Subnetting issue

I have a problem with Subnetting question in my exam on Networks and i am currently trying to prove that the answer to the question is incorrect

The question is :

Suppose you are given an network ip adress 128.20.224.0/20
and asked to subnet this network into two subnets by allocating 1000 computers for first subnet and 750 for the last

And the answer is :

First of all - this network ip is one of 14 subnets of the network 124.20.0.0 ( classs B), it means that the network 124.20.0.0 was subnetted and now i need to subnet one of subnets, but using the algorithm of standard subnetting 2^n - 2 = number of subnets desired.

Now, I am not agree with that because the standard subnetting deals with division of the networks of class A,B,C and not of their subntes.
And there is another algorithm for recursive subdivision of subnets which is called VLSM, where i allocate bits for sub-sub-nets according to my need ( 2^n = number of subnets )

Am i right?

Slaks

my answer would be 128.20.228.0/22 and 128.20.232.0/22 each would be able to hold 1022 hosts.

not sure where you are going with the 124.20.0.0 that is a class A address.

here is why my answer is what it is...

128.20.224.0/20= range 128.20.224.1 - 128.20.239.254 broadcast=128.20.239.255

now lets put into binari 10000000.00010100.1110|0000.00000000
the | shows where your network mask ends now barrow two bits...
10000000.00010100.1110|01|00.00000000
the second | is the new subneted mask barrowing bits from the 128.20.224.0 hosts which added together = 128.20.228.0/22

And i am not agree with your answer cause pay attention that we are talking about subnet adress 128.20.224/20. This is one of 14 subnets from 128.20.0.0/16 right? There is two algorithms for subnet division : 1.Subnetting and 2.VLSM
It seems to me that you are working with the first one but subnetting doesn't allow subnet subnetting becasue length of the bitmask must be the same. If you take VLSM it needs only 1 bit to represent 2 subnetworks. But then you can not control number of computers in the sub-sub net because it is the same number.

believe me I did classless routing (vlsm). I dont know how to explain it to you any better then I did. you are confused on classfull and classless routing. and what is your point with it being one of 14 subnets? in classless routing you just need to be under the person supernetting your block.

That is the point - you did classless VLSM ( excluding all-zero,all one adresses) but we didn't learn VLSM on course, we learned classfull subnetting only. As far as i know it is impossible to subnet subnets in classfull subnetting because the algorithm doesn't permit to do so ( static adressing)