Layman term please.
The txout is an output of any transaction that credits its destination address with some value.
The PoS works this way: The wallet iterates through all available txout's that are not spent. For each txout it calculates its Proof-of-Stake hash value that is a function of several data elements associated with this particular txout. If the calculated hash value divided by that txout's coinage is less than the current PoS difficulty then the so-called PoS kernel is found, so a wallet constructs a PoS block using this txout and submits a block to the network.
When the txout is spent its coin age is reset to zero be it a PoS block generation transaction or regular spent transaction.
Thus, the less unspent input transactions you have, the less blocks you could create. Provided that the unspent txout becomes eligible for PoS block creation not sooner than it gets 7 days old (the period we are going to select) and the target time is 1 minute, we need to have at least 10080 txouts to sustain the target time of 1 minute at the base PoS difficulty. The difficulty is a measurement of a network security, the more txouts eligible for staking a network has the more is networks security and the more complicated it be to make an attack on a network.
If every of the initial 170 holders would merge its coins into a single txout then the network would generate 170 blocks in a week period.
Thus, to ensure that a network creates at least 10080 blocks in a week period we need to force holders to keep their stakes split enough. The only possible way to achieve this is to limit the maximum reward per PoS block. Hence the maximum reward per block should corresponds to the txout value of ~100 coins (1000000/10080). To make some reserve for difficulty to decrease we need to lower that value even more. I would propose to set this limit to the value that corresponds to the 50 coins input at its age of 30 days (the maximum coin age after which the probability to find a block is not growing anymore).
To 'bootstrap' the network initially, so the holders would not need to split their stakes manually, the coins distribution script that I have to develop will generate the series of transactions per holder address.
Okay finaly i understood something. Since this is the initial parameters, we can test your proposal out. Shouldnt be a problem imo.