In the current crypto-landscape it is difficult to distinguish yourself as a crypto-coin from the rest. All of them (with some exceptions) use Blockchain technology. In the beginning, for 2014, it was always a blockchain of its own but nowadays most new crypto-coins are released on a Token platform like Ethereum, NEO, Cardano, etc.
The new tokens are practically indistinguishable in technical terms because they all use the same platform. So we see that in order to seduce the buyers’ favour, there is a lot of buzzing websites, lots of so-called “partners”, long lists of team members, advisors and supervisors, a roadmap from here to there and blitz looking “white papers” in which the objective often remains a bit vague and you can regularly ask yourself whether this coin contributes to a better world or only serves to fill the pockets of the starters of the coin.
The Crypto coins that come on the market with their own blockchain can of course distinguish themselves from the rest with their own technical specifications.
So, one of the things is the number of transactions per second that their blockchain/network can process, the so-called TPS. Very proudly they mention that they can process more transactions than crypto X and Y and the biggest confectioners talk about more transactions than Paypal or Visa can handle. Especially Visa is for many an example they want to mirror. The central Visa network now handles on average about 4400 TPS. source creditcard.nl
Let us now examine the TPS in more detail.
The mother of the crypto coins is of course Bitcoin. This coin has a block time of 10 minutes and a block size of 1Mbyte. Because a transaction is 420 Bytes on average, we can calculate that Bitcoin can handle 1000000/420/(10*60)=~4 transactions per second on the network. By introducing SegWit this has been increased to an average of about 7-8 transactions per second. Here a graph of blockchain.com
One of the busiest crypto network handles 254,456 (november, 28 2018) transactions per day. That means about 2.85 TPS. That is about 30% of the capacity of the bitcoin network can handle.
Several coins have changed the specifications with respect to Bitcoin and come with higher numbers. For example, Gulden and Litecoin have a block time of 2.5 minutes and can handle therefore 4 times as many transactions as Bitcoin. And Bitcoincash has adjusted the block size to 32MB and can process up to 32 times as many transactions per second as Bitcoin.
So it is important, if you want to use TPS as a USP (unique selling point), you will be above 7. Well, we have known that lately….. The new coins rolled over each other to present ever higher numbers.
- Bitcoin, 4–10 TPS
- Gulden, 16–40TPS
- Dash, 30–56 TPS
- Digibyte, 300–560TPS
- Neo, 1000 TPS
- Eos, 5000 TPS
- Nano, 7000 TPS
- Komodo, 20.000 TPS (1.000.000 in the future)
- Ternio, 1.000.000 TPS
In addition, several existing coins have announced “improvements” to increase the number of TPS. Ethereum also wants to shard to 1,000,000 TPS with possibilities to expand to 100 million TPS. But for the time being these are theoretical values that have not yet been realized in practice. It is expected that Ethereum Sharding will only be able to roll out in 2020.
Is all this necessary now?
Bitcoin could handle 4 TPS and in 2017 it ran into the limits. But that wasn’t caused by the theoretical maximum number of TPS but mainly because several miners mined “empty” blocks so the maximum capacity couldn’t be used. However, with the introduction of SegWit an average of 7–8 TPS is now possible and that has been proven to be enough for the time being for Bitcoin. But for all the other crypto currencies it doesn’t seem very sensible to already lay down a method that can handle 1 million transactions. Except if you do that for marketing purposes of course. It sounds nice but serves no practical purpose.
Below are examples where the urge to score in this area has had negative effects.
For example, to take Digibyte, it has chosen a block time of 15 seconds and a block size of 2 MB. At this moment the DGB network has about 0.2 transactions per second while 560 TPS is theoretically possible. Each block now has an average size of 585 bytes. This means that the blockchain of DGB is growing very fast with an average of 1.4 Gbyte per year and is already >8GB in size. So now it is mainly overhead what DGB stores in its blockchain. The max. capacity is a factor 2800 larger than is needed now but the blockchain is growing very fast, which results in a slow working wallet. Starting the desktop wallet takes minutes to synchronize to get the last blocks in.
In addition, researchers have determined in 2013 that the network latency ensures that it can take 12.6 seconds before a new block propagates to 95% of the nodes. See here the paper. The block time of DGB is regularly only a few seconds, so that it can happen that one part of the network lags behind another part of the network. This leads to extra orphaned blocks.
In Ethereum, which uses the same block time, this is solved by allowing so-called “Uncle” blocks. But it all leads to useless network traffic that actually disrupts the proper functioning of the Blockchain.
Neo has largely given up his decentrality in his search for more TPS. They have devised a system where there are Consensus nodes that control and delegate the blockchain traffic. There are now 7 consensus nodes working under the responsibility of the NEO foundation, which means that this way of working no longer produces a controllable safe decentralized blockchain and thus tends towards the centrally structured Fiat money traffic with its banks from which we wanted to stay away with crypto money and blockchain.
Bitcoin also tries to do something about the slowness of the transactions. The lightning network has been developed for Bitcoin. A layer on the bitcoin protocol that does not handle transactions via blockchain but via separate channels. This allows transactions to be done “off-chain”. Only the final results (balances) of a group of transactions are recorded in the block chain. But this form is also at the expense of decentrality and the history of all transactions is no longer exactly kept. The blockchain ensures that only the final balances remain correct but all in-between balances in blue and brown disappear.
The desire to mirror the capacity of Visa is not necessary for any crypto currency in the current crypto market. One of the busiest network (Bitcoin) is a factor of 1000 less pressure than Visa, while wanting to get such large numbers of transactions can be very negative side effects for the blockchain network.
Of course it makes sense to think about solutions for busier traffic but the need to implement them is not really there yet for most of the currencies, except for Bitcoin and Ethereum.
It makes more sense to take the time to develop a thorough decentralized solution for this issue than to now burden a Blockchain network with a half faulty solution because it sells so well from a marketing point of view.