We've been teasing our Beach Vault in the last few dev logs, and today is finally the day that it's published!

🔗Browse the Saltwater Skies

A collection of South African beach HDRIs.

To download them, make sure you're signed in with your Patreon account.

As a reminder (since it's been a little while since we published a new one), vaults are asset collections that are temporarily exclusive to our Patreon supporters. Each has a goal - a target number of patrons at which the vault will be unlocked and released freely for everyone. This way, our Patreon supporters directly help fund Poly Haven's mission of creating free assets, get (temporarily) exclusive content, and help release that same content for everyone.

Think of it like reverse-crowdfunding, we make the content and trust you guys to help fund it retroactively <3

Being inland dwellers, beach HDRIs were previously something scarcely found in our library, and looking at our search statistics on the site, we though it would be a great idea to make a few trips around the South African coast to capture some beaches.

Dario and Jenelle both made separate trips earlier this year while it was warmer, and captured over 100 HDRIs in total. Most of these were for this vault, though some were for another upcoming vault or to publish freely.

Perhaps later this year it would be fun to go back and supplement this HDRI vault with some texture scans, rocks, and new palm tree models. Comment below if you think this would be worth doing :)

Many of these are also eligable to be turned into Pure Skies, so perhaps we can do a big update at some point!

---

In general, the whole "vaults" idea has been working really well for us, both in terms of funding growth towards our goal of releasing our add-on for free, and also by helping us make themed content packs that people actually want. In the process, we're finding that working in defined and focused "sprints" on particular projects to be quite effective.

In the spirit of transparency, I thought it would be good to track our budget on this project and compile a bit of a report here for anyone who wants to know more practically how our Patreon supporters are funding things...

1. Trip Costs

Contrary to what most people think, travel expenses are by far the smallest cost of any asset collection.

This was a specifically local trip, too, which means exceptionally low costs.

On the surface, $2k to capture 70 HDRIs is crazy, that's just ~$30 per HDRI! And if we count the HDRIs we captured that aren't for this vault, it goes down even further.

The real cost of the HDRIs actually comes from the work involved in processing them, and the time spent there...

2. Processing Costs

We've been using Notion for some time now and track roughly how much time we spend on everything we do using a semi-agile workflow.

Being a numbers and spreadsheets guy, one of my favourite benefits of this system is being able to look at timelines and charts of each project like this.

Using this data, we land at a more realistic number of about $17k total (~$240 per HDRI) accounting for all the processing time, reviews and revisions, admin, and other misc tasks on the project.

3. Past and Future Costs

This doesn't account for all the camera and computer equipment we already have, nor the rest of our pipeline tools already developed, nor the future costs of hosting the files. But it's certainly validation of the fact that HDRIs are by far the cheapest type of asset for us to produce (when compared with textures and models).

Also to be considered is the fact that Dario and Jenelle are (or were) fairly new to the workflow of stitching HDRIs, and many of them were also especially challenging to stitch due to the clear skies, open horizons, and moving waves. So next time, we might process HDRIs a bit quicker, but we'll probably also have different challenges to face as well.

We also consistently hit our heads against what I call "the butthole of the sky", which is a phenomenon I've encountered before but was particularly present in many of these HDRIs.

I believe it's caused by unintensional partial polarization of the images, possibly from the camera sensor's micro lens array itself, resulting in subtle exposure shifts near the zenith depending on the camera's orientation relative to the rising/setting sun.

Super interesting, and also super annoying.

Going forward, we plan to include a dedicated zenith shot to work around this. Even though it's not needed for full 360° coverage, it'll help avoid blending issues near the pole.

Setting the Patron Target

When setting the goal for how many patrons we want to have before we unlock a vault, part of the equation is how many new patrons we need to cover the costs of making it.

Given our average patron historically supports us with around $41 in their lifetime before cancelling their membership, it takes about 5.8 patrons to fund one HDRI. Given the 70 HDRIs in this vault, that means we'd need 415 new patrons to cover the costs of the vault itself.

To help subsidize the other asset types and more expensive projects (remember, HDRIs are by far the cheapest type of asset for us to produce), I thought it's worth setting the goal a bit higher than this.

We also considered the bigger picture of our roadmap, the general incentive of the vaults and our other Patreon rewards, and the eventual target of releasing our add-on for free as part of the decision.

Ultimately, we landed on setting the patron target to 550 patrons more than the previous vault, i.e. 4250. I think this is reasonable and realistic to achieve within less than a year given our current growth rate.

---

Anywho, if you have any questions about the project, feel free to leave a comment below or poke us in our Discord. We always love to chat :)

Thanks for reading, and again thanks for your support!

-Greg