A Shining Gem amongst Crumbling Ruins

The Castle Chronicle: Volume 16

Author

Atomist
May 11, 2023 • Est. Reading Time: 23 minutes

The markets are bleeding post-CPI, with memecoins taking a hard hit. However, amidst the chaos, a single tweet from Elon featuring a meme with Milady has sent the price of NFT collections soaring.

Step into the Castle realm and unravel this extraordinary tale… 👇️ 

 👀 What’s to Come in Volume 16

🔑 Key News: Bittrex bankruptcy, LVL hack and Other Headlines, Exploits, Continued Adoption, DeFi Updates, New Projects & more from CJ

🔥 Hot Narratives: Elon tweets a Milady meme, Arbitrum fee switch proposal, and post-CPI mayhem

📈 Macro Outlook: Simplified Round-up and Implications from Hansolar

📉 HTF Price Action Scenarios bullish or chop? from Vlad

⛓ On-Chain Sleuthing: Algod trying to obfuscate his activity, Appodial switching to MUX, Smart Money accumulating crETH2, Smart Farmers leaving AVAX and more brought to you by Cl & Amir Ormu

🔬 Part 2 of Digesting Zero-Knowledge Biggest Paper from Herres

⚔️ A Call to Arms

Think you have what it takes to enter the Castle and contribute to funding and advising projects across the space?

🔑 Key News

Major Headlines

Exploits that happened this week

Adoption is not stopping!

Weekly DeFi Updates

Some interesting projects to look out for

Courtesy of CJ - check out his telegram channel for daily news, projects updates and new releases

🔥 Hot Narratives

Arbitrum Revenue under DAO Ownership

With all the commotion around the launch of the Arbitrum DAO and ARB token, many missed one of the most bullish parts of this new entity. Arbitrum has committed to sending all surplus revenue generated by transaction fees to the DAO. They are the only rollup to do so.

This revenue surplus was just distributed, where the DAO received 586 ETH from the L1 fee surplus and 1,477 ETH from the L2 fee surplus totalling 2036 ETH ($3.75m). Obviously, there is already a proposal to distribute this fee surplus to ARB token holders.

One thing to watch out closely for in the future is the impact of EIP-4844 which changes Ethereum’s fee market mechanism. This is predicted to dramatically reduce fees across the L1 and its L2s (according to some projections reducing L2 fees by 10x). The question will be whether the L2s equally reduce their fees and keep their margins the same or look to increase their profitability.

Milady Storm Continues

Milady continues to dominate the NFT market, up another 50% in the last week. As a DeFi-native, I am simply enjoying the NFT ‘bluechip’ communities cope and seeth.

One fantastic demonstration of this NFT rivalry is the MILADY/BAYC chart which has been absolutely UpOnly. Remember, you can leverage trade Milady to the moon or BAYC to zero on NFT Perp private beta.

A recent tweet by Elon Musk was only fuel on a burning fire and set Milady collections off on an absolute godsend into mainstream attention:

Post-CPI Mayhem

As always, we saw a fair amount of fuckery in the market with the CPI print yesterday. The FUD was further flamed by Arkham sending out another incorrect notification of the US Govt. selling BTC. Whipsaws in price action were also seen in traditional charts such as SPX, DXY, and Gold.

I shouldn’t need to remind you, don’t trade these events.

Memecoin Szn Over?

Hopefully, one thing we can all rejoice over soon is the death of meme szn. The glorified PEPE is now down ~63% from its highs. We can only pray for a return to fundamental value investing in this completely irrational market.

Memecoins have shown time and time again that they probably should have a place in your portfolio, but you are probably very late allocating now.

Once again this reaffirms the fact that crypto moves FAST. In recent months successful traders have moved from range trading a crab market to swing trading narratives to memecoin domination. Who would have thought narratives themselves were just a narrative for a time…

As always, to make it in this game you need high conviction but also high adaptability, with the ability to flip that conviction elsewhere at the drop of a dime.

