Sep 5, 2023

What We’ve Learned in Early Access (So Far)

Jumplight Odyssey - Chan

Greetings Captains, new and old!

Buckle up, this is a long one!

A few weeks ago, LoG launched a game for the first time in nearly a decade. We put everything that we had on the table, and as an indie team – for complete transparency and honesty – the weeks leading up were absolutely exhausting. But that’s what we’re here to do, and what we love - making games.

Reaching the point to be able to launch into Early Access is already a huge milestone. For those of you who have a little insight into gamedev, you’ll know that it's a miracle any game gets made, really. A game of this scope even more so. It’s also exceedingly rare for indies to be able to get as much traction as we have, considering how saturated the market is these days (this month!).

In every way, to every person that loves Jumplight Odyssey as much as we love creating it: thank you thank you thank you.

This little game has our whole hearts, minds and souls inside. The stories that you have shared, your crew’s devastating failures and glorious victories bring us so much joy. We have made a board in our studio sharing all the kind things you’ve been saying to us - we look at it daily!

Overall, the launch into Early Access has been… A little rough to start. A little is maybe an understatement. Somewhat rough? “Mixed,” perhaps?

The resounding sentiment from folks seems to echo what this reviewer has said:

And we agree. The bones are there, JLO is on the right course, but it often feels confusing, hard to understand and – sometimes even painful. The initial instabilities, the lack of player agency and the lack of game feedback are all undermining your experience as a Captain.

Note: If you’re still experiencing critical save bugs or crashes, please make sure you have updated your game! We have deployed literally dozens of fixes in the past week that address a lot of the instability seen at launch.

As we’d hoped, we have learned SO much from this first week of Early Access.

We knew that when we dropped Jumplight Odyssey, we were punching completely above our weight as an indie studio. So, in the interest of complete transparency with you, our community, we thought that we would take the time to delve into exactly What We’ve Learned from Early Access (So Far)…

Learning #1: Responding Quickly & Decisively to Critical Bugs Had a Greater Impact Than We Ever Imagined

This isn’t our first rodeo. We knew to expect the likelihood of a rapid hotfix at Launch, especially for a game as big as JLO.

We knew coming in to launch that there were some bugs, but unfortunately, the intel that we were able to get internally from our own testing couldn’t foresee how prevalent and far reaching the issues with save games would be.

(Shoutout to Raven/Half an Unkindness for this meme – it still makes us laugh)

The Save Game & Other Stability Bugs

Bugs are expected in Early Access, but critical bugs that eat your saves and crash the game… Yeah, they need to be zapped ASAP.

With the state of save/load in the game being so much worse than we foresaw, the need for a rapid hotfix, and the complexity of the task at hand, increased ten fold beyond our expectations.

We pulled out everything we had to make it happen.

The real worst part about this bug was the damage it caused to the faith our greater community had in us as a development team. We received commentary like:

“The game has potential but I think League of Geeks have shot themselves in the foot by releasing prematurely."

"This game is riddled with game breaking bugs which are made fatal without a working save function. It's almost as if no-one bothered to test it at all, perish the thought."

Among many others in a similar vein.

We’re not going to lie, or shy away from the truth: reading that stuff hurts! It is so hard to see so much love and dedication from the team shrivel in the face of the mega-save-bug. Especially when we hold ourselves to such high standards.

We steeled our resolve, came together as a studio, broke down the problems and supported our Production, Engineering, QA and Community teams in coordinating that hotfix as fast as possible.

The hotfix was deployed only 48 hours after initial launch!

For an indie studio, and faced with such complex bugs (save/load bugs are notoriously brutal) that is a huge accomplishment. We’re very proud of that. Not only did it resolve most of the large-scale instability bugs, we managed to also add some critical quality of life improvements and band-aid fixes the game desperately needed. It was a meaty hotfix.

And you all loved it. I think some of you were more amazed than we were. Haha!

But there was more work to be done. So...

Within a week we patched the game again. This time with over 100 (!!!) changes, fixes and improvements.

Most amazingly, the response from all of you to these two rapid patches and our constant engagement has been so much more positive than we expected. So we're going to continue to focus (as much as is reasonable and possible) on rapid fixes over the development of big, shiny new features. You’ll hear a bit more about what we mean by that throughout this post.

