PAT19 Twilight

Welcome back, to another post on a Stars Reach pre-alpha update. This time, the update is called “Twilight” - and the name has been chosen, because night-time in the game is less dark now.

Well, that seems a rather fruitless endeavor to me - because there’s always people with old/bad monitors, or mismatched contrast/brightness settings between OS and monitor, or playing in a too bright room, or just with bad eyesight… you can crank up brightness as much as you want - it’ll always be too dark for someone. There is only one way to fix this: a gamma slider, so that everyone can adjust the brightness to their own liking.

Night is slightly less dark now - it’s more noticable in fullscreen

But that was a minor change anyway. To me, the actual headline topic, the big one, is that the materials stats now affect crafting. That’s the juicy part! And I’ll talk about that extensively a bit further down. This update is a pretty big one. As was the previous one - lots of bigger changes and moving parts around.

The Issues

With more and bigger changes and larger updates, we are also seeing more and bigger bugs and issues. The last update ran into severe performance problems, which caused the planet Herastag to be taken offline completely. Herastag is now back and is now one of the least lag-affected planets of all. But I’m getting pretty low framerates on Akacala now instead. That’s a planet that has a number of player buildings on it - which is probably the source. The huge starship being built in space is a source of severe framerate drops as well. But there seems to be more going on than that. I dropped down from playing at 4K to running the game at 1440p - and despite that drop in resolution, I often got lower fps now, than I had on 4K previously.

There are more performance improvements in the pipeline - and the new shaders (which were planned for, but didn’t make it into this update) are supposed to improve things quite a lot. But performance issues aren’t the only issues I’m seeing.

• Yesterday my game crashed hard, after already having 60 gigabytes of RAM and wanting still more. But there simply wasn’t more to be had.
• The new UI is great - but with the UI now being half old, half new, there’s a load of issues as well.
• Tool selection sometimes doesn’t match up what’s actually selected, tools often don’t show up at all (neither in the characters hands, nor the UI) - but they still work.
• The issue of sounds sometimes cutting out seems to be back.
• Sometimes things are just white and have no textures on them.
• I’ve experienced all the trees and plants flickering in and out of existence. (Epilepsy warning on that one!)
• Severe lag waves and rubber-banding seem to happen more frequently.
• The rubber-banding strangely also affects the camera - which can be a bit nauseating. Imho the camera should always, strictly move with the mouse, and only the mouse - and without a server-roundtrip delay or anything.
• Several people have reported loosing items by swapping them in and out of the toolbelt.
• The maker scanning function previously got forgotten completely. Now the waypoints are gone instead.
• Seasons had to be turned off because they were causing issues.
• Some of the one-time info popups I’ve already seen 10 or more times, and they keep just randomly coming back.
• When playing with fixed reticle mode, there’s an increased number of situations where you end up with no free cursor when you need (because a dialogue box is open) - or the free cursor remaining in place after the dialogue has been closed, with pressing “T” to mode-switch not working either.
• We’ve seen a few character corruption issues again - not as bad as “Player Name” or “0-0-0” - but with people getting stuck and being completely unable to move.
• Two items have bonuses not assigned to any PQRV value
• The search function gets weird if you search right after opening your inventory - but works fine if you wait a few seconds before searching.
• The crosshair is now overlapping my character (if it’s visible at all)
• Terraformer tailings are broken everywhere except the wild planet
• Probably more stuff, that I never noticed or forgot to mention…

I know it’s pre-alpha, which is a quite early development phase, where all these sort of issues aren’t unexpected. I’m just noticing that it seems to be getting a bit worse than it used to be during earlier tests. Which of course can be explained by the increase in pace with faster development, more changes and larger patches. The last two updates both brought more changes - and more problems.

I was always sceptical about the time plans for Early Access and Release - because even with a great team and a great framework, you can’t rush certain things, or you just start making sloppy mistakes. And right now, things certainly feel a little rushed. And if that’s the current plan - to keep to that “EA launch in summer” plan, by rushing it… then I will probably take a little break when the launch happens, and only return after the worst has blown over. Just for the sake of my own sanity.

With all that out of the way, let’s move on to the good stuff.

The Item Revamp

The Terraformer, in it’s previous iteration, was quite overpowered. So extremely useful actually, that when it came to balancing the starter packages, I thought it was worth at least as much as three of the other tools. And even then, players, when given a choice, would probably often still prefer getting a single terraformer over getting three other tools.

At the same time, the terraformer was just completely overloaded with functionality. Hitting “TAB” to switch modes was basically exactly the same thing as hitting a number key to switch to another tool. It did effectively do exactly the same thing. And it wasn’t even that all the available actions like digging, depositing, heating, lithifying, eroding, assaying, prospecting and dumping could be tabbed through - no, each mode came with multiple actions, which again often behaved differently to each other. The Instaformer even had another sub-mode on Q (place/delete) - that interfered with the usual Q-E forwards/backwards cycling through materials.

