Should you own any AVAX?
Avalanche has been rising up the crypto charts lately and I thought I’d take a deep dive on how the protocol works. Specifically I wanted to know how it compares to Cardano and whether it meant Cardano was already outdated as some were already saying.
Disclaimer
Now as a disclosure, I currently don’t own any AVAX, the currency of the Avalanche network, and I do own a lot of Ada. But I did this research to help me decide if I should be moving some of my ADA into AVAX so consider it a fair critique of both blockchains.
Also, while I did my best to read through all the research material I could get my hands on in the past few days, I’m by no means an expert in the field, so I may have interpreted some of the academic writings incorrectly. If you feel I’ve done that then please leave a comment on my YouTube channel to let me know.
Finally I will do my best to explain key concepts of how Avalanche works, but there are two excellent videos I strongly recommend watching beforehand.
Avalanche
Fast finality
The biggest thing you will hear about Avalanche when you look into it is the super fast settlement times. For those not familiar with how blockchains work, for a transaction to be considered final, a high majority of nodes on that network need to reach an agreement or “consensus” on what that transaction was. It normally takes a number of iterations before consensus is reached and the transaction is considered final, aka finality.
Ethereum and Cardano take several minutes to reach finality. Bitcoin can take even longer. In Avalanche it is 1 second which can be considered near instant.
How does Avalanche achieve this? Well it uses something called Byzantine Fault Tolerance protocol (BFTP) which allows it to employ a clever sampling method. Each node in a network will do a random survey of around 10 other nodes in the network as to whether a transaction is valid or not. Based on what it finds out it then adjusts its own position. This then repeats very quickly until the entire network has converged on the same view.
Bitcoin on the other hand uses Nakamoto consensus, which basically boils down to having a leader, in this case decided through POW that then broadcasts the block once it’s solved the cryptographic puzzle. I’ve been doing some digging around and from what I can tell the current version of Cardano’s consensus protocol (Ouroboros Genesis) does not use a BFT protocol as it requires knowing how many nodes there are running in the network upfront. This would then result in lock up periods such as in Avalanche where funds are locked up for a minimum of 2 weeks.
Avalanche has two version of this protocol Avalanche and Snowman, but one is just a slight variation of the other. The BFT based consensus protocol is definitely impressive, quite intuitive, and super fast, so I have to give kudos to Avalanche here. We’ll discuss Cardano’s finality later on.
Mixed Multi chain architecture
The second bit of innovation from Avalanche is that they have three different blockchains that work together. I imagine these would be called sidechains in other blockchains except here they can all be considered the main components of Layer 1. Each blockchain can be configured in its own way, For example the X-Chain which is purely for the purposes of managing and exchanging tokens is built on a Direct Acrylic Graph (or DAG) architecture which is renowned for its speed.
The P-Chain is for managing other blockchains, aka subnets. This is a similar concept to sidechains in Ethereum or Cardano, and parachains in Polkadot, although there doesn’t appear to be a fixed capacity that needs to be auctioned off as there is with Polkadot. Each subnet can be private or public. What this means is you can have industry specific blockchains set up which are private and for use only between participants in that industry. This is the second innovation that I’m quite impressed with. It’s something that would have a great appeal to businesses. Now Charles has talked about having other sidechains on Cardano that could be similar, but with Avalanche it seems like the idea is a little more fleshed out.
And finally the C-Chain is the computation layer of Avalanche. It is essentially an Ethereum Virtual machine that utilizes Avalanche for its security layer instead of Ethereum
Tokenomics
Ava Labs that is developing Avalanche have been raising funds through private sales. I therefore decided to take a look at the token distribution for AVAX:
From this there was nothing that really stuck out to me -it certainly looks a lot better than Solana’s token distribution and I don’t have any issues with VC funding in itself., though Cardano’s Catalyst funding mechanism is more innovative.
Cardano
So now let’s turn our attention to how Cardano compares with Avalanche.
UTXO
Avalanche may have a different consensus mechanism but it still utilizes a UTXO based model making it similar to Cardano in this respect. I find it reassuring that both projects consider the Bitcoin accounting model the best one to build on as opposed to Ethereum’s. In fact Avalanche should consider joining the UTXO alliance too.
Fast finality
Speaking of consensus, the Hydra paper is very clear about fast finality being one of the key features of the hydra layer 2 solution.
” One aspect where Hydra really shines is fast settlement: as soon as all parties have signed a transaction, and the sending node has aggregated a valid multisignature, this multisignature provides a guarantee that the transaction can be included into the ledger of the layer-one system.”
