Sabitlenmiş Tweet
We built a novel PRNG.
V1 hit 16TB+ PractRand with zero failures.
V3 is live now — tests running.
We're challenging cryptographers to break it. 🧵
English
SOPH Entropy
69 posts
SOPH Entropy
@SOPH_Entropy
Verifiable entropy generation. Provably fair draws with cryptographic proof your users can check. Waitlist open. ✉️ [email protected]
United Kingdom Katılım Mart 2026
54 Takip Edilen0 Takipçiler
SOPH Entropy retweetledi
I like this except I think that it would be best summarised as “All tests are basically memorise the answers and don’t tilt” and that’s really hard, either component, for most. @patio11 and @tptacek had an entire startup based on “We have talent, hire it” and it didn’t work.
alz@alz_zyd_
Growing up in Asia, it took me a while to understand American "jock culture". Why would top companies hire athletes with bad academic records over top students? I still don't totally understand it but it feels a bit different from Asia or even Europe My current theory is that a lot of high general ability, quick-thinking, high discipline and work ethic ppl choose to invest a lot in being athletes until around college. Many of them do something else afterwards. But the competition is intense enough that if you hire former successful athletes you get flexible and very high general ability people You might imagine this could work for any skill. Like, if competitive league of legends were sufficiently popular, you might imagine hiring ex-pro-gamers, if you believe that the general component of ability is strong enough that if you hire someone that's the best at something a lot of others are trying to do, they will be good at pretty much anything I also feel like sports selects for a broader, and more realistic, at of skills than academic tests do. Sports tests physical ability, strategic sense, perf under pressure, game theory, and teamwork, among other things. Tests are more or less just memorize all the answers and don't tilt
English
@twekaone Nice — if you need a verifiable entropy source for the generated tokens themselves, we've built an auditable API for exactly this. Every output has a public proof trail.
English
Cooking up a tiny entropy primitive, Shannon entropy for strings 🔐
Measures Randomness Density, best used for generated tokens, secrets, and API keys
github.com/ekaone/entropy
English
@khettanyAb Nice work — impressive stats. We've been running a similar challenge, hit 16TB PractRand with zero anomalies on V1, V3 currently in progress. Completely different architecture. What's your throughput ceiling?
English
BARZAKH-521 CSPRNG passes PractRand 32TB (0 anomalies)
BigCrush 160/160 NIST SP 800-22
Quadratic torsion over M₅₂₁=2^521-1
2605-bit state | 7.5MB/s | Z3/Bitwuzla timeout
Paper👇 @khettany/barzakh-521-a-novel-csprng-via-quadratic-torsion-and-anti-linear-dispersion-over-a-mersenne-prime-cbb0ed6741c8?sk=266739d2890f270f8035aeab80dfd0e0" target="_blank" rel="nofollow noopener">medium.com/@khettany/barz…
Code👇 github.com/khettany/Barza…
#CSPRNG #Cryptography #PractRand #BigCru
English
We think it's unbreakable.
Prove us wrong.
If you like what you see — we're building verifiable entropy for draws, gaming, and security.
sophentropy.com/waitlist
English
Rules:
→ Answers SHA-256 sealed before launch
→ 6 month deadline
→ Submit via GitHub Issues
→ Methodology + code required
→ Everything public and independently verified
github.com/SOPH-Entropy/s…
English
We built a novel PRNG.
V1 hit 16TB+ PractRand with zero failures.
V3 is live now — tests running.
We're challenging cryptographers to break it. 🧵
English
We think it's unbreakable.
Prove us wrong.
If you like what you see — we're building verifiable entropy for draws, gaming, and security.
sophentropy.com/waitlist
English
Rules:
→ Answers SHA-256 sealed before launch
→ 6 month deadline
→ Submit via GitHub Issues
→ Methodology + code required
→ Everything public and independently verified
github.com/SOPH-Entropy/s…
English
@qzino_official @marbles Using live CCTV traffic as a randomness source is a genuinely interesting approach. How do you prove the outcome wasn't influenced after the fact? The entropy source is novel but the verifiability layer is where most platforms fall short.
English
New provider @marbles is now available on Qzino 🏎️
Real-time games with pure chaos — live CCTV traffic, marbles, ducks and unpredictable paths.
First stop — Rush Hour. Place a bet and leave your Qzer ID in the comments — 5 users will receive $250.
The intersection is yours.
English
@_RaffleKing All solid on-chain signals. What about off-chain raffles - brands and platforms that can't use smart contracts? Same trust problem, no equivalent standard yet.
English
Green flags when signing up to a new raffle platform 🚩
✅ Smart contracts lock the prize before entries open
✅ Chainlink VRF picks a provably fair winner
✅ Frictionless on-chain payouts
English
@iiiFish2 @HerstoryWeb3 @0xbalala @the_web3compass Nice progress on VRF subscriptions - tricky to get right. Worth noting: Chainlink VRF solves on-chain perfectly but there's no equivalent for off-chain draws yet. Interesting gap.
English
📑 Day 22 in becoming a Solidity developer
- Integrated Chainlink VRF with `RandomWordsRequest`, subscriptions, and confirmations to get secure randomness,
- Built a lottery flow with states.
#BuildinPublic #Solidity
@HerstoryWeb3
@0xbalala
@the_web3compass
English
@Dollartree_1 @lighthct Provably fair for casino games is solved. The gap is everywhere else - brands, agencies, and platforms running off-chain draws and competitions that will never touch a blockchain. That market has no verifiable randomness standard at all.
English
@lighthct the entire thing is an absolute scam. provably fair is so easy
no one needs to create a new provably fair system. its like trying to reinvent the wheel. it already works and the new wheel is just there to scam you
English
You are dropping plinko balls.. all of a sudden a few bets timeout.. those bets are refunded to your account.. but.. what if those nonces were a 1000x?
You can check the PF to see if they skipped your 1000x nonce, but.. what if the PF verifier didn't allow that?!
Welcome to @MonkeyTilt where they skip nonces and you can't even check them
English
@btcheir Great breakdown. Chainlink VRF solves on-chain raffles perfectly. The gap is off-chain - brands, agencies, and platforms running promotions that will never touch a blockchain. Same trust problem, no on-chain solution. That's where most raffles actually happen.
English