I had riled against these tab-cycled tool modes in the past, because just they seemed unneccesarily complex and confusing. And confusing they were - as became regularly evident by videos of new players, questions they asked in text or voice chat, and feedback they gave after first playing. Getting used to the Terraformer and it’s modes took a bit of time and effort.

Still in love with those trees!

The new system is SOOO much better in that regard - it’s an absolutely massive improvement. Now, every tool is just one thing. You switch between them with the number keys. TAB (when available) always opens a material-picker (which is so much more convenient than cycling through long lists of materials). Q and E (when available) always are alternate actions to the left-mouse-button’s primary action.

It’s so much cleaner, more consistent and more comfortable to use, that I honestly don’t even mind the downside of the terraformer now needing three of my precious toolbelt slots, instead of just one. I always carry 30+ camps with me anyways - and can just drop one down, if I need to swap out tools (or want to save my inventory). I love it!

Paint it in oil, hang it in a museum as abstract art - nobody would notice

If we now could only get rid of this needless "tool is active but holstered, highlighted but you can't use it, displays it functions and keys but they don't work" state as well… (instead having the tool either out and usable - or no tool selected at all - but not: "selected but unusable and UI basically lies to you about that"). That would be like a second dream come true!

Item Stats: PQRV - Potential, Quality, Resilience, Versatility

And now for the topic I’ve been itching to talk about the whole time: Item stats. That’s a really big one - for item crafting, the economy and the entire resource system.

All materials come with four different attributes (Potential, Quality, Resilience, Versatility) which will affect different result statistics, depending on what item you are crafting. On weapons this could be things like damage and fire rate, on tools it could be things like increased yield and mining/harvesting speed, on shields things like damage absorption and recharge speed. It is not the case yet - but in the future the values for these material attributes will be different from planet to planet. To players of Star Wars Galaxies (SWG) this will feel quite familiar, as materials in that game had similar attributes with similar crafting effects as well. The big difference being, that in Stars Reach, new materials with new attributes are not going to spawn in on regular intervals. Instead, planets will only receive their resources once - when they are being generated - and in order to find new materials, new planets have to be discovered. That’s at least how it works for non-renewable materials like ores and minerals. Trees and plants on the other hand, can be farmed and regrown. Unlike ores and minerals those will not inevitably run out - although players can theoretically cause their extinction through over-harvesting.

This is what I wrote about the item stats back in May, in PAT15. Each of these stats can theoretically range from 0 to 999 (not 9999 like in SWG - that was just overkill and caused database bloat) - but in practice some materials might have their own custom min and max values (but never exceeding that 0 - 999 range). When an item is crafted, these numbers than translated into specific bonuses for each item:

Item	Potential	Quality	Resilience	Versatility	unassigned
Terraformer	+Scan Result	+Excavation Magnitude	+Battery Regen	+Max Battery
Companion Orbs	+Node Ring Count	+Boomerang Shot Count	+Battery Regen	+Max Battery
Fortress Shield	+Capacitor Size	-	+Shield Duration	-
Gravity Gun	+Primary Beam Knockback	+Boomer Radius	+Black Hole Radius	+Max Battery + Battery Regen
Harvester	-	-	+Battery Regen	+Max Battery	Planting Battery Drain
Healix	+Surge Radius	+Mass Heal Radius	+Battery Regen	+Max Battery
Omniblaster	+Homing Radius	+Turning Radius	+Battery Regen	+Max Battery
Paver	-	-	+Battery Regen	+Max Battery
Reflection Shield	+Capacitor Size	-	+Shield Duration	-
Trailblazer	-	-	+Battery Regen	+Max Battery
Xyloslicer	-	+Chopping Battery Drain	+Planting Battery Drain	-	Battery Regen
Pruning Shears	+Pruning Speed	+Growth Speed	+Battery Regen	+Max Battery
Depositor	-	+Deposit Efficiency	+Battery Regen	+Max Battery
Chronophaser	-	-	+Battery Regen	+Max Battery

All of this stuff is preliminary and just a first pass - so I definitely would expect to see a lot of changes happening here over time.

To explain how it works, let’s craft a Terraformer as an example:

The Terraformer needs 3 materials

The crafted item’s stats are calculated form all materials used in the recipe. For each stat individually, the lowest number is dropped - and the average of the remaining numbers is taken:

Slot	Ingredient	P	Q	R	V
Common Metal	Manganese	434	735	470	644
Reactive Gas	Fluorine	310	466	721	600
Uncommon Metal	Tin	333	751	603	359
Expected Result	Average best 2	383.5	743	662	622

The Result

And that means, our brand new Terraformer gets these bonuses:

The Item Stats

• Scan Result +19.2 (P)
• Excavation Magnitude +152.3 (Q)
• Battery Regen +66.2 (R)
• Max Battery +746.4 (V)

But what does all that mean?

