Hey profuno. It's legit :). See here for self-custody considerations: [https://medium.com/loopring-protocol/we-take-the-ultimate-non-custodial-test-b5528fafbec2](https://medium.com/loopring-protocol/we-take-the-ultimate-non-custodial-test-b5528fafbec2)
Or here for full technical spec: [https://github.com/Loopring/protocols/blob/master/packages/loopring\_v3/DESIGN.md](https://github.com/Loopring/protocols/blob/master/packages/loopring_v3/DESIGN.md)
Thanks for the links!
Some follow up questions that you may be interested in answering.
What do the people, who don't want Ethereum to succeed, say about this and why are they wrong?
Do you think there is a need (or is it advantageous) to have this running in the wild for a few months/years before we claim that this is one of the true scalability solutions we have all been waiting for?
Hey. Good questions. I will answer and provide more links :).
1) They can say, Loopring can censor transactions. That is true. Loopring.io's single operator can indeed produce zkRollup blocks as it wishes. Ignoring an account's trades or transfer requests, for wtvr reason. But of course that is just bad business. And really, the main value Loopring provides is that a user's funds are always safe and in their self-custody, Operator cannot freeze those, or steal those. So if you're censored, you exit the system. All this with high speeds, low fees.
Importantly, Loopring Protocol actually DOES support a multi-operator model which solves this (if one operator ignores, the next will accept). One reason Loopring.io isn't running that way is because our users don't exactly demand that right now. Performance and asset security are priorities. I speak about this exact topic on today's Into The Ether podcast :) https://podcast.ethhub.io/loopring-scaling-trading-and-payments-on-ethereum-with-zkrollups.
2) Of course that could generally be good practice in everything on Ethereum. But as far as these things go, Loopring takes the most secure design decisions at every fork. So if you wanted to get onto a layer 2, a zkRollup presents the safest way to get up there. In fact, you can even call it layer 1.5 :). I also talk about that in the podcast. This post is also a good quick one on those different decisions: https://medium.com/loopring-protocol/we-take-the-ultimate-non-custodial-test-b5528fafbec2
I think this ITE podcast is the best explanation I've heard on ZK and Optimistic rollups, and the difference between...
No they are not private. The zk crypto here is used to allow for efficiently verified proofs of some computation happening but making it private is much more involved.
Edit: Looking into it it seems there are some rollup variants that do claim transaction indistinguishability, but I don't believe you don't get it by default.
Hey UE. LRC stakeable by normal users to receive a portion of protocol fees that flow through the network ([https://staking.loopring.org/](https://staking.loopring.org/)); and also must be staked by DEX or payment owner/operators as a bond for service level guarantees (kind of like economic security before the proofs are published for Ethereum level security).
You can read more here about it: [https://medium.com/loopring-protocol/loopring-3-0-lrc-utility-model-d7da9ac79d3d](https://medium.com/loopring-protocol/loopring-3-0-lrc-utility-model-d7da9ac79d3d). But as you see, that article is nearly one year old, and a lot of advancement has happened on our part - with payments, but even more with upcoming protocol versions. So the specific mechanics are very likely to adapt, and we have in fact began discussions around optimizing it. The core will remain same.
>Loopring does only work with normal transactions and does not help scale other smart contract functions afaik.
Thats a good point, I wonder if zkRollup can support more complex transactions in the future.
I've been wondering this too. I guess the answer is a) democratizing things so you don't need an ASIC to participate in the blockchain and b) saving some of that PoW energy because you still need the main chain to run for these L2 solutions to work
Yeah right, whats stopping them then? BTC has given up on innovation a long time ago. BTC will not revolutionize anything. My bet is it will end up as sth like stamps where a few nerds collect and trade.
If you read through the bitcoin-dev and lightning-dev mailing list you'd see that there's a constant new stream of ideas being tossed around. New developments on chain happen rarely as all possible tradeoffs need to be discussed and tools to take advantage of new new features need to be developed before the features are deployed on the live network in order to properly understand the properties of new features.
It's a completely different approach to development than what the Ethereum community takes, and due to the extreme caution being taken development appears to be from the outside much slower. But in reality the result ends up being incredibly versatile and performant without some of the unforeseen externalities you sometimes end up with in Ethereum. One example being the rise of centralized services like Infura being used as an interface for the network instead of more people wanting to and being able to run full nodes.
I know that there is a lot of ideas being tossed around. And objectively nothing at all has happened. The few failed attempts to make something happen suggest that it will stay that way. The stuff thats currently in production on top of Bitcoin is just as centralized as Infura.
My observation is that the great minds that want to make things happen are on Ethereum. Why would you try to write a game on a pocket calculator if you have a normal computer at hand. The "caution" is just an excuse for incompetence or laziness. You cant disrupt the banking system if you dont move.
