Marcelo Salhab Brogliato (@msbrogli) header
Marcelo Salhab Brogliato (@msbrogli) profile picture
Rio de Janeiro, Brazil
Joined March 2009
Photo by Marcelo Salhab Brogliato (@msbrogli)Photo by Marcelo Salhab Brogliato (@msbrogli)Photo by Marcelo Salhab Brogliato (@msbrogli)Photo by Marcelo Salhab Brogliato (@msbrogli)Photo by Marcelo Salhab Brogliato (@msbrogli)Photo by Marcelo Salhab Brogliato (@msbrogli)Photo by Marcelo Salhab Brogliato (@msbrogli)Photo by Marcelo Salhab Brogliato (@msbrogli)Photo by Marcelo Salhab Brogliato (@msbrogli)Photo by Marcelo Salhab Brogliato (@msbrogli)
Our new consensus finalized 61 million transactions in 10 minutes. Every validator averaged over 101,000 a second, and every transaction was confidential. 25 validators, 9 regions, 4 continents. We ask every chain five questions about TPS. Our answers:
4
14
3
29
1,297
Next: longer runs, and more per second. Past 100k, bandwidth is the next wall. At 100k, each validator sent about 1.1 Gbit/s to the other 24, and that grows with the rate.
1
2
1
5
74
When we asked Sui and Tempo these five questions, we promised to answer them for our own. Here they are. Sui: nitter.cf/msbrogli/status/210827… Tempo: nitter.cf/msbrogli/status/211111… Ask us more in the replies.
In 2018 the BMJ published a randomized trial of parachutes. Jumping from a plane with one was no safer than jumping with an empty backpack. The trial was careful and correct. The planes were parked. Many blockchain speed records are set the same way. Here's how to tell. 🧵 Average altitude 0.6 m. Average speed 0 km/h. Nobody in either group died or was seriously hurt. The trial ran in the BMJ's Christmas issue, and its lesson is serious. The point: a test can be done right and still say nothing about the hard part, if it runs where the hard part can't happen. Tempo's published benchmark is one example. Its methodology page: "two local validators in consensus on a dedicated bare-metal server."
2
1
5
88
Marcelo Salhab Brogliato retweeted
Hathor Evolves, HTR Expands. In 30 mins! 👇 🚨
7
17
1,297
What could a large quantum computer do to a blockchain? Take hidden amounts: the commitments still hide them, but whoever finds h (H = h·G) can mint coins out of thin air. Be the quantum computer on a laptop sized curve. Reply with your solution, or how you would solve it. 🧮
1
6
13
573
Pedersen Commitments Challenge 2 p = 2^61 - 1 = 2305843009213693951 curve: y^2 = x^3 + 7 mod p (43·q points) q = 53624256071278747 (G and H have prime order q) G = (1217468943615328232, 2065245357365722203) H = (121249104770771710, 552385845561251560) r = 1234567890123456 C = 1·H + r·G = (600904362434283989, 1383882975168263828) Attack: turn Alice's coin C, worth 1, into two coins worth more than 1 in total. Find v₁, r₁, v₂, r₂ with v₁·H + r₁·G + v₂·H + r₂·G = C and 1 ≤ v₁, v₂ < 1000 (r₁, r₂ any numbers mod q). Reply with them and how you solved it. Don't forget to say how long your code took to solve the discrete log. The first valid answer gets credited, AI answers included. H was hashed onto the curve, so nobody knows h (recipe on request). Challenge 1 and its solution: nitter.cf/msbrogli/status/211474…
Can anyone tell two 40 USDT payments are equal? Not by comparing their commitments. They match with probability 1/q, about 2⁻²⁵⁶, whether the amounts are equal or not. No complete proof came in, so it's on the card. Can anything else give them away? Challenge 2 coming soon.
1
1
73
Can anyone tell two 40 USDT payments are equal? Not by comparing their commitments. They match with probability 1/q, about 2⁻²⁵⁶, whether the amounts are equal or not. No complete proof came in, so it's on the card. Can anything else give them away? Challenge 2 coming soon.
Two payments of 40 USDT on a confidential chain. Can anyone tell they are equal? Pedersen Commitments Challenge 1. Proofs in the replies; the first clean one gets credited in the solution.
5
14
547
Two payments of 40 USDT on a confidential chain. Can anyone tell they are equal? Pedersen Commitments Challenge 1. Proofs in the replies; the first clean one gets credited in the solution.
2
7
2
22
1,900
New to the notation? A quick primer: • Points on an elliptic curve can be added. r·G means G added to itself r times. • G generates the group: adding G to itself again and again reaches every point, q of them, and q is prime. Then it cycles: q·G is the zero point, so a·G = b·G exactly when a ≡ b (mod q). • H₁, H₂ are fixed public points, one per token. C = v·H + r·G hides the amount v behind the random r. • Going back from r·G to r is the discrete log problem. Nobody knows how to do it fast on a 256 bit curve. This is the math under hidden amounts in confidential payments. Levels 2 and 3 here: nitter.cf/msbrogli/status/211339…
On most public chains, every payment shows who paid whom, how much, in which token, and where the coins came from. Analysts follow that trail to real names. Privacy isn't on or off. It comes in levels. Six levels, what each one hides, and what still shows. 🧵
6
12
139
On most public chains, every payment shows who paid whom, how much, in which token, and where the coins came from. Analysts follow that trail to real names. Privacy isn't on or off. It comes in levels. Six levels, what each one hides, and what still shows. 🧵
2
12
2
32
1,774
Level 6: hide the network. Your wallet sends each payment to a node, and the first nodes to hear it can guess your IP address. Tor or I2P hides that address. Dandelion++, used by Monero since 2020, passes the payment along a random path before it spreads. This one works at any level: Bitcoin Core supports Tor and I2P.
1
4
12
73
At level 6, what's left is that a payment happened, when, how many inputs and outputs it has, and the fee. Most levels add proofs, and every proof costs every node work to check. Which level would you pay that cost for, and where would you stop? The names, to look up: Pedersen commitments with range proofs (Bulletproofs), asset surjection proofs, stealth addresses, ring signatures, zk-SNARKs, Dandelion++. Used in Liquid, Monero and Zcash. How the trail is followed: the café thread (nitter.cf/msbrogli/status/211226…). Try it yourself on Hathor's privacy testnet: nitter.cf/HathorNetwork/status/2…
PRIVACY TESTNET IS LIVE! ⚡️ You can now send, receive, hide and reveal your transactions on Hathor Mobile Wallet! Watch the tutorial which summarises: 1. Update mobile wallet to v0.40.0 2. Switch to shielded testnet network 3. Get tokens: faucet.testnet.hathor.networ… hathor:native
4
15
80