Why is the cost of memory non-linear, while the cost of storage is linear?
I wonder why the cost function of memory vs. storage differ? While storage is associated with a constant cost, this is not the case for memory. Why is that?
Ethereum Stack Exchange is a question and answer site for users of Ethereum, the decentralized application platform and smart contract enabled blockchain. It only takes a minute to sign up.Sign up to join this community
Its due to the way that the memory market works. The average buyer of memory tends to buy it with the computer system. This means companies like Dell and other PC builders tend to buy in bulk but the amount that they put into the average PC is in the 1-16Gb range. The mainboards that most PC builders use will accept a single memory DIMM that is in the 1-4Gb range - so that's what they order - and hence what the chip manufacturers make.
This means that memory manufacturers align their factories to the production of DIMMs in that range so if you want a higher capacity DIMM (i.e more memory) - they need to reconfigure their plants to produce the higher capacity chips - which has a cost associated with it that must be passed on to the consumer - i.e higher capacity memory costs more.
As time goes on though and production is ramped up they will be able to produce large quantities of higher capacity RAM at lower cost - but only if there is associated demand to justify it.
With hard drives - the innovations tend to be in the density of data that can be stored on the same basic device (apart from microwave drives - the basic design of platter drives has stayed the same for a while). This means that the changes to increase density cost much less to implement and can generally use the same plant equipment - so (almost) linear cost.