View Stats:
 This topic has been pinned, so it's probably important
Misfits Attic  [developer] Aug 22, 2015 @ 4:45pm
Breaking Upgrades design discussion with dev (spoilers)
(Edit: 2nd set of continued thoughts from me HERE)

(Edit: continued thoughts from me HERE)

OK, so this is a tough one we may have to iterate on, but I think we can get the best of all worlds eventually. (slight spoilers)

Note: some of this discussion was going on in THIS thread, and others were commenting in other threads, so I'm trying to consolidate things here.

The problem:
Players are feeling like upgrades breaking is unfair/annoying, especially because it feels random.

The intent behind the current design:

Why break?
I want players to have to adapt to survive. As such I don't want them to be able to find a favorite upgrade/strategy and rinse-repeat. By shuffling their hand, players are forced to continually adapt. I *think* players understand this part and are ok with this.

Keep in mind that this also means repairing an upgrade needs to be somewhat cost prohibitive (or have diminishing returns) unless absolutely necessary, otherwise you can circumvent the whole system of shuffling upgrades and scavenging.

Why isn't breaking currently based on direct usage?
I didn't want people not using upgrades for fear of them breaking (I always hated this in RPGs) so a continual deterioration seemed the solution (smoke 'em if you got 'em).

I also tried to make upgrades quantity based, but some upgrades I wanted you to be free to use willy-nilly (sensors) and others quantity didn't work (I like that Stealth is like holding your breath, where it recharges over time).

Other items you rarely use (like tow) and so making it wear down during use seemed like a small window, and breaking during a tow could be heart-breaking (maybe ok?).

Possible Solutions

Some of these are easy to try out, while others are a little more complex. We could try some and see how they feel. They're also not mutually exclusive.

No breaking on the same mission as an error
If this was the case then you'd at least know going into a mission that an upgrade *could* break (if it was showing errors). By the same token you'd know an upgrade couldn't break (if it was not showing errors before the mission). You also wouldn't feel time rushed after errors started showing on an upgrade.

Upgrade States only change between missions
Again this avoids the issues mentioned above, but requires notifications after the fact and feels a little odd.

More info on an upgrades status
It could be that knowing the state of an upgrade a bit better could alleviate some frustration. Perhaps if the upgrades have an indicator of status from 0-10 players can make more informed decisions, and feel that they gambled by bringing that upgrade down in the first place. In this case we could remove the erroring warning and just have upgrades break somewhere between 0-3 or some such. This could also allow us to make them behave oddly at lower states which could be fun :)

One issue with this is we need a solid UI indicator in a somewhat crowded UI.

Base the states on usage
This would make people feel better that degradation was tied to usage, but it might be tricky for some upgrades, and encourages players to not use upgrades, which I'm not in love with. It could also be a mix of things that included usage as one factor eg: # of missions, time, and usage.

I have a lot more thoughts but let's discuss. Help me make an awesome game :)

If you haven't played the game for long please indicate that in your comment.

Thanks for your help!

- Tim (Duskers creator guy)
Last edited by Misfits Attic; Aug 27, 2015 @ 6:03pm
< >
Showing 1-15 of 220 comments
PeanutGalaxy Aug 22, 2015 @ 5:02pm 
Im thinking the first solution should be combined with an in-game limit for the max number of errors per mission. i mean even if all my stuff works for the rest of the mission i cant really pull myself back from having 4-5 errors in one mission.
DemonofRazgriz Aug 22, 2015 @ 5:22pm 
Yeah... I had 4 upgrades die in 5-10 minutes with no prior errors that sent me scrambling back to mothership on my previous mission. That was almost a little too catastrophic, haha. I am for that happening to maybe only one or two max per mission on a big interval.