Steam Reviews & The Almighty Algo

If you have been around the indie game space for very long, you know that Steam Reviews act as a form of tribute to the unholy Steam algorithm. Steam will automatically promote (or not promote) your game in different places around the store, based on your Steam review score.

Upon our initial launch we jolted up to the “Positive” and “Very Positive” rating in Steam Reviews – which was a HUGE win for our team!

But then folks started to experience the Save Game issues, other launch instability, and the Three Player Pain Points (more on that later), and so our review score started to tank.

Right now, at the time of writing, we’ve dipped down into “Mixed”.

Which we totally understand – don’t get us wrong for a second here! We’re grateful to everyone who’s taken the time to play and review the game. Steam reviews are simply another feedback input for us.

The unfortunate kicker is that it can be very hard to recover a Steam review rating once they decline. But we're confident that every issue folks cite in Steam reviews is something we can address in time.

There have been so many games that have launched into EA completely buggy, with radio silence from the Development Team and no resolution offered for a long, long time. We are steadfast and eager to demonstrate that Jumplight Odyssey is NOT one of those games. We’re determined to deliver upon the promise of this game, recover your faith in us and with it, that Steam rating.

It was – and will always be – integral to the foundations of League of Geeks that the community’s voice is heard. Our Community Team goes to excruciating lengths to ensure all of your bugs, feedback and suggestions are captured, collated, prioritised and communicated to the team.

So far, at the very start of week three for Early Access, we have:

268 bugs captured from Steam, Discord, Reddit, Twitter & the In-Game Bug Reporter
1200+ pieces of feedback & 500+ suggestions
All of which are processed, compiled and passed onto the rest of the team

A tiny glimpse of one of our working boards!

And with just the last two patches we have already seen a slow but meaningful uptick in positive reviews, going from 61% to 64% positive since the release of the patch.

As we continue to listen, learn and implement the necessary improvements to the game, our hopes are that this rise not only continues, but accelerates!

Concluding our First Learning:

To conclude the first learning, with the hotfix and patch launched, and our commitment to engaging with you in meaningful and impactful ways, we actively hope that your faith in our ability as a studio only grows from here - and that the game becomes progressively better as rapidly as possible.

It’s so vital to the foundations of our community and to your ability to provide real impactful feedback, that you are actually able to play the game without running into game-save errors and other fatal crashes.

Interestingly, 50% of our Negative Steam Reviews mentioned that the game has promise/potential. To those community members: We see you. We understand. And we’re going to work our hardest to deliver a game deserving of your positive review – as soon as we can.

Learning #2: The Three Player Pain Points

This game is huge. The brief was a bold one for our studio. For any studio. Take FTL and Rimworld, smash ‘em together, but with a level of polish and aesthetics greater than Two Point Hospital and Evil Genius.

In order to achieve this impossible task, we came to Early Access, well, as early as we could. Some things we knew were going to be hard lessons to learn. This is one of them.

With a sprawling set of complex systems that interlock with each other, and such a breadth of content, y’all taught us preeeeetty quickly that almost every issue you were encountering would result in one major issue…

Failing through no fault of your own.

Yeah, the ultimate videogame sin.

When diving into your feedback we found there were three major areas causing players to fail runs through no fault of their own, and in the worst scenarios, all of them are affecting you at once. These have come to be known internally as the Three Player Pain Points.

The first one is stability, but we’ve spoken about that already, so…

2. Lack of Player Agency - Control as a Captain

There is – perhaps – nothing harder in the world than standing idly by while people struggle to do something that you know to be simple. To watch the crew – your crew, who you’ve vowed to steadfastly protect and shepherd to the safety of the Forever Star – struggle to do their jobs, let alone even take care of themselves, is beyond frustrating. We feel it too.

Crew AI problems are so multifaceted and intrinsic to the fundamentals of JLO that to examine each part, you must have context on the other parts. So far, we’ve identified a few key problems among the community:

Crew aren’t taking care of themselves. They won’t prioritize their own needs, and will sometimes drop dead rather than simply… Drink some water, or go to the galley to eat. This is especially prevalent if they’re On Shift.
Crew need to prioritize important functions – Power, Water and Construction. They also need to be “smarter” in understanding that certain occasions call for a bit more priority than other functions (moving bodies to the Morgue is a great example here).
Crew’s pathing needs more work. Not only can crew struggle to navigate the ship if buildables aren’t spaced obnoxiously far from each other (a minimum of two spaces, usually), but in rare instances they can cluster (shoutout to The Clump and Moshpit bugs respectfully), get stuck (on furniture, spinning around, etc.) or simply lag around – doing nothing.

