A. Microsoft recently announced a change to the support statement for clusters of Hyper-V servers. Windows supports up to 16 nodes in a Failover Cluster. Previously, the support statement was 64 virtual machines (VMs) per node, which meant 960 VMs per cluster (15 times 64, because you need a node's worth of hosting available to support the VMs of a failed node, thus 15 and not 16). The new support is a blanket 1,000 VMs per cluster, which means if you only have an eight-node cluster, you could have 125 VMs on each node. (You don't need to keep a node's worth of VM hosting empty in this calculation because you can run more than 125 per node). If you had a 4 node cluster, you could have 250VMs on each node.
Remember that just because you can run this many of VMs doesn't mean you should, or that your hardware actually could. Make sure you perform proper capacity planning, and remember that the 384 per node limit still applies, as does the eight virtual processors per logical processor support limit.