Day: September 12, 2019

Have you joined the RyanSchultz.com Discord yet? More details here. 

---

It’s only natural to want to look at the similarities and differences between the first two blockchain-based virtual worlds to launch, Cryptovoxels (CV) and Decentraland (DCL). While Decentraland is still in closed beta testing, I was among the first group of people who was invited to visit and explore this new platform. Therefore, I have decided that now would probably be a good time to compare and contrast the two virtual worlds, in an effort to provide the best information to current and potential investors in both platforms.

Most of the information I am reporting here comes from two sources:

• the Crypto Cities report on HackMD, last updated Sept. 1st, 2019;
• the Decentralized Virtual Worlds report by CL and Jin, dated July 25th, 2019.

Where information differs between these two reports, I have chosen the more recently updated version, the Crypto Cities report.

---

Decentraland and Cryptovoxels are two virtual worlds that currently exist on the Ethereum blockchain. Both of these virtual worlds are divided into square pieces often referred to as parcels, aligned on a grid to form a city. In both virtual worlds, land parcels are a non-fungible asset maintained in Ethereum ERC-721 smart contracts.

Project Background

Unlike many other social VR projects such as Sansar and High Fidelity, where the company has built the virtual world over time in anticipation of earning future income from users, Decentraland started with a well-timed, highly successful Initial Coin Offering (ICO) of their cryptocurrency token, MANA, in August 2017, raising US$24 million in less than a minute! This was followed by two successive auctions of virtual land parcels (called LAND), which were also very successful. Today, MANA has a market capitalization of approximately US$50 million. Decentraland is based in Argentina, and currently employs an estimated 45 people full time.

Contrast this with Cryptovoxels, which started in 2018 as a part-time project by a single New Zealand software developer, Ben Nolan. Cryptovoxels has been funded to a total of approximately US$140,000 worth of Ethereum (ETH) over the course of one year of virtual land parcel sales. This profit has recently enabled Ben to be able to work on the project full-time.

Project Size and Maps

In terms of overall size of the projects, Decentraland is approximately 23 times bigger than Cryptovoxels:

Decentraland itself is about half the size of Manhattan in New York City:

Jin reports on the differences between maps:

Decentraland’s atlas hasn’t changed much since the auction. The content that’s currently deployed into the world is not displayed on the marketplace map. Some wonder if this may have been a factor leading to several anomalies of parcels having sold for enormous sums of money.

We’ve analyzed the blockchain a few times since September 2018 to see how much content was deployed to Genesis City.
– In September 2018 there was ~63 parcels with content deployed
– In January 2019 there was ~100 parcels with content deployed
– In July 2019 there were 24,000 parcels deployed*

(*see UPDATE at the end of this blogpost)

The Cryptovoxels map shows content that’s currently deployed to the city as well as analytics and other useful features. Anyone can jump into the world right now and try before they buy.

Content Creation Pipelines

In Decentraland, the content creation pipeline is asynchronous and somewhat difficult to master: publishing custom content requires users to know command-line and editing JSON files. For any custom models you will have to rely on using the SDK and setting positions of objects manually through code. Earlier this year, a simple drag-and-drop editor for novice users was created, called the Builder.

The editor for Cryptovoxels appears in-world when you press the Tab key. You can edit and publish to the content server seamlessly and in real-time, similar to games like Minecraft. You can add or remove different types of blocks to build any shape you want. You can even further decorate it with images, audio, art, texts, hyperlinks, ERC-721s, GIFs, etc. Changes to the parcel are saved automatically so that if you log out and log back in you see the changes persist. 

Land Sales

Decentraland has had two massive auctions of land parcels. Parcels in Decentraland were auctioned in December 2017 at prices averaging around 1,000-2,000 MANA, where a record breaking US$28 million was spent on virtual property. All the MANA spent on LAND and staked into Districts (themed areas) was burned after the auction, lowering the overall supply of MANA. Individual parcels have been sold on the secondary markets for very high prices, with some premium lands going for as high as US$32,000 in MANA just this year. In one extreme case, I reported in February 2018 that someone had actually spent over US$120,000 on single parcel of Decentraland’s virtual land!