Ultimately, there is a long way to come with fleshing out the effectiveness within our AI, and admittedly some of these things are bugs. But, there are also impossible questions to answer when dealing with AI; Should a crew member leave the metal fabricator to eat, or should they sit and starve? The answer is different depending on the context in that moment, in that run, in that alert mode, even with that particular crew member. And those combinations are infinite. The answer is even different depending on the player; does this particular player value narrative that emerges from crew adhering to their characters, or is it a player who is simply playing to win and demands ultimate effectiveness from their worker drones?

Impossible problems, yes. But still, the problems we have to (and are excited to) solve. The immediate answer for us is cleaning out the bugs and objectively undesirable behaviour as soon as possible, and then providing you with more meaningful agency and ways to manage and direct your crew - all through the context of being a starship captain. And we have a lot planned.

You, as the Captain, should be able to be alerted of a Critical Condition, then be able to easily find that person specifically. Then, you should be able to actively direct your officers to remedy the situation in a way that is meaningful to you and your run.

You should be able to influence where items are stored, and should be able to actively select who is the most eligible for Away Missions at a glance.

Additionally, we are looking specifically at more meaningful ways to effectively manage your resources and affect agency over them.

Power specifically has been brought forth by many in the community as a frustrating resource to manage. If you are not immediately kicking off your run by creating a Nuclear Reactor, you’re bound to almost immediately run into some problems. This has given us a lot to think about not only with game balance and feedback-player-loops, but also has given us insight into what information is integral to provide in the Tutorial for new Captains.

It is undeniably hard to balance a game as expansive as Jumplight Odyssey. Additionally, part of the fun of being in Early Access is stress testing and gauging whether the balance “feels” right. The game should be challenging -- hard, even – but not impossible, and it shouldn’t feel like it is tricking you, withholding information or robbing you of success.

3. Ineffective Communication of Game State Information

How the hell is a Captain meant to make mission critical decisions when they don’t have the information required to make mission critical decisions?

A lot of the recurring feedback we’ve received over these first two weeks of Early Access has been to do with the intricate, almost impossible dance between managing your Crew, Water and Power.

Many Captains have spiralled out and have lost runs because there simply isn’t an abundance of useful information that supports making effective decisions as a player. Sure, they can choose to purchase a rice crop over a pea crop, but unless they have hours they’re willing to sink into the game to trial the difference, there’s no source of truth to provide insight on the efficiency of their decision.

On top of that, there’s no source of truth to provide insight on whether that decision is causing a massive, catastrophic failure somewhere else aboard. Perhaps, in building three new algae crops to support the Crew, the Captain does not realise that their current water systems aren’t able to support so many new water-sucking additions until it is too late – and they have no water.

Or maybe you have to rotate your Shuttle crew out because they’re likely very tired, but you’re forced to guess, and cannot see the needs of the many crew you have to choose your replacement pilots from.

All of the above, compounded by save file issues and general Early Access instability, means most players then find themselves at the start of a deep, deep whirlpool down. Imagine like Pirates of the Caribbean, where the whirlpool opens up. That’s the Death Spiral that begins once one system fails. And without the information you need, only the most hardcore of players can recognise when it’s begun - and manage to survive it.

Y’all remember this one? Yeah, this is what the Death Spiral looks and feels like.

We’ve received lots of feedback that Captains want a control panel of sorts, where they can see at exactly what rate main resources are being consumed – especially both Power and Water. This would help players make decisions on where exactly their focuses should lay, and could even show important trending information that a Captain would want to know about their starship.

We actually already have this planned on the Roadmap (a Situation Report screen where you can get an overview of various things and trends, as well as numerous improvements to UI and tooltips) - but it's a big and complex thing to build. For us, knowing your pain points here is an example of the exact kind of thing we were hoping to learn in Early Access - as now during the Early Access phase, we can build it based on better, more accurate information from you, our players. Instead of making educated guesses in the dark as developers.

We understand this doesn’t take away from your valid frustration, but we do hope it gives you solace that we’re paying very close attention to all your feedback (and feedback needs) here.

We know we’ve said it a bunch already, but that’s precisely why we’re here in Early Access. Which brings us to our next point…

Learning #3: Early Access Means Different Things to Different People

We’re no strangers to Early Access. Our first game, Armello, had a very successful period in Early Access that changed the game for the better in so many ways, and galvanised our community in the process.