I’ve been trying to understand this a bit better, and one user on Reddit told me:
“Hydra is state channels, similar to Bitcoin Lightning Network, not a new consensus mechanism. When you do a state channel transaction with someone else it is not finalized on-chain, instead it gives you a promise that when you’ll eventually close the channel and settle the funds you’ll receive the correct amount into your wallet, that is what Hydra’s fast finality means, it is a fast promise of finality but not actual finality like on-chain transactions.” (source)
So in short, if I’ve understood this correctly, the user experience of anything that uses Hydra will be that of fast finality, but in the background it will still be relying on Cardano’s layer 1 solution to do the settlement which takes about 5 minutes. But does mean that seriously fast transactions are coming to Cardano. Charles btw has said that Hydra will be entering 3-5 week development cycles, meaning its full steam ahead with development!
Fast finality through Layer 1 improvements?
Charles’ in his recent AMA did however talk about how input endorsers, as mentioned in the original Ouroboros research paper, can speed things up significantly and that they’d be releasing a blog post on it in due course. I went back and read through the original 2016 Ouroboros paper, and can explain to you what an input endorser is – it basically verifies what the slot leader is doing, the slot leader being the node that is generating the current block. So it was discussed from the point of view of preventing malicious actors. There can be multiple input endorsers so I can see how it might form something of a DAG but it wasn’t clear to me what the big innovation for faster finality here would be – perhaps they can apply the BFT protocol to input endorsers so that things are validated and propagated much faster in the blockchain, but I’m really talking beyond the limits of my knowledge here and I look forward to Cardano’s blog post.
Sidechains
Charles mentioned that the infrastructure for side chains will be coming through hard forks next year. I don’t have much more info on this however so please comment below if you know more about what this means and how it compares to subnets in Avalanche.
Application layer
I liked the fact that Avalanche’s EVM solution was already developed out – we keep hearing about an EVM for Cardano being developed by IOHK or run-time verification, but I haven’t seen anything materialize – apologies if I’ve missed the announcements. Thankfully however we do now have DCSpark and their Milkomeda project. It’s already in the testing stages and early signs are that its going well. So when it comes to compatibility with Ethereum DApps I’d say Avalanche and Cardano are more or less on par.
Which got me thinking – is Ethereum compatibility the way of the future or will it be seen as more backwards compatibility with retro dapps ? Who knows – but if EVM is to be the way all dapps are written then Avalanche must bank on Ethereum’s planned upgrades being delayed or abandoned, as else why would anyone want to use Avalanche if Eth 2.0 finally launches? To be fair the answer might be because of Avalanches other innovations, but Ethereum has an incredible brand power behind it, it’d be like trying to mimic the iPhone and go up against Apple.
Of course in that sort of scenario you can say the same about Cardano, why would you want to use EVM on Cardano when you could use the real thing? But Cardano at least offers an alternative – its Plutus application framework. Which led me to think – people may complain about Plutus being harder to learn than Ethereum’s solidity. But given a chain that has EVM and Plutus and one that only offers EVM, I suddenly want the one that has both. Why? Because Plutus offers a refreshing alternative to EVM. Plutus is designed with mission critical applications in mind, so it’s a lot more rigid in what it allows. There is a chance therefore that Plutus would be what big business would want to build their multi-million dollar applications on, given the history of bug and hacks on Ethereum dapps.
Of course I might be wrong and the EVM may continue to dominate – but I still prefer a chain that has both options.
Conclusion
It was great learning about Avalanche, and I’ve been impressed with its innovations. If I had new sources of funding I’d buy some AVAX. However am I going to sell my Ada for AVAX? Definitely not. You only have to listen to Charles’ most recent AMA to see that IOHK know what they are doing with blockchain technology. There are certainly parts where I could do with more information (specifically what is the amazing Layer 1 improvements that has been alluded to) but that will be presented in due course.
Cardano has got immense momentum – at least a hundred DApps will be launching in the next 3-6 months, DCSpark are launching their EVM side chain for it soon, it has a programming model, Plutus, that is designed with mission critical systems in mind, it is being deployed in real life enterprise settings with millions of non-crypto users getting onboarded, it’s putting in place a certification system for DApps, it has a burgeoning governance system, a multi billion dollar decentralized investment vehicle controlled by Ada holders in Project Catalyst and an identity solution in place that positions it well for on-coming crypto regulation.
In short tech is just one piece of the picture. Many “Ethereum killers” touting amazing tech have come and gone over the years but Cardano has steadfastly remained. I will have to do more research on the non-technical aspects of Avalanche so don’t take this to mean they are sitting around doing nothing – for one thing they have a $200m fund for eco-system development. And I definitely want to have some AVAX in my portfolio. But the breadth of IOHK’s roadmap for Cardano is mind blowing, and I don’t want to be sitting on the sideline watching it unfold in 2022.
If you liked this blog post, please hope over to the YouTube version of it and give me a thumbs up!
ReddSparks Crypto Blog 