I like option 3 with the info on an item's status. It could just be a status bar of some sort or a percentage, nothing too complex. Maybe give an alert around 50%, then 30% and from 30% and under, it can have that unpredicable factor that way so it leaves the user gambling essentially.
Misfits Attic  [developer] Aug 22, 2015 @ 5:22pm 
Currently each upgrade chooses 3 numbers:
- number of missions before an error is possible
- amount of time until it errors
- amount of time after it errors till it breaks.

With this randomness it's unlikely to have 4-5 errors in a mission (unless your rolling deep with 12 upgrades). I'd love to not mess with the randomness if possible. I could also add that as an option as well as a lot of other difficulty/gameplay settings...

[Edit: ok looks like that happened, I could limit it I suppose, but then it might back up everything in the Queue I suppose]
Last edited by Misfits Attic; Aug 22, 2015 @ 5:24pm
PeanutGalaxy Aug 22, 2015 @ 5:27pm 
perhaps if we had more warnings. its not just working -> error -> broken. maybe error -> malfunction (still works, just not very well. like motion sensor will start giving false detection and tow will begin to randomly detach when in use) -> performance continues to degrade -> final warning -> broken.
DemonofRazgriz Aug 22, 2015 @ 5:29pm 
I wish I could go back and take a screen shot of that mission... I know it was at least 3, but I do remember 4 with no errors die. I did have 4 drones in with 3 upgrades a piece, some with the same (2 with generators, gathering, and motion)

So that all could have been a factor for sure as you stated.

I do love the randomness of it all; I guess it is just finding the right balance.

And if I hadn't said so before, I love the game and the idea of it all! Great work!!
Last edited by DemonofRazgriz; Aug 22, 2015 @ 5:35pm
LAJ-47FC9 (Liara) Aug 22, 2015 @ 6:07pm 
While the breaking/repair mechanic fits this game perfectly, it can sometimes really break the progression of the game if it happens to something like an interface or a tow module. Plus, it's just kind of annoying in general. At the very least, I would prefer it if modules didn't have a chance of breaking on the same mission as erroring.
Misfits Attic  [developer] Aug 22, 2015 @ 6:26pm 
Yeah, I think that's the minimum change to see if it works better. Actually I could get that in really quickly...
Nate Aug 22, 2015 @ 6:52pm 
Originally posted by DemonofRazgriz:
Yeah... I had 4 upgrades die in 5-10 minutes with no prior errors that sent me scrambling back to mothership on my previous mission. That was almost a little too catastrophic, haha. I am for that happening to maybe only one or two max per mission on a big interval.

I like option 3 with the info on an item's status. It could just be a status bar of some sort or a percentage, nothing too complex. Maybe give an alert around 50%, then 30% and from 30% and under, it can have that unpredictable factor that way so it leaves the user gambling essentially.

Maybe the most "fun" way to do this would be to have an adjectival indicator of status, like:

Stealth (New)
Stealth (Like New)
Stealth (Functional)
Stealth (Worn)
Stealth (Damaged)
Stealth (Failing)

That way, you'd never know *exactly* when it's going to totally break, but you'll at least be able to look at it and make the call whether to repair it or not.

Don't forget, too, that keeping things "like new" can become a Scrap sink, which I think will keep people from doing the rinse/repeat thing on a strategy. As it is now, the expense makes it a bad ROI to repair something; this gives more control to the player without really unbalancing (I think), especially if the repair only kicks you up one level.

You could also semi-randomize the adjectival rating to add more uncertainty, like:

New = 100%-85%
Like New = 90%-75%
Functional = 80%-55%
Worn = 60%-35%
Damaged = 40%-15%
Failing = 20%-1%

Just randomize it after each mission, never letting it go higher once an adjective has been chosen, and after each repair. This can cause some evilness like "I spend 3 to repair my Failing item to Damaged" which actually results in only a small % improvement but a full adjectival rating increase.