💹 Macro Outlook

Past Week Events

  • 3/17 BTFP Starts 👉 SPX, Gold, Crypto rally

  • 3/22 08:00 FOMC: 0.25% 👉 SPX & crypto fell, but gold rose and bond yields fell

  • 3/31 12:30 PCE Inflation: A 0.3% F 0.4% P 0.5% 👉 SPX & crypto rally on softer print

  • 4/12 12:30 CPI(MoM) : A 0.1% F 0.2% P 0.4% Core CPI A 0.5% F 0.4% P 0.5%

  • 4/12 18:00 FOMC Minutes: Admission of light recession towards EOY

  • 4/13 PPI: A -0.5% F 0.1% P 0.0% 👉 Large fall leads to risk-on rally

  • 4/20 UK 👉 10.1% inflation numbers spark a sell-off

  • 4/28 Core PCE, Personal Income/Spending: A 0.3% F 0.3% P 0.3%

  • 5/3 23:00 JOLTs Job Openings: A 9.59m F 9.68m P 9.93m

  • 5/4 03:00 FOMC: A 5.25% F 5.25% P 5% 👉 SPX sell off BTC and GOLD up

  • 5/5 12:30 Nonfarm payrolls A 253k F 180k P 165k | Unemployment Rate: A 3.4% F 3.6% P 3.5% 👉 SPX up Gold and BTC down

Upcoming Events

  • 5/10 12:30 Inflation Rate MoM: F 0.4% P 0.1% YoY: F 5% P 5%

  • 5/10 12:30 PPI: F 0.3% P -0.5%

  • 5/12 14:00 Michigan Consumer Sentiment: F 63 P 63.5

  • 5/16 12:30 Retail Sales MoM: F 0.6% P -0.6%

BTC, ETH, and Macro

  • With regional bank fears subsiding SPX is not showing much direction at local highs around 4100.

  • The CPI print today could offer some volatility in the otherwise slow markets. Current forecasts are at 5% YoY. This figure has seen a steady fall from the beginning of the year and it would be disheartening if it started getting sticky around this area.

  • Truflation(real-time inflation monitoring service) points to lower inflation numbers around 4.5% for April and the PPI, a forward indicator of inflation, last month came in substantially lower which could also filter down to a lower print for this month. However, the job market has continued to show strength. Nonfarm payrolls on the 5th overshot forecasts by 40%! This could cancel out some of the disinflationary trends.

  • At the end of the day;

    • Above 5% 🐻ish for crypto

    • Below 5% 🐂ish for crypto

Courtesy of Hansolar - check out his telegram chat for all things options, crypto and DeFi

📈 HTF Price Action Scenarios

This will be a short and sweet one as price is pretty much in the same position as it was a week ago. We got a push to the upside out of this marked range and price is now correcting that push. My instant thoughts here are re-accumulation. The expectation is still for price to either continue its bullish push from here or to break down lower towards the daily demand zone.

Overall still bullish expecting more upside. Re-accumulation could be longer though and we might see some chop for a while.

Courtesy of Vlad - trend-based trader and MentFX student

⛓ On-Chain Sleuthing

Algod trying to hide his activity

The team at Castle Capital has found the addy where the famous Bahamian, Algod, has sent his esGMX bag. He seems to have tried to hide the destination by doing 2 transfers (via the GMX transfer function) between different addresses.

The first transaction. The second transaction. Leads to a fresh multisig, not even 3 months old.

Appodial selling GMX and now LP’ing on MUX

Appodial, the GMX bull & important figure in the GBC community seems to have derisked and sold 1/3 of his GMX bag.

Appodial selling GMX for USDT.

Additionally, he is now LP’ing on MUX.

Appodial is providing liquidity for MUX Protocol.

Gainzy doxxes his wallet

In a recent tweet, Gainzy showed one of his wallets containing 2 million in USDT. Here is a link to his debank page, you might find some alfa there.

