Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Sherlockkat]

@biomatter:

Also, what happens when you fail to kill the Data Miner? I had read a little about him in this thread and so I had a go at him, and my Ripper downed him in one hit. After escaping past those damnable Enhanced Grunts, I took a breather. My mind will be blown if I really just got the location of every staircase henceforth.

If you fail to kill the data-miner, he runs down a escape hatch and sets his grunts on you anyway. You don't get the location of every staircase. You only get the location of every machine. But, in access, I think the entire exit setup with the shell and sealed door counts as a machine or something. Because, it shows up in the map. I also saw the a terminal + door setup show in the factory. That's the extension. So, I think you download terminal locations and the doors associated with those terminals. I dunno.

You mentioned that Lower Caves connect to both the lower Factory floors. Does this mean I can hit, say, Zion from Factory -7, and then Data Miner from -6? And that in a future run, it might be reversed?

Yup. That's exactly what I meant. K mentioned it somewhere in this thread.

Re Research branches and Armory: I have seen locked stairs to Testing and Section 7. Section 7 does not show up when you access(branch) according to Zhirov. I have never seen Quarantine. Armory is a locked branch in factory. It is behind a double door and is guarded by a bunch of sentries.

Re Perun, Svarog and Zhirov: I will swing at them and let you know what happens. Though, as you mentioned, bringing a scan processor along might be the smarter. But, who's got time for that? This might not be until Alpha 9 though.

Re Archives: Something is going on there. The terminals are always locked out when I show up. I think there is some plot information that I am missing and K did mention that very few ppl have seen everything that happens that way. Hub_04(d) is interesting. But, I don't think there is any lore there. But you can do/see some pretty cool stuff there.

Looking forward to your interpretation of the lore . Edit: Actually, I just checked out the wiki and was going through the Zhirov stuff and have started to make some more connections. Especially the bit about Sigix taking the fight below ground and certain "Derelict" adapting alien technology. I believe those self proclaimed derelicts are tinkerers. Wonder whether warlord is being used by the any surviving Sigix in order to overthrow MAIN.C. Man, even the data miner could be in cahoots with the Sigix.

@zxc: Nice win. Where did you get that system-mapper from? Wastes? Man, I should really start getting 3 power slots too. I always go with two power slots and go really greedy on utilities. The result : Late into the game, my build is just this dynamic puzzle and I need to constantly figure out how to keep everything powered.

The adv signal interpeter got nerfed. You need the exp signal interpreter to see the late game hunters. It's the reason I find myself relying on optical arrays more and more.

[Kyzrati]

The adv signal interpeter got nerfed. You need the exp signal interpreter to see the late game hunters. It's the reason I find myself relying on optical arrays more and more.

It wasn't technically a nerf--all the Hunters just got a boost to their sensor cloaking ability, and I added the Experimental version of the interpreter as a way to still detect the late-game Hunters. This was to put all of the Hunters slightly ahead of the interpreter curve, and more importantly make late-game Hunters require a more rare interpreter, one that you can't get from Watchers. The aim was to make the late game a little tougher for speed runners who can otherwise avoid most everything by simply acquiring interpreters from Watchers. Now you can't be so sure

Oh, and I was uploading the scores today and noticed zxc was back for a win; nice job

[zxc]

@zxc: Nice win. Where did you get that system-mapper from? Wastes? Man, I should really start getting 3 power slots too. I always go with two power slots and go really greedy on utilities. The result : Late into the game, my build is just this dynamic puzzle and I need to constantly figure out how to keep everything powered.

3rd power slot wasn't actually being used at the end, but I got it to be on the safe side as I've learnt not to go all-out on utilities these days. My extra slots for propulsion came in very useful (I wanted only 4 for Imp Flight Units). Towards the end I was just having issues with heat as I never seem to find heat sinks or coolers after early factory...

The adv signal interpeter got nerfed. You need the exp signal interpreter to see the late game hunters. It's the reason I find myself relying on optical arrays more and more.

