Daily XRD radix thread

Don't lie, I was scrolling down the thread and before I read your comment I could smell the curry coming from it.

Early investors are up x1000

Holochain>radix

10 pajeets were paid to shill this on biz

don't fall for the scam

Bought in the public sale. Comfy hold for the next 12 months. I think people will be shocked by the bullish price movement of eXRD, simply because this kind of hard Vesting is a new experiment for a crypto token. It reduces the sell pressure dramatically.

How do I buy this in the US?

scam, don’t bail out The seed round lads they just wanna dump in your face

Damn those Avax and Polka shills are really pulling out all stops recently. Radix will blow you, and probably Eth, out of the water; get used to it.

Attached: unlockedxrd.jpg (600x471, 29.83K)

razor is outperform chainlink and radix, hear me what i tell you, or be sorry for eternally.
Hrishikesh is a man of honor.

Attached: CEO of shitstreet capital.jpg (1976x1108, 669.76K)

Less than 4x actually, after 7 years of holding. Dan's early supporters paid 3,000 BTC for their tokens - as anyone who has read the economics whitepaper knows. Your IP address will now be traced for the libel lawsuit. RIP

I have only had a quick look at Radix but for reference these were my initial thoughts:

They have been building a product for 7+ years and were meant to release in 2019 but had to pull it because was insecure and changed the consensus. Now they are using Hotstuff for each shard (which performance degrades past 100 validators). It doesn't support total ordering (which is normally required for smart contracts). They claim to have unlimited scalability by having multiple shards (yet every validator must also validate a global shard and with hotstuff performance degrades significantly past 100 validators, so good luck scaling that to millions of validators lol)

The tps marketing figures are based on completely independent shards with no cross over between them. The application layer becomes the bottleneck then. Likewise writes may be fast to each of the different shards but reads become a problem then trying to piece together information from lots of individual shards. They say a query layer will be used but little info on how it will be done and supposed to be launching next year.

Sybil protection and weak backing were few of the issues which mean they had to cancel their 2019 release. The whitepaper doesn't go into too much detail
"Rather than having equal voting power across all nodes, each node will have voting power relative to the amount of stake they have registered in order to compete for selection as a validator"

Holochain is the worst scam in the history of crypto scams, fully abandoned by devs. Are you joking Pajeet?

So voting power is based on weight within shards (no mention of limits), so what happens if one shard has a validator with far higher amounts staked than others assigned to the shard? they can just have majority voting power by themself (or small group working together rather than needing usual 2/3 of votes?

Leaders appear to be key to enable the braiding between shards - but also vulnerable to DOS attacks as known in advance

They are launching an ERC20 token first (when it will have zero use case until mainnet launches (supposedly mid 2021). For every $0.02 the token price increases they unlock 210 million tokens until all the minted 4.41 billion tokens are released (and market cap of $1.8 Billion). The max supply is 24 Billion.

Of the 4.41 billion tokens that are minted at the start, few community members provided them with 3000 BTC in 2013 to fund the project and gets 3 billion RADIX tokens (around 70% of the minted supply and 12.5% of the maximum supply) which gets released every $0.02 the price increases

IMO they will just end up launching a single sharded blockchain using hotstuff of 100 validators (nothing special) and they still have lots to work out for multiple shards (and that’s with 7+ years already working on it)

It's hilarious that this pseudo-technical FUD keeps being copied and pasted (I know for a fact it originated in the Avalanche Telegram). Since Dan has already answered it, let me copy and paste his response:

"It doesn't support total ordering (which is normally required for smart contracts)."

Cerberus is designed to NOT have that painful limitation of enforcing total global ordering. Now, total order of applicable *state* is required, but that's very different from needing total order of *all* state, applicable or not. For example if a contract has a particle P representing the latest state in shard X, and two actions happen originating from shard Y and Z trying to consume the state P in X, the validators in shard X will define the order that Y and Z are applied and one will fail. The state is now P' and Z' and Y' actions are applied and one of them will fail. There is verifiable applicable state order P -> Y -> P' -> Z' .... smart contract safu and global state order is, well, whatever it is after the action happened!

"They claim to have unlimited scalability by having multiple shards (yet every validator must also validate a global shard and with hotstuff performance degrades significantly past 100 validators, so good luck scaling that to millions of validators lol)"

No not every validator must validate the global shard, a subset of all validators will be selected to validate the global shard, and all other validators will verify the correctness of that output. That subset can be larger than 100, and have some conditions (e.g the top 300 validators with the most stake which cover all shard space) and it churns every epoch (a period of 24 hours for arguments sake). Unless there are 100000s of staking, destaking events every day, it'll handle the load just fine.