However, that was 8 years ago.

Early Access itself had literally just launched on Steam the year before. We knew that now, in 2023, the landscape of Early Access, and player’s expectations of polish, quality and degree of completion, had changed. But we were honestly surprised by how much.

We knew that the systems in Jumplight Odyssey were not fully realised. This was, honestly, on purpose.

Critical bugs notwithstanding, we came to Early Access as early as possible to collaborate with you all. You’ll see in our comms leading up to launch that we messaged this, and our excitement in doing so.

But it’s clear that as much as we can communicate our intentions, we underestimated how much players expect from Early Access titles in 2023. The degree of completion that games are now launching into Early Access with is far greater than previously. Games are almost soft launching into Early Access, as opposed to launching at a point where real, meaningful development can occur in partnership with the community.

With such a huge vision for this game, we still believe in the power of Early Access and getting it into your hands as early as possible. Despite our game sitting in Mixed right now, all of your feedback has been invaluable and we’re so glad to be here with you all, developing this game together.

Launching a game is hard. Launching into Early Access as a self-publishing indie during one of the busiest times ever in videogames is astronomically hard. When there are so many freaking cool games releasing every month, trying to find the perfect launch date AND cut through that noise is almost impossible.

Honestly, how many Game of the Year candidates have released this month alone - and we’re not even in the AAA holiday release season yet?! Wild stuff.

We were in a position where – financially – we needed to launch this year. Any earlier, we would’ve encountered the Baldur’s Gate 3 phenomena (and it was questionable we could release earlier anyway). Any later, we hit the ramp up of Starfield’s multi-million dollar marketing budget (its a busy year for Space games). Later than that, we’re smack-bang in the midst of the AAA holiday release season.

We wanted you all in the game as early as possible, but honestly, for our studio, the options weren’t as vast as you might think. We couldn’t wait another 5-6 months for the next viable window to release. Especially when we’re aiming for a full launch in the first half of next year.

Of course, a lot was done to prepare for launch and ensure it was as stable as possible. We had even reached out to some of our wonderful, core Discord community members who had time to do some last minute bug hunting out of the kindness of their hearts (shoutout Gamebreakers! Your unhinged bug investigation knows no bounds!).

Our goal is to build our dream game and make sure it’s also one everyone wants to play. In order to achieve that and remain afloat as a self-publishing independent studio we needed to make the decision that afforded us the opportunity to maximise our time in Early Access.

Conclusion: We’re on the Right Course, Thanks to Your Guidance

There are so many more insightful learnings and interactions with you all that we haven’t even been able to capture in this update. Honestly, having a Community filled with honest (sometimes, brutally so!) and eager players is an absolute blessing. We’re truly grateful.

And the great news is that, resoundingly, you all believe in this game. From the survey available from the game’s main menu, we have found:

86% enjoyed their experience playing JLO and scored it 4/5 on a /5 scale!
67% say they would recommend JLO to a friend!

Even those of you who are frustrated enough right now to leave negative reviews…

Over 50% of our negative Steam reviews mention that the game has promise/potential.

One of the kindest comments that we received on our reviews was:

“Despite being early access, this is one of the most well thought-out and intricate games I've played in a while... The concept of being forced to flee while desperately caring for your crew never loses its edge. It just FEELS good. If you buy this game currently, you will encounter some minor issues. The developers are very vocal and connected with the community and constantly take note of potential issues. They are always listening to our feedback to help themselves stay informed.“

And they’re right, we are listening. We know many of you have been eagerly waiting for even more patches and improvements that resolve these core problems in playability. We are absolutely, and without hesitation, on it. Each iteration of Jumplight Odyssey will be better than the next.

This week, we plan to release our Starmap.

This Starmap will not only capture our hopes and dreams for the future content of JLO – which you can expect to be released over the Early Access period – but will capture your voice, and the subsequent improvements from what we’ve heard. Our development roadmap is intended to be an evolving thing, not something fixed in place. In fact, a lot of it has already shifted based on what we’ve learned since launch. Our Starmap will be less a direct course or roadmap, and more a look to the stars of what the future holds - what our dreams are, and a vision for how we’ll get there, together.

Of course, we want to delight you with new content, new surprises, new stories.