DigitsDAO is a PEPE LP

DigitsDAO is now LP’ing PEPE/WETH on Uniswap v3 in an attempt to profit off the high volume this pair has been generating.

Digits’ LP position in Uniswap V3.

Smart Money Moves - Tokens

  • In the past week, Smart Money has shown significant interest in the memecoin category, with most of their activity directed towards it.

  • Specifically, many Smart Money addresses have interacted with $Lotus, a new memecoin that is designed with a flywheel effect.

  • Additionally, despite crETH2 still being depegged, Smart Money continues to accumulate it.

  • While Smart Money is currently farming on Chronos, it appears that they are not interested in buying or holding $CHR, opting instead to sell their rewards immediately causing the price to plummet.

Smart Money Moves - Farms

  • This week, Smart Farmers have turned their attention towards The Granary, which has just begun $Grain emissions. By supplying funds on The Granary, you can earn $Grain.

  • MUXLP continues to be a popular choice among Smart Farmers.

  • Some Smart Farmers have recently been farming on The Ennead (beta.ennead.farm).

  • Smart Farmers have left the AVAX chain in the past week, after squeezing all of the juice in Glacier Finance and SoliSnek farms.

  • Farming volatile pools on Velodrome remains a popular choice for Smart Farmers, while those looking for farming stablecoins turn to Wombex Finance and USD+.

  • Finally, Smart Farmers have shown a preference for the WETH/ARB pool on 3xcalibur too.

Research and analysis by Cl & Amir Ormu

🔬 Digesting the Biggest Paper on Zero-Knowledge: Part 2

In volume 14 we kicked off a research series on the biggest paper that covers Zero-Knowledge. We explored mathematical proofs and non-traditional proof systems. We highly recommend going back to volume 14 and reading the first part to properly understand chapter 2 which is covered in this volume. It is a great way to learn more specifics on zk, tech that will increasingly become more important.

Chapter 2: Introduction

In this section, the author highlights how randomness contributes to the power and efficiency of surveyed proof systems. Before diving into specifics, the importance of randomness is emphasized in enhancing algorithm efficiency. A scenario involving trusted parties, Alice and Bob, illustrates this concept as they collaborate to compute a function from their inputs.

The subsequent paragraphs introduce randomized algorithms and demonstrate how randomness improves efficiency through specific examples. Reed-Solomon Fingerprinting, Freivalds' Algorithm, an alternate perspective on both, and Univariate Lagrange Interpolation will be explored in Chapter 2.

These algorithms play a vital role in scaling blockchain systems by reducing communication requirements between nodes. With blockchain's decentralized nature and extensive data, transmitting information efficiently is crucial.

Reed-Solomon fingerprinting and Freivalds' Algorithm enable nodes to communicate using smaller data sizes while ensuring data verification, thereby enhancing scalability and overall system performance.

These algorithms are instrumental in advancing blockchain technology for widespread adoption and success.

2.1.1 Reed-Solomon Fingerprinting

Imagine you have two very large files, one is a book called "The Cat in the Hat" and the other is a book called "Green Eggs and Ham". Instead of sending the entire book to your friend, you create a fingerprint of each book by taking a few selected pages and creating a code (using a Hash) from the letters on those pages.

For example, you might use pages 10, 20, and 30 from "The Cat in the Hat" to create a code. Then, you take pages 15, 25, and 35 from "Pepe the Frog" and use them to create a different code. These codes are much shorter than the entire book but still unique to each book.

You can then send the two codes to your friend instead of the entire book. Your friend can compare the codes and if they match, then it is very likely that the entire books are the same. However, if the codes don't match, then the books are different.

Reed-Solomon fingerprinting is a more complex version of this idea that uses advanced mathematics to create even shorter and more unique fingerprints of large files to increase efficiency. It's used in many different applications, from comparing software updates to detecting errors in data transmission.