"The tps marketing figures are based on completely independent shards with no cross over between them. The application layer becomes the bottleneck then. Likewise writes may be fast to each of the different shards but reads become a problem then trying to piece together information from lots of individual shards."

Wrong, the tps figures are based on every event being cross shard, as it has to be due to shard placement of particles that represent the UTXO being hash random. Validators in shard X don't need to know anything about the information being processed by shard Y, they simply agree or disagree that the action can be applied to shard X. Likewise the validators in shard Y agree that the action can be applied to shard Y. X doesn't need Y info, Y doesn't need X info, semi-stateless. X and Y send the results of the vote to each other and 2/3 of validators in X and 2/3 of validators in Y are required for a commit, otherwise it fails.

"Sybil protection and weak backing were few of the issues which mean they had to cancel their 2019 release. The whitepaper doesn't go into too much detail."

The whitepaper focusses on consensus only. Proof of stake is a well known sybil method, so it shouldn't be too taxing to join the dots on the basic functionality.

"So voting power is based on weight within shards (no mention of limits), so what happens if one shard has a validator with far higher amounts staked than others assigned to the shard? they can just have majority voting power by themself (or small group working together rather than needing usual 2/3 of votes?"

Correct, *if* you adopt a crude and non-intelligent means of selecting validator sets for shards. The staking information in the global shard allows the selection process to be smarter, as all nodes have that information and can verify the correctness of it. Therefore validators with similar amounts of stake can be assigned to the same validator set as it's not how much stake that is in a particular validator set that matters, but that a minority in that set doesn't have a majority of voting power.

"Leaders appear to be key to enable the braiding between shards - but also vulnerable to DOS attacks as known in advance"

Not really a new problem, nor specific to us. Many protocols have to deal with DDOS where the validator set is a known entity and there are a bunch of conventional solutions.
*********
FUD 0/10 must try harder. I know you're having a bad time with Avax price in the gutter, but that's what you get when you buy shitcoins

This shit copypasta has been debunked so many times

what the fuck are you nerds bickering about?
In plain English please, how do I p-r-o-f-i-t from this? Fucking nerds man.

Easy, buy Radix on Uniswap on 17 November and hold for a year.

By the way are you a Ferengi?

Kek you mongoloid.
The cerberus consensus is only from this year I believe. They used to have a consensus called tempo but had to cancel it as it wasnt secure due to weak backing and sybil protection problems, so they based it on hot stuff instead.

Hot stuff is limited to around 100 validators before performance degrades and with Cerberus as it braids multiple instances of hot stuff shards together for certain transactions them this increases the message complexity significantly

"Notably, this form of parallelization introduces a tunable tradeoff between security and scalability: reducing the number of shards each node covers increases parallelism and scale, but decreases the validator set size of a given shard."
"With emergent Cerberus instances, cross-shard consensus requires a separate QC per shard. Thus the
authenticator complexity with correct leaders is O(s * n / T) where s is the number of shards in a given command
(and s must be 1

MONEY GRAB

FUCK OFF

Despite sybil protection being one of the reasons they had to cancel the launch in 2019 and change consensus, theres not a lot of info in the whitepaper

"Rather than having equal voting power across all nodes, each node will have voting power relative to the amount of stake they have registered in order to compete for selection as a validator"

If voting power is based on weight within shards (with no limits mentioned), what happens if one shard has a validator with far higher amounts staked than others assigned to the shard? they can just have majority voting power by themself (or small group working together rather than needing usual 2/3 of votes?

"An in-depth description of the application of PoS to Cerberus within the Radix public network is left for future work."

"With our proposed leader election mechanism, future leaders are known well in advance and thus particularly
vulnerable to denial-of-service (DOS) attacks...A deeper analysis of possible applications of these strategies within Cerberus for Radix public or private network deployments is left for future work."
I have only had a quick look at Radix but for reference these were my initial thoughts:

It doesn't support total ordering (which is normally required for smart contracts).

The tps marketing figures are based on completely independent shards with no cross over between them. The application layer becomes the bottleneck then. Likewise writes may be fast to each of the different shards but reads become a problem then trying to piece together information from lots of individual shards.

"This mapping provides good consensus performance and security on transaction writes. Transaction read queries, however, such as requesting a complete transaction history for a given account, are more difficult since every transaction is addressed by content rather than by a single identifier."