Love the game!
Professor Hugedix Aug 22, 2015 @ 7:37pm 
adding even one state of disrepair in between "showing errors" and "busted beyond repair" would alleviate a lot of frustration I reckon, having things go from working perfectly to being in danger of disintegrating any moment is very jarring. I like the idea of there being a stage where the module starts to actively malfunction, but it could become annoying given a few repetitions

what about a duct tape module, that lets you patch up another drone's upgrades to guarantee that they won't malfunction any further during the current mission? the upgrade would still need to be repaired afterwards but you could at least save your most precious module from spontaneously exploding and causing a bunch of heartache
Blu C Fox Aug 22, 2015 @ 8:17pm 
I commented on this in another thread, so I'll just add that here (although what I have to say on the matter is similar to Nate):

Malfunctions/Failures. Overall I actually really like this mechanic and very much hope it stays in, but at present it feels too random. I've had thing I've never used before malfunction without ever being used.

My suggestion is that drone upgrades have a semi-obfuscated condition level (0-100%). Players would just have a general indication of the condition, without knowing exactly how good it is. From 90-100% there would be no failures, however as condition degrades malfunctions gradually become more common. Playing into this malfunctions would be cleared automatically at the end of a mission, simulating basic maintenance post-mission, however actual failures would need to be repaired (likely fairly cheap for most components).

The condition would degrade slightly at the end of every mission, to signify general wear and tear, with additional degredation possible for actual use. In some cases this may be a per activation (say scan, motion, probe), other times it could be by time activated (gattling, shield, stealth). The exact degredation could be dependant on the type of module (a scan is likely to degrade faster than a probe, as most wear happens on the actual probe itself which is lost upon deployment).

The above could ultimately play into the modifications available for different tools. As an example, I could envision a gatling that you could upgrade so that it can be left on while moving, however it would degrade much faster than a normal gatling seeing the same amount of use. In effect you trade reliability and increased maintenance costs for better capabilities.

Finally, this could play into the salvage equation and risk/reward quite a bit. Most upgrades you find will be of varying levels of quality, but perhaps certain locations are known to generally have higher condition upgrades available to salvage.

In a second comment in the same thread I realized I had forgotten the following:

Players would have the (expensive) option to refurbish parts and restore their condition, however this would carry some sort of penalty. Perhaps each time you refurbish a part it gets a little more expensive. Or perhaps it isn't available for a set number of days after, since it is an intensive process. Perhaps both!

This would let players weigh the risks of using worn equipment. Do they just repair the failure for cheap, knowing that as time goes on there will just be more and more failures that they will have to repair each time? Is it too soon to refurbish it and was that malfunction just bad luck? Maybe you can't afford to restore the part's condition until after this mission, and now need to plan around the real possibility that something important could fail at an inopportune time.

Sorry for a simple copy/paste from stuff I said elsewhere, but it's quite a bit to type out again so even though it echos some earlier sentiments I figured I would just add it all after I was pointed in this direction.
AntiBlueQuirk Aug 22, 2015 @ 8:22pm 
As I mentioned one of the myiad other threads you mentioned, this is the method I think I personally like the most, and keeps a lot the current perks and style, without the RNG stealing things from the player.

  • Items have three states, normal, errored, broken.
  • Items progress from normal to errored just as they do now. (overtime/random, etc.)
  • Items only permanently break if the player uses them while errored.
  • The chance of permanent breakage is high, but there's still a chance you'll get away with it.

This has the following effects:
  • This keeps the "ship happens" style the game has, forcing the player to adapt.
  • The player is only permanently penalized if they push the odds.
  • A player may continue their current mission without fear of their equipment breaking, but they must change their style immediately to do so.
  • Bringing backups is now important as opposed to penalizing the player.

Personally, I'd like the option to repair equipment that isn't broken as well. This would reset the timers and wear counters, allowing the player to do "preventative mainantence" on critical equipment. This doesn't have to be any cheaper, for balance reasons. (You don't know what's broken, so you just replace everything.) I also don't see any reason to tell the player when stuff is going to break. I'd think that's its rather difficult to tell the status of equipment you've scavenged from random places, and then strapped together with duct tape.

