Just how much money is there in the world? According to the Bank for International Settlements, the total amount is about \$5 trillion, but the CIA believes, the total amount is \$80 trillion if you include “broad money.”

The total supply of Burst is 2,158,812,800 with each Burst being divisible into 100,000,000 parts. These smallest units of account are called “Planck”. So this results in a maximum supply of

215,881,280,000,000,000 Planck

If 1 Planck had to represent 1 US-cent for being a viable unit of account, the whole current world economy would have to grow 27-fold before Burst would “run out of Plancks” covering every single US-cent of “broad money”. At this point in time 1 Burst would be worth

It’s safe to assume this will take some time. Now that we had our share of fun, let’s look at the technical reasons for this article.

Internally, the Burstcoin wallet uses 64bit values to store Burst amounts, where in fact these amounts are handled as Planck. These 64bit can represent a space of 18,446,744,073,709,551,616 values, whereas we know from the above that Burst only could make use of 215,881,280,000,000,000 distinct values – provided all Burst/Planck in existence would assemble in just one address. So for our purposes we need only

log2(215881280000000000) = 57.5830157595391890309

bit to represent all Planck in existence. We have to round up, so it’s 58bit. Therefore, we have 6bit “wasted” or “reserve”. Now these 6bit could be used in a future refinement of Burst granularity and Burst would not be divisible only into 100,000,000 Planck, but we could actually make it divisible by 6,400,000,000 – say “HypoPlanck”.

If we did that, the world economy had to grow 1728-fold of its present level, before we would run out of said “HypoPlanck” each representing 1 US-cent (and 1 Burst being worth \$64M). All of this, of course, assuming Burst would run the whole world economy.

I therefore propose to treat the Planck granularity as final and the maximum required bit width to represent a Burst amount (in Planck) as 58bit. This means we do have on many places the opportunity to make use of additional 6bit of information whenever a Burst amount is represented.

Let’s take for example the data section of a Multi-Out transaction. CIP4 defines it as

<Id1><amountFor-Id1><Id2><amountFor-Id2>….

where Id and amount use fixed 8byte (64bit) fields. We now know, that the most significant 6 bits of the amount fields are free and in a future hard fork could be used for some additional information. I do not want to suggest any specific usage right now, but rather use the opportunity to stipulate some brainstorming/discussion if and what these 6bit could be used for. Consider this being a RFC.

## Subscribe

### 1 comment

1. CurbShifter says:

[ repost of my comment on reddit, seems more appropriate to it post here ]

I am no proponent of this idea. Leave the full 64bit purely appointed to the amount.

With the total amount of tx to date of 7646727. this should amount to a negligible ~5.7 MB in ‘wasted’ space in the db.. ?

However, when those 6 bits are bit-masked in BRS there should be no discrepancy if its ‘wasted’ or designated ‘reserved’ space. And it would be(come) backwards compatible if those bits are going to be used at a later point.

If using those 6 MSB is a real consideration I would start by masking them now if they are not. It would make a HF less H.