But the first thing we need to do, and our first focus, is to address the Three Player Pain Points, and get Jumplight Odyssey to a point that will keep you coming back for that nostalgia fix. The only way we will achieve that is through your support, play and bug reports.

For those who left us a Steam review (positive or negative, either is an honour!) thank you. Your contribution will help us gather visibility in the Steam algorithm and share our little game with other folks just like you.

For our core community, our Stargazers, those who have diligently stuck it out, who have lost thousands of crew but have dared to do better each time: thank you.

Onward to the Forever Star.

<3 LoG

Sep 5, 2023

Demo FAQ

Super Dungeon Designer - Squish Studios

Hey Dungeon Designers!

Our Demo is live now! This post will serve as your one stop for anything and everything related to the Demo. This Demo has some items and features that are disabled and only available in the Full Version of the game. One of those features is the ability to upload your dungeons.

Kickstarter: https://www.kickstarter.com/projects/squishstudios/super-dungeon-designer

How do I access the Demo?

There is a link to the demo in the middle of our Store Page, after clicking that link you should have a game called "Super Dungeon Designer Demo" in your Steam Library.

Where can I learn how to play?

Check out the in-game tutorial found in the menu. If you're having trouble figuring something out please reach out either by replying to this post or jumping into the Super Dungeon Designer Discord Channel to ask. Every item has a tooltip, just hover your mouse over the item in the build menu to learn what it does!

What does this button do?

Here is a UI cheatsheet:

How do I browse dungeons others uploaded?

Click the "Browse Dungeons" button from the main menu to check out all the dungeons available in the demo!

How do I play co-op?

First, click on this button at the top of the left menu while building your dungeon

Next, click on this button to add another player. Click it three times to add up to four players

How do I add an exit to my dungeon?

Add an "End Level Tile" to add an exit to your dungeon. You can place multiple "End Level Tiles" and hide them or make them a part of a puzzle!

How do Links work?

Place down some objects you want to link then enter Link Mode by clicking the Hammer/Pencil icon in the bottom left. Click on the Magic Wand icon at the bottom of the screen so your cursor is set to link objects together (when you click on Link Mode it should set to that by default). Then, click on the object you want to trigger a link, then click on the object you want it to trigger. For example, if you click on a Bat enemy, then click on a Treasure Chest - they are officially linked! The Treasure Chest won't appear until the Bat is defeated. Links affect a lot of things in different ways but generally items/enemies will appear when linked to. Some objects will appear or change state when linked to. For example, the Spike Trap will spring up when linked to.

How do Groups work?

Groups are used with linking so you can trigger multiple things at once. Click on a Group Color from the left side menu when in Link Mode. Then, click on any objects you want to add to that group. Now, when you link to or from one of the objects in that group, that link will affect the entire group. For example, if you link from a group of unlit torches to a Treasure Chest, the Treasure Chest will only spawn when all the torches in the group are lit!

How do custom boss fights work?

Click the Link button and hover over a Boss to see various icons. You can hover over these icons to view a tooltip telling you what they are for. You can then click on the icon to link from it. After clicking the icon, click on something else to create a link. The linked object will trigger when the icon's description is met. For example, if you link from the "Boss Splits In Two" icon to a Lizard enemy, the Lizard will spawn when the Boss Splits in two! (the Boss splits when hit 6 times). You can also adjust which icons are triggering what from the Links menu on the right side after creating a link.

How do I play with a controller?

Controllers should register even if you connect them while playing, we recommend syncing controllers to your computer before you start the game. Bring up the menu by hitting ESC and click on "Controls". From there you can assign controllers to different players and also remap any buttons you need to remap. You can also remap keyboard and mouse controls from this menu. Note that controllers are currently only supported while playing a dungeon (they don't work in Menus or build mode). We will have menu support in soon!

How does Steam Remote Play Together work?

Steam Remote Play Together is a Steam feature that we support. For more information check out this link: https://help.steampowered.com/en/faqs/view/0689-74B8-92AC-10F2
We recommend a strong internet connection, especially when you add more than 2 players.

Where can I report bugs or feature requests?

Please report all bugs!! It will really help us out. You can report bugs by replying to this post, or in our Discord channel (link above). You can also email us at contact@squishstudios.net. If you have a feature request you can post that here, or create your own post in the forum to get a discussion going around it. We take feature requests seriously, if we see a lot of hype around a new feature we'll definitely consider it for the game.

We're excited to build some dungeons with you!

-James