No idea about Scan Result, but it has something to do with prospecting. Maybe the amount of cubes you see, or something. But I do have an idea what the other three do.

If you dig into a single material, you will always dig out exactly the same volume (create the same size of hole) and get exactly the same amount of loot, until your hopper is full. And for regular materials like Granite, Marble and Breccia - that amount you get is exactly 100 units. (For all skills in the Mineralogy tree maxed out).

So, you always remove the same amount of ground, get the same amount of material and the same amount of tailings. Then you empty the hopper and repeat. The only thing that changes, is how long that takes.

Digging Granite with 4 different Terraformers - take note the 766 Total!

Excavation Magnitude increases the amount of ground you remove each tick and the amount of material you get each tick. (Each time you “shovel” and loot flies towards you.) It also increases the radius of the tunnel you create - which conversely means the length of tunnel created per hopper does get less. Overall you do create more tunnel in less time though - because you dig faster - i.e. it takes less time to fill up one hopper and get those 100 materials.

Battery Regen decreases the amount of time needed to recharge your battery. That means shorter breaks in between bursts of digging. That also means that overall you dig faster - taking less time to fill up one hopper and get those 100 materials.

Max Battery increases how much charge your tool can hold. It means you can dig longer before needing a break - but it also means longer breaks, as there is more to recharge. In fact these two cancel each other out, and thus Max Battery would have no effect on overall dig speed. But it does, because there’s a two second delay every time before regen starts, and thus having to regen less often, means fewer of those 2 second waits.

Overall, Excavation Magnitude has the greatest effect. If you really want to dig the most, in the shortest amount of time, that’s the stat you want to prioritize getting as high as you can. Battery Regen takes a good second place. But the effect of Max Battery is so small, that it’s barely even noticable over the lag-caused run-to-run variation.

Digging Marble with 4 different Terraformers

The other thing is, that if you are serious about mining - you should absolutely equip two Terraformers. Then you need only halfways decent Regen, and one Terraformer will easily charge back up again, while you empty the other - thus allowing you to mine without any breaks, and giving you twice the hopper capacity. And that makes Excavation Magnitude the clearly best stat to raise on this tool.

Digging Breccia with 4 different Terraformers

Overall the tools seem to have gotten nerfed somewhat. This was to be expected, since everything we had so far, was quite overpowered for testing purposes. When the game releases, you probably shouldn’t be able to max a character in just a few hours - therefore things have to get slower than they are now.

I think the tools we had in previous tests were in some cases more substantially differnt though, than just being what the new tools would be, if they had all stats at 999. In the case of the Terraformer for example, that previously could hold a LOT more tailings - and hopper size is not a thing you change through PQRV.

Digging even more Granite with even more Terraformers - now at over 5000 granite

I’m fine with the new Terraformer though. I’m used to switching between two of them now, and that works pretty fine - and on the wild planet, when it comes to dumping the hopper, it’s actually nice that there isn’t that much stuff in there - even when having to run to the dump more often can be annoying.

The new Harvester is something I struggle with more - because unlike with the Terraformer, you can no longer use the Harvester by just holding the mouse down and waving it around. That drains your battery within just a few seconds. And it then takes longer to charge back up, than draining it again takes. The reason for that is, that the Harvester looses a lot of energy even when it doesn’t actually harvest anything. The Terraformer will just dig on every tick - but the harvester will more often fail because the targeted plant is already done harvesting, or there’s nothing in range, or it measueres soil pH instead - using the full amount of energy each time. The harvester is now really only usable as a single-click tool, not as a click-and-hold tool.

A Closer Look At The Numbers

If you’ve read my post on table-top rpg combat, you will be already aware that I love staring at numbers. If not, better brace yourself, because I’ve got some numbers to stare at!

