold is actually too aggressive for the multicast traffic pattern in use, or whether one specific multicast source is genuinely oversubscribing the card. For example, cardNrtsMcastDiscardCount climbing steadily prompts the admin to check whether the multicast discard threshold needs raising or whether one multicast source needs its own rate limit.