That's correct. Currently it can be used as a decentralized exchange and for direct payments of ETH and ERC20 tokens.
>As announced last week, we extended the Loopring Protocol and Relayer to support not only zkRollup DEXs, but also payments!
Isn't the philosophy behind an OR that the assumption is every transaction is valid, until proven otherwise? With a BFT PoS setup, once a transaction is confirmed, then it can't be reversed.
Maybe there's something about OR's that I'm missing
I love Ethereum and everything about it, but even I know that there are things Ethereum can't do. For example, it can't allow someone to deploy their own chain with their own consensus rules. These blockchain specific apps can perhaps have requirements where total decentralization isn't ideal or feasible, and the consensus rules need to reflect that. So by running your own consensus engine within your own chain, you can also gain the benefits of connecting to Ethereum through Cosmos. I think this is a niche build case that Ethereum or Polkadot can't do with their shared security model. But that's ok, because all of these chains do provide some technical benefit to the entire space, so they really ultimately work together. Where the ecosystem value will eventually accrue is obviously still an open question since I see this playing out over 5-10 years. All I know is that Ethereum is king, and Cosmos is kind of interesting, so we'll see what happens.
No, it's not. Someone still needs to pay for the computation (ZK is intensive) and on-chain balance-updates. It's less than a cent per transfer, so for now Loopring is paying for it, but that won't last forever.
I think you have to do 1 on-chain transaction to join the Loopring Pay system, then you can transfer to anyone else on that has joined the Loopring Pay system, instantly and for free.
Like depositing and withdrawing from a DEX, but you remain in full custody.
**UPDATE**: I just create a Loopring account and it costed me $2.39 just to open the account. (you just have to sign an on-chain transaction to open the account)
There is a 55.0 minute delay fetching comments.
**Defaulted to one day.**
I will be messaging you on [**2020-06-07 11:20:11 UTC**](http://www.wolframalpha.com/input/?i=2020-06-07%2011:20:11%20UTC%20To%20Local%20Time) to remind you of [**this link**](https://np.reddit.com/r/ethfinance/comments/gxms0a/loopring_pay_is_live_zkrollup_transfers_on/ft3rrdu/?context=3)
[**CLICK THIS LINK**](https://np.reddit.com/message/compose/?to=RemindMeBot&subject=Reminder&message=%5Bhttps%3A%2F%2Fwww.reddit.com%2Fr%2Fethfinance%2Fcomments%2Fgxms0a%2Floopring_pay_is_live_zkrollup_transfers_on%2Fft3rrdu%2F%5D%0A%0ARemindMe%21%202020-06-07%2011%3A20%3A11%20UTC) to send a PM to also be reminded and to reduce spam.
^(Parent commenter can ) [^(delete this message to hide from others.)](https://np.reddit.com/message/compose/?to=RemindMeBot&subject=Delete%20Comment&message=Delete%21%20gxms0a)
hey. that's right. there is a tiny cost to Loopring Relayer (backend+proving system) to process a batch/publish the proof, but it is so negligible per tx, we are fine with making it free at launch (and potentially for a long, long time). if some spammer tries to fill our zkrollup blocks by sending payments back and forth, we will reconsider. for now, we hope the word 'free' entices folks, as the zkRollup really becomes more useful with more people on it. (and to some extent, actually cheaper for us, since we don't want to process blocks with \~10 txs in it... we want 1000 txs to amortize the cost :).
Hey UE. Sorry for delay here, been a busy few days!
Costs can be seen in depth here: [https://github.com/Loopring/protocols/blob/master/packages/loopring\_v3/DESIGN.md#results](https://github.com/Loopring/protocols/blob/master/packages/loopring_v3/DESIGN.md#results)
*Note the above is for TRADEs. For Transfers (Loopring Pay) my rule of thumb is everything is half as expensive (or double as efficient).*
So for trades: cost to publish is \~220k gas. That's a fixed overhead cost that is amortized over all trades in the block. (up to 4096 trades, so \~8000 transfers, although for a block that big it may be 400k gas). Then there is also an incremental cost per tx of 365 gas, because we put the data on-chain (on-chain data availability needed to be a true zkRollup).
If those numbers don't mean much, this article puts it in dollar terms: [https://medium.com/loopring-protocol/zksnark-prover-optimizations-3e9a3e5578c0](https://medium.com/loopring-protocol/zksnark-prover-optimizations-3e9a3e5578c0)
TLDR: $0.000124 per trade, or 1 million trades for $124. For a transfer, figure $0.00006, or 1 million transfers for $60.
That includes the gas costs described above, plus the off-chain ZK proving costs. As you see, very, very, very cheap :). (note the articles don't consider gas at 30 gwei, but you can scale the cost and it is still super cheap.)