But that's my two zorkmids.
Last edited by AntiBlueQuirk; Aug 24, 2015 @ 12:21pm
Blu C Fox Aug 22, 2015 @ 8:39pm 
With the copy and paste out of the way, I'm going to add a bit more:

I'm clearly a fan of having some sort of a condition rating for different upgrades so I figured I'd lay out why.

1) It makes having a reserve of parts more important. Right now the only reason to keep spares is to offset the pain of losing the upgrades with a drone. The condition rating makes building a pool of spares more important as things gradually wear out and either break perminantly and need to be scrapped or be repared. This adds an extra dimension to consider where now players with extra upgrades can just scrap them for a bit of extra resources. Doubly true if different upgrades wear out at different rates.

2) It adds a maintenance and risk/reward dynamic. On the maintenance side making sure your drones have good condition upgrades increases their reliability during a mission. On the risk/reward dynamic there is the reality that losing a good condition upgrade will hurt more than something that is really worn out. So that being the case do you risk 'good' upgrades on very risky missions or do you take something that is beat up even though it may fail and make dealing with the challenges that much more difficult.

3) It really adds a lot of depth to the game. Newly discovered upgrades can be of different conditions. Things like mining ships may have more upgrades lying around, but they are also probably well used. On the flip side something like a merchant ship probably carries like new stock. This gives players other things they can weigh when considering what mission to take (the safe mine ship for a lot of easy to grab spare parts, or the merchant that looks hazardous but has a chance of getting a couple like new things).
Lastpick72 Aug 22, 2015 @ 9:01pm 
Okay, How about something a little more original. Perhaps, much like you have already added effects like the video going out why not take that concept and expand on it for each of the upgrades. So instead of outright failures leaving one up the creek without a paddle, one could Macgyver their way through a mission if they have the perserverence. I could see things like the tow randomly disconnecting while in use, or the probe and motion detector reporting false images or maybe none at all [imagine being fooled by that], the gatling could misfire targeting a friendly, the teleport could send your probe to the wrong location or just teleport the drone without any of it's upgrades. Just don't make it the defacto standard to ruin well laid plans because of a rng time I can't see, it's really annoying to have a drone prepped and safely stowed in the ship suddenly break for no apparent reason just when your finally ready to use it. I'd argue that a player should always have a way out if they use what is given to them, it is a game after all.
Digital Aug 22, 2015 @ 9:11pm 
How about a combination?

So you have whatever random criteria in the background that gets you to the point where you start getting an error. Once the error is reported, using the module is more risky.

So if your tow is getting an error for example, you'd be fine if you just don't use it for the rest of the mission. But if you need to use it, you're putting it at risk of breaking.

You could either set it up so that each use increases the risk... or just have it be a flat percent. I kind of like a flat percent, because it encourages you to push your luck.

Also lower the percent based on how often the module tends to be used. You tow maybe 3 things in a good mission, so give it a 50% chance... however you use things like stealth more so have it be a lower chance that evens out. The idea being if you continue on about your day ignoring the error, you've got a pretty solid chance of breaking it.

This fixes the frustration of it "not being my fault" when it breaks, and keeps the random. You want to avoid people using the same combo constantly, this would randomly put components at risk.
damien Aug 22, 2015 @ 9:21pm 
Losing your only gatherer module is very close to a game-over condition. Losing your only tow or interface module isn't game-over, but it's a dire enough possibility that I'd abort a mission rather than risk it.

So unless the goal is for players to evacuate as soon as a critical module starts showing problems, I'd vote for never having a module go from showing errors to breaking on the same mission.

Alternatively, perhaps modules could wear down and show errors after some amount of time, regardless of use, but only break when used. So using a damaged module would be risky, but having your gatherer module show errors wouldn't trigger an immediate run for the exit.
< >
Showing 1-15 of 220 comments
Per page: 15 30 50