Honestly, that is the strangest thing about this whole article. Why would the Soviets mint a coin that can't be combined with other coins of the same type to make a full ruble? It just seems really unnatural.
whenever you see something that makes you think "why" like this, 99/100 you can blame it on PR/marketing.
I can't think of a single other currency with a non-divisible coin (although I'm sure they exist historically). Even the old pre-decimal pounds/shillings/pence were all neatly divisible.
https://ru.wikipedia.org/wiki/%D0%9F%D1%8F%D1%82%D0%BD%D0%B0...
The basic gist of it is that it's of 18th century, pre-Soviet origin, back when kopeks were srs money and there it was useful to have denomination between 10 and 25 kopeks.
In Soviet times, retail prices were set centrally and printed or sometimes physically stamped on the products themselves so things adding up to a ruble is not something central planners had to worry about too much. You could guarantee prices in cast iron as seen on the handle of this skillet:
https://www.rbth.com/history/332037-ussr-vending-machines-vi...
We're all raised with stories of the famines and the gulags and all the horrible things that the communists did, so to see things like the old arcade machines or sodas or the best ice cream in the world (apparently!) is always really jarring in a positive way.
More relevant to this community, it always made me think about the abstract math of change making algorithms.
With the coin denominations we have in the US, the optimal (fewest number of coins) change can always be made by going for the largest coin available less than the remaining balance due. It's entirely possible to make a currency system that was not the case.
As a trivial example:
In a hypothetical system with coins denominated 20, 15, 10, 3, 2, and 1 (basically the Soviet-era model without the 5-kopeck coin) the optimal way to make 25 units (15+10) is not the "largest first" way (20+3+2)