The Terraformers I have crafted, and the bonuses they got from their PQRV stats:

P	Scan Result	Q	Excavation Magnitude	R	Battery Regen	V	Max Battery
250	+12.5	250	+51.3	250	+25	250	+300
650	+32.5	536	+109.9	705	+70.5	533	+639.5
498	+24.9	663	+135.9	743	+74.3	460	+552
760	+38	371	+76.1	719	+71.9	502	+602.4
555	+27.8	726	+148.8	531	+53.1	602	+722.4
377	+18.9	716	+146.8	755	+75.5	607	+728.4
522	+26.1	686	+140.6	779	+77.9	485	+582
502	+25.1	759	+155.6	483	+48.3	591	+709.2
612	+30.6	743	+152.3	649	+64.9	502	+602.4
554	+27.7	420	+86.1	766	+76.6	588	+705.6
733	+36.6	650	+133.3	484	+48.4	635	+762
769	+38.5	672	+137.8	605	+60.5	221	+265.2
384	+19.2	743	+152.3	662	+66.2	622	+746.4
706	+35.3	594	+121.8	525	+52.5	709	+850.8
475	+23.8	743	+152.3	537	+53.7	502	+602.4

Thanks to sgisk for the Terraformer with V221. Because of the average only taking the highest numbers, it’s not easy got such low results. They are helpful though for the statistic.

So, what we can clearly see at first glance, is that Battery Regen equals R*0.1 or R/10 if you will. Scan Result seems to be P*0.05 or P/20 - but there’s a rounding error: 733 is 36.65 which should be rounded to 36.7 but shows as 36.6 for some reason (floats? these really should be integers, just multiplied to the needed level of precision).

Similarly Max Battery Regen seems to be V*1.2. And Excavation Magnitude seems to be Q*0.205. Why that weird number? I really don’t know…

Of note here is, that everything is perfectly linear - just simple multiplications. The exact values to be multiplied with are probably subject to change and will get tweaked again and again, over time.

The experiments I did to figure out the actual effects these number have, involved me taking a Terraformer, digging into pure material until the battery is empty. Waiting until the battery is full again, and then repeating the process until the hopper is full. There is of course some variation to the results, due to lag and human reaction times being not the greatest.

Digging into Granite

P	Q	R	V	Time needed until Hopper full
250	250	250	250	04:56
250	250	250	250	04:47
250	250	250	250	05:02
250	250	250	250	04:31
377	716	755	607	02:16
377	716	755	607	02:16
498	663	743	460	02:07
498	663	743	460	02:11
502	759	483	591	02:14
502	759	483	591	02:27
518	634	625	684	02:26
522	686	779	485	02:32
522	686	779	485	02:42
555	726	531	602	01:59
555	726	531	602	02:02
612	743	649	502	02:19
612	743	649	502	02:12
650	536	705	533	02:31
706	594	525	709	02:44
706	594	525	709	02:40
733	548	648	436	02:35
760	371	719	502	03:17
760	371	719	502	03:18
769	672	605	221	02:38
769	672	605	221	02:20

Digging into Marble

P	Q	R	V	Time needed until Hopper full
250	250	250	250	02:54
706	594	525	709	01:34
522	686	779	485	01:02
502	759	483	591	01:02

Digging into Breccia

P	Q	R	V	Time needed until Hopper full
250	250	250	250	02:39
706	594	525	709	01:17
522	686	779	485	01:19
502	759	483	591	01:09

We can clearly see that Q has the most impact on the dig speed, followed by R. We can also see, that lag can cause quite some variation in times - and that that variation increases the smaller the battery (i.e. the more breaks have to be made - which makes sense, as every single break the lag can add a little extra delay). Thus bigger battery seems slightly less affected by lag variation.

Also of interest might be, that granite is a hard material and thus rather slow to dig. Marble and Breccia are more on the average side, and sand and dirt are a lot softer and will be even faster than that. And while using the Terraformer on Granite felt okay to me - granite takes twice as long to fill the hopper as Marble and Breccia do. I haven’t tested soils yet, but if those go down to like 30s for one hopper-filling, it really might feel like you have to empy that hopper excessively often. Also - with numbers that small, the differences caused by the PQRV values become less noticeable, and potentially smaller than the difference lag can cause. PQRV would really need to show a bigger effect.

The Stripped Terraformer with the 250-250-250-250 values feels really bad to use. Like, I don’t think you can cut that back any further if you don’t want to risk scaring away newcomers right out of the gate. I haven’t even tried a 250 Harvester - that must be real pain to use. Despite starting that low - PQRV is still really, really fine-grained. I doubt any player could reliably tell in a blind test, which of two Harvesters is 700-700-700-700 and which is 710-710-710-710 - if you don’t hand them a stop-watch. Add a little lag, and they probably couldn’t tell, even when using a stop-watch.

Makes me question if 0-999 is even needed, or if 0-99 would work just as well? I mean… if the players can’t tell the difference anyways… (yeah I know, it’s more about the placebo effect of seeing those bigger numbers - allows for more competition between crafters - no actual difference in performance needed).

But between the PQRV on tools, the unlockable skills, eventual buffs and consumables, and group play and other bonuses - it’s gonna be really hard to give all those things a noticable effect - without making a player who has it all, be massively more powerful than a newbie with nothing.

See you next time!