Will Hydra drop electricity per transaction?

Dear all,

Since Hydra will vastly improve performance, I am presuming it will also drop the electricity required per transaction. Can someone check that? If any data can also be provided to support that, this would be greatly appreciated!

Thank you in advance!

Hello @nosuic

Hydra will allow a node operator to basically open a “temporary side chain” and move bunch of transactions and calculations off chain. So, electricity consumption should still be the same except now it’s off chain. So, per transaction electricity usage is probably going to be the same.

However, this means that some transactions will not have to be part of the ledger any more (since only the end state will be reported back to on-chain). So, a smaller ledger may lead to less storage and there fore less electricity to store and propagate. I have no way to calculate difference in storage size vs electrical consumption. I would assume that such power efficiency may be very, VERY small at current state.

Also, first implementations of Hydra will be very local, so any metrics and data may be unusable. As per IOHK:
" a Hydra Head is a very local construct within a small group of participants. These groups will initially be independent and thus, looking at the sum of their individual metrics as a whole is misleading." Source: Implementing Hydra Heads: the first step towards the full Hydra vision - IOHK Blog

Maybe after Cardano reaches 100 million or a billion users it may be significant enough to notice.

Note: These are all just my assumptions and I have no actual data. :smiley:


Thanks so much Neo for taking the time to reply and for explaining Hydra. I read that blogpost you are quoting and it does explain it all clearly, thanks.

May I ask; Do you know whether the Vasil fork could drop the cost / Tx given that it will increase throughput?

Thanks again :pray:

There is a proposal that may be implemented in Vasil HFC CIP-33 that modifies Plutus so it can attach reference scripts. This will reduce the size of some Plutus contracts. Smaller Tx size will make those specific transactions cheaper. You can check out CIP-33 here:
There was another proposal to make collateral easier for Plutus contracts. This would allow contracts to have exact amount of collateral. This doesn’t make it cheaper, just makes it require less up front ADA. Also, the same proposal will allow users to use other tokens besides ADA for collateral. Not sure if this upgrade is 100% going into Vasil HFC since they need to change Plutus to do it.

As for regular Tx fees, they should stay the same. Cardano fees are based on 3 things. Those are Tx size, cost per size and base fee.

Base fee (0.16ADA) is just a constant that can be changed at any time. The main reason it exists at the moment is to prevent spam attacks. So, if value of ADA goes up by a lot that number can easily be lowered.

As for cost per byte. I think it’s 44 Lovelace per byte. This is also a protocol parameter that can change easily. So, anything that lowers transaction byte size will also reduce transaction fees.

Formula looks something like this at the moment:
Base Fee(0.16) + CostPerByte(44 Lovelace)*TxSize. = Fee
The first 2 are protocol parameters and can be changed easily (with out hard fork) any time. The last one (TxSize) can be reduced by making efficiencies to the protocol. So any time you hear Tx size is going down it means the fees are also going down.

Check out Cardano fee structure doc here:

As for Hydra, anyone running a Hydra head could reduce fees to 1 Lovelace, however that would apply only for that instance of Hydra Head. This makes micro transactions possible with in created instances of Hydra Head. However, if the Head is closed or collapses there would still be regular fees applied for the final state. You could have, for example,100 users complete 10k transactions with in one head and when it collapses the fee would be only for 100 transactions instead of 10k transactions. (Note: This is just a very simplified example).

So, Hydra fee reduction will be very specific to those apps/side chains that are running it.

I’m not sure if there are any other components planned for Vasil HFC that would reduce transaction size/ fees.