There was no massive auction for Cryptovoxels land parcels; the project started off as a very small community that has grown bigger over time in an organic fashion, as parcels are minted slowly outwards from The Center. CV has grown by leaps and bounds within the past few months, as can be seen from this comparative illustration:

Jin reports:

Average land sales from Cryptovoxels are beginning to catch up to the Decentraland market. However, it is worth noting that the entire market for DCL post-auction is now second hand. Cryptovoxels did not have a massive auction and instead mints new lands with procedural generation scripts for the size of each parcel and road.

Average Cost of Land Parcels

Land in Decentraland is significantly more expensive than Cryptovoxels. Currently, the lowest price for parcels in Cryptovoxels is about 20-25% that of the lowest price of parcels in Decentraland.

Client Software

Decentraland started with a client based on A-Minus in 2017, then a Babylon JavaScript client in 2018, but has since switched to a web-browser-based Unity client, which is currently in closed beta. There is also a JanusWeb client, which is unreleased. The majority of deployed content features a low-poly cartoon aesthetic but the SDK supports any glTF models that fit within the scene limitations.

Cryptovoxels has a variety of clients in development: the default Bagbylon JavaScript client (https://www.cryptovoxels.com/play), which requires a standard web browser or the Oculus Quest default browser. There is also a JanusVR client (unreleased), an Exokit client (https://github.com/exokitxr/exokit), a Substrata client (http://substrata.info), and an open source Unity client and plugin. Over this past summer, there have been experiments with importing Cryptovoxels into VRChat, the largest social VR platform in terms of user count and the best platform for a custom avatar experience.

One significant difference between Cryptovoxels and Decentraland is that Cryptovoxels supports users in VR headsets, while Decentraland does not, and it is unlikely that the platform will do so anytime in the near future.

Some Final Figures

	Cryptovoxels	Decentraland
Market Cap	US$142,000	US$35,800,000
Current/Accessible Supply of Land	1,246 (Current)	45,000 (Accessible)
Total Supply of Land	3,026 parcels	90,601 parcels

---

This blogpost would have been impossible without the tireless work of CL and Jin, from whose reports I drew most of this information. Thank you!

*UPDATE 4:16 p.m.: Apparently, Decentraland (the company) is very unhappy with this blogpost, and I have been approached by a representative of the company who tells me that “your latest article contains lots of discrepancies and out of date data”.

The company feels in particular that Jin’s portrayal of Decentraland is unfairly negatively biased, but when I asked the representative for a list of concrete examples of errors made in this report, all he could give me was one figure, “In July 2019 there were 24,000 parcels deployed“, to replace one of Jin’s statements, which I have now inserted above.

I appear to have gotten myself caught in between two sides of a dispute, with Decentraland (with whom I felt I had a very good working relationship) on one side of the argument, and Jin (with whom I have worked before without incident) on the other side. I hate being caught in the middle like this, and I don’t appreciate being caught in the cross-fire.

From my perspective, the blogpost I wrote today seems to be very even, not painting either company in a bad light in any way whatsoever. I pride myself on being as accurate as possible in my reporting, especially where facts are concerned, and if a company has a serious problem with something I have written, then I will certainly address the issue, BUT I NEED A LIST OF WHAT THE FACTUAL ERRORS ARE AND WHAT THE CORRECT FACTS ARE SUPPOSED TO BE. And so far, I have only had one factual error pointed out to me, and not “lots of discrepancies and out of date data”, which is what I was originally told by the Decentraland representative. The company seems to be very upset about how this blogpost makes them look, when I think it makes them look pretty good. I’m very confused. What did I do wrong here? This episode has just left a bad taste in my mouth.

SECOND UPDATE Sept. 13th: I have since received an apology from the DCL representative, which I have accepted. He had been at the end of a very, very long workday, and was not at his most diplomatic in asking for corrections, and I took what he said the wrong way. We are both moving on from this unfortunate episode. This is just a bump on the road forward.

There is still much left to write about Decentraland, and (as my regular blog readers already well know) I will not shy away from reporting both the good things and the bad things as they happen, at all the companies working on the various social VR platforms and virtual worlds I cover on this blog.

Sometimes we just need to take a step back and appreciate just how far we’ve come in the development of all these projects. Both Cryptovoxels and Decentraland have come a long way in a very short time, and both are truly pioneers. I look forward to seeing how both develop and evolve over time and I wish both teams the best in their future endeavours.