We’re very close on the merge release but still dealing with some last minute issues.
FWIW, the merge may sound easy, but it is an incredible amount of work – 100s’ of Monero unit tests have to pass and then GRAFT’s on top. Every change potentially has consequences in other places that have to be found and reconciled.
We’re getting there though. At this point we’re looking into next week for public Testnet.
We’re almost finished with the M13 merge – it has been a LOT of work – picking apart and putting back together a very complex and not that easy to follow code base that is Monero, following and retesting all the changes with the code we’ve added, but we’re pretty much there though with no blocker issues left at this point and only few non-blockers to get through.
Some devs are taking a break this week and part of next week for national and family holidays, so will push out the network update rollout until mid May when everyone’s back and are able to support any issues that might come up.
We hit a big first milestone with LYRA DPoS architecture this week – a very early first proof-of-concept version was shown by Slava. LYRA combines delegated proof-of-stake (DPoS), private transactions, and chain collection (no single blockchain!).
Why are we working on LYRA now?
There are several reasons for that:
1) we’re in the field that is developing extremely rapidly and favors projects that not only deliver the functionality, but also (and perhaps most importantly) offer technology leadership.
2) A DPoS architecture will be useful in different ways over time, starting tentatively as a sidechain solution for merchant tokens / loyalty programs.
Staking is done via “cold wallet” staking transactions as described in the guide and install instructions. Staking transactions expire based on your preferences and can be added up for a higher tier.
RTA sample is chosen randomly using the staked blockchain based supernode list embodying the real RTA implementation.
The next few months will be spent building up the network using stimulus transactions, flushing out any bugs and vulnerabilities and finishing the workflow. We will then roll out mainnet production-ready external RTA API support along with new wallets around May-June timeframe, to run on top of a robust and stable network.
Public testnet seeds are being upgrade later today at block 287770
The mainnet network upgrade is scheduled for March 20, around 14:00 GMT, @ block 308460
With that, we would like to invite you to take a minute to celebrate with us!
(virtually for now, although we are thinking of doing a GRAFT user conference in not so distant future)
Supernode communication and tier-based authorization selection
After extensive deliberations, community feedback, research and testing, we are settling on a new tweak called CryptoNight v8 ReverseWaltz for the time being – it is based on Waltz’s number of iterations and small change in shuffle operation. We will be monitoring a promising CN-gpu algorithm with potential switch to it in the future.
The fork is tentatively set for block height 299200 (03.07.2019, appr 11:00 CST US)
We’re made several pull requests into popular mining software projects, waiting for Xmrig in particular to approve the pull request as critical path to mining support.
Nicehash attacks have been a real nuisance and been cause for frustration for the miners community and community at large, long confirmation times at the exchanges, and other disruptions. Last week we implemented a tweak to the current GRAFT algorithm which we dubbed CryptoNight Waltz (we reduced the current number of iterations to 3/4, hence the name “waltz” as waltz is written in ¾ timing). This change was selected because it is unique, easy to implement and test, and prevents the attackers from using available Nicehash hashing power.
Some community members suggest that GRAFT should go beyond this change and implement different algorithm. Specifically, new CN-GPU is suggested. Note that introduction of completely new algorithm would induce the risk of blockchain failure due to unexpected consequences, which means more comprehensive and prolonged testing is required. Also, if we take this algorithm without any tweak, there will be more than one blockchain on the same algorithm, which contradicts the original goal.
With that said, we would like the community to express their opinions in a form of questionnaire, and in addition to simple voting for particular algorithm name explain the reason, benefits, possible downsides, and risks. Please submit your suggestions here. You can also join discussion on Telegram GRAFT Mining group. It’s important that everyone gets a chance to “chime in”!