A more elaborate description which is needed to give emphasis on the cost of the protocol is the following: In the Reed-Solomon fingerprinting protocol, Alice picks a random element r from Fp (a set of integers modulo a prime number p) and uses it to compute a hash function hr that creates a fingerprint of her file. The value v is the result of applying hr to Alice's file, which is also called the fingerprint of the file. Alice then sends both v and r to Bob, who uses the same hash function hr to create a fingerprint of his own file and compares it with the fingerprint received from Alice.

2.1.2 Costs

In the Reed-Solomon fingerprinting protocol, Alice only needs to send two pieces of information to Bob: v and r. This is a big improvement compared to the deterministic protocol where Alice would have to send n times the log of the number of possible characters in the files.

Assuming that the prime number p is not too big compared to n, Alice can send v and r using only O(logn) bits. This is a big deal because it means that the amount of information that needs to be sent over the network is much smaller, which makes the process faster and more efficient. This shows that randomness can be very powerful when used in the right way.

2.1.3 Discussion

Reed-Solomon fingerprinting is a protocol used for verifying data between two parties. This protocol is based on the use of a hash function that generates a unique identifier, known as a "fingerprint," for each input data. Reed-Solomon fingerprinting uses a hash family H that has the property that for any x ≠ y, the probability of two different inputs having the same fingerprint is small. The protocol is named after the error-correcting encoding, called the Reed-Solomon encoding, used in this protocol.

A field is any set equipped with addition, subtraction, multiplication, and division operations that behave as they do over rational numbers. For any prime number p, Fp is also a field, where the field operations are addition, subtraction, multiplication, and division modulo p. For every a ∈ Fp \ {0}, there is a unique element a −1 ∈ Fp such that a · a −1 = 1. Division by an in Fp refers to multiplication by a −1. Later in the manuscript, the fact that there is a unique finite field of size p^k, denoted Fp^k, is exploited.

2.2.1 Freivalds' Algorithm

Freivalds' Algorithm is a probabilistic algorithm for verifying matrix multiplication. Given three matrices A, B, and C, the algorithm determines whether AB = C. The straightforward way to do this is to compute AB and then compare the result to C. However, this requires O(n^3) time, which can be too slow for very large matrices. Freivalds' Algorithm is a randomized algorithm that can do the verification in O(n^2) time with high probability.

The idea behind Freivalds' Algorithm is to use randomization to reduce the amount of computation needed to verify the correctness of AB. The algorithm works by picking a random vector x and checking whether Ax = B(Cx), where B(Cx) denotes the matrix-vector multiplication of B and Cx. If Ax = B(Cx), then ABx = BCx = C(Bx), which means that AB = C. If Ax ≠ B(Cx), then AB ≠ C with high probability. The probability of error can be made very small by repeating the algorithm several times.

Freivalds' Algorithm has applications in computer science and other fields where large matrices need to be multiplied and verified. It's used in machine learning, image processing, and cryptography, among other areas. Note that in the field of blockchain scaling, we use a lot of matrices that are bound to these types of algorithms to make them more efficient.

2.2.2 Cost

The cost of Freivalds' Algorithm is O(n^2), which is much faster than the O(n^3) cost of the straightforward algorithm for matrix multiplication. This is because the algorithm only needs to perform matrix-vector multiplication, which takes O(n^2) time, instead of full matrix multiplication, which takes O(n^3) time.

The algorithm is also very efficient in terms of space complexity. It only needs to store three matrices A, B, and C, and a single vector x. This means that the algorithm can be used for very large matrices that would not fit in memory using the straightforward algorithm.

2.2.3 Discussion

Freivalds' Algorithm is a simple and elegant way to verify matrix multiplication using randomization. The algorithm has the advantage of being very fast and efficient, making it useful for a wide range of applications. However, the algorithm is probabilistic, which means that it can make errors with some small probability. The probability of error can be made very small by repeating the algorithm several times, but this increases the running time of the algorithm too.

Research by Herres

Thanks for reading, please give us a follow at Castle Capital and subscribe to The Castle Chronicle for an update each week!

Virtually yours,

Atomist & The Castle

Reply

or to participate.