PORTING COLORGASMS ONCHAIN

Thank you for signing up to this newsletter, i will from now on use this channel to announce all my projects ahead of social media.
You will receive detailed articles about all there is to know coming projects. Every articles will also be published on my website for easier access.
Back to web 1.0 ? Nope ! I think this is how a proper web 3.0 should arise. Because building a diffuse peer to peer network is now at anyone’s reach with the help of decentralized technologies.
I decided to radicalize this shift, and integrate on my website every part of the production to distribution chain related to my practice.

« Le centre est partout et la circonférence nulle part »
Pascal – Pensées - 1670

This is a series of 100 unique abstract compositions made using glitch techniques.
I’m currently in the process of porting this collection fully onchain with a burn/redeem mechanism which is described in detail at the end of this article.

As human beings, while we grow up, our perceptions are progressively modeled, manufactured and channeled into deterministic arrangements of concepts and labels that converge toward shared meaning.
I’ve always been fascinated by the idea of rolling back this phenomenon and produce a conscious experience of this emergence.
At the dawn of life, as a newborn, the brain is still untrained and doesn’t solidify all the stimuli it receives. it’s probably a visual chaos, in the original sense, where all the lights triggering the retina are perceived as one single living matter.
In 2008, I wrote a poem about this will to break the finitude of perceptions to reach a sense of a deconceptualized depth of the world.
Then it sat there for 15 years, till 2023 when I found this way to illustrate it.

Similar to a reverse engineering process, i rolled back the the visual perception phenomenon looking for its source.
I took videos of railroads and subway tunnels with an old phone. Dirty footages of residual infrastructure geometries, something banal no one pays attention to. Despite being of low cultural value, it is a hyper-semiologized environment : it doesn’t carry a charge of potential discovery beyond an eventual pittoresque satisfaction, which would inevitably reinforce the structural abstract meaning apposed to it.

I used a video editing software to alter the source footage in 4 cumulative steps.
1. the original chromatic palette was mapped on a new gradient. This exposed the tones inner variation and broke the first shell : the conceptualisation of color.
2. the second step applied a blur to the whole image, vanishing all perception of contours.
3. the third step dispersed the previous result in two directions by using the original tone map a reference. This destructive step created missing parts filled with black. A new vocabulary of shapes emerged as a distant derivation of the original.
4. The last step used a GAN model to generate a 4 times upscale. This increased the derivation process of the previous steps by adding plenty of new details based on the pixel size of the original, consolidating the image final texture. the GAN pass is a shift from deconstruction to reconstruction.

This conclusive step marks the inflexion point of the whole alteration process : it’s a new start, a climax, an orgasm.

At the time of the release of the collection in July 2023, I had little knowledge about how fully onchain project worked. I learned that few months later by doing the cypherdudes project, along with all the ideas about digital preservation.
As this work is very important to me, I decided to look for any means possible in order to give the highest level of technological permanence to that series.
The main issue was the file size. Each piece from the original collection is an png format image between 50 and 90Mb for 4320 by 7680 pixels.
Altering the image was not an option. Reducing resolution, dithering colors, or even finding a way to generate an approaching result, etc.. all these classic options for onchain works did not fit as the works were already published, and their level of detail were essential to the main concept.
I started looking for a way to compress the file itself like a zip file would do. This is when i found out about AVIF (AV1 Image File Format) : the AV1 intra-frame compression system predicts image blocks from neighboring pixels using spatial prediction. The amount of data that needs to be stored is not the list of all pixel values, but a list of quantized coefficients necessary to reconstruct the whole based on a compressed portion of the full image data.
The result is almost identical to the original images, for 1 % of its size.
The whole collection is now 63Mb. For onchain data as a total its a lot, but as individual pieces are now 50-90kb, its much more affordable.

In 2023, I deployed and minted this project as a series of erc721 tokens using manifold studio. This platform provides modular smart contracts that are extensions of a core erc721 contract (the token issuer).
This modularity allows for the registration of new extensions that can be used to mint new tokens under the same original contract.
I created three contracts : an extension, a renderer and a storage contract.
The extension contract contains the Burn/Redeem function where the previous token is burned and the new one is created. It also connects the new renderer contract to the original contract.
The renderer and the storage contracts work together.
The last one is used to store onchain all the images by creating one pointer for all the storage units per file (the image is divided by chunks of 24kb : max size for ethereum contracts). The renderer calls this pointer to get the image blob data and format it in a readable way for any browser.
All these contracts are available on my website here to download, reuse, and adapt to your needs.

The burn/redeem action is the last irreversible step to port the token fully onchain. This function of the new extension contract, only callable by the token owner, burns the old token and mint the new one in a single transaction.
Prior to this, the image data corresponding to a token has to be stored on chain before the burn/redeem call can be unlocked.
As the artist i’m taking charge of storing all the image data onchain. This is a slow process. In deed, in order to minimize deployment costs, i’m waiting for gas price to be very low (around 0.05).
I decided to open the burn/redeem before all the image data are stored onchain. I started with the first 13 works from the series. From now on it will be on demand. You can notify me that you own a certain token that you want to have onchain. I will answer promptly, and shortly after the image data will be stored onchain and the redeem button will be unlocked.
There are around forty available on my marketplace at 0.042 ETH, that you can of course collect before sending me the request.

As a kind collector proposed to help cover the costs for the tokens they own, i’m sharing this here too in case you’d like to help too. The cost to bring one image onchain is between 0.005 and 0.01 ETH depending on its size.

In the porting interface, the legend shows the status of each pieces from the series. Offchain - Stored Onchain - Burned / Redeemed

Once that step is done, the process is complete !
Check the onchainchecker page to inspect the properties of a token i just burned and redeemed.

Thank you very much for taking the time to read.
I hope you enjoy as much as i do to have more room to unfold every details of my works in this communication format.

You can contact me on social or by email for any questions or inquiries.

See you on the blockchain !

Felix