It wasn't technically a nerf--all the Hunters just got a boost to their sensor cloaking ability, and I added the Experimental version of the interpreter as a way to still detect the late-game Hunters. This was to put all of the Hunters slightly ahead of the interpreter curve, and more importantly make late-game Hunters require a more rare interpreter, one that you can't get from Watchers. The aim was to make the late game a little tougher for speed runners who can otherwise avoid most everything by simply acquiring interpreters from Watchers. Now you can't be so sure

Oh, and I was uploading the scores today and noticed zxc was back for a win; nice job

That's too bad about the adv signal interpreter, but it's probably justified. Still, sensors all the way!

Thanks folks.

[zxc]

Minor requests and points including some that almost certainly aren't worth K's time but I'll list them in case they offer ideas.

* One-time hacks on terminals like Locate Traps should gray out if you complete the hack, exit the terminal and re-open it (likewise if you take an operator's key and discover where the traps are).
* After discovering a secret hack through lore like trojans, maybe list that hack in all terminals with that player profile afterwards.
* Operators can be abused quite thoroughly (see https://youtu.be/SzLc5l15RpA?t=719) but maybe this is OK as it's tactical and not always possible.
* Running using shift probably shouldn't keep going through doors (and ideally should stop at 45 degrees to doorways, machines and wall edges to allow for diagonal movement, and also should stop at items if it doesn't already).
* I can't help but think that there ought to be another way to get to exits with sentries camping on them without fighting them and without doing a weird dance to get them to notice you and walk off the exit.
* The best flight units are still an early-game prototype!
* I swear heat sinks and coolers don't spawn often enough late-game, but it could be the result of a really small sample size.
* I found a weird sentry which was camping beside an exit instead of on top of it (see https://youtu.be/SzLc5l15RpA?t=6501).
* Still getting a noticeable pause when making my first move on a new floor in Factory onwards.
* Recycling machine hacks seem overly difficult for the reward.

[Kyzrati]

Thanks for the feedback! Having a list of little stuff to do is good for those extra days shortly before release when I'm mostly in the testing phase so no big features can be added. But the Little Stuff list is already quite long

Here we go

* After discovering a secret hack through lore like trojans, maybe list that hack in all terminals with that player profile afterwards.

This is planned, though not in this way. It will be a separate console you open. It's just low priority right now because the current system works and this will be a lot of work.

* Operators can be abused quite thoroughly (see https://youtu.be/SzLc5l15RpA?t=719) but maybe this is OK as it's tactical and not always possible.

Yeah, I wouldn't worry about that yet.

* Running using shift probably shouldn't keep going through doors (and ideally should stop at 45 degrees to doorways, machines and wall edges to allow for diagonal movement, and also should stop at items if it doesn't already).

Run movement isn't fully implemented yet, though it got pushed down rather far on the list after the initial implementation. Thanks for bringing it up, since I hadn't yet collected any opinions on what players want it to do (all it said in my notes is to collect opinions ).

* I can't help but think that there ought to be another way to get to exits with sentries camping on them without fighting them and without doing a weird dance to get them to notice you and walk off the exit.

I can think of about 4-5 other methods that already exist

* The best flight units are still an early-game prototype!

Huh, we'll definitely be fixing that before long then, especially now that fabrication is coming up! It's tough to balance them into the movement system with everything else--not enough room for common vs. prototype flexibility with respect to flying :/

* I swear heat sinks and coolers don't spawn often enough late-game, but it could be the result of a really small sample size.

Right. That would be the item distribution trends which involve more and more items as you get closer to the surface, meaning a relative reduction in the chance of finding dissipation-related parts. In the latter half of the game the best way to acquire them is to fight, since as you know most combat robots carry them and tend to drop them by the dozen . But then, as a speed runner you'd be trying to avoid those situations...

According to the algorithms and part counts, the chance of an item being a cooling item by floor is... -5: 1.3%, -4: 0.6%, -3: 0.2%, -2: 0%, -1: 0%.

* I found a weird sentry which was camping beside an exit instead of on top of it (see https://youtu.be/SzLc5l15RpA?t=6501).

Did it worry you? . I can think of one situation where that's possible--sometimes guarding units will permanently relocate themselves if they leave their original position for some reason and come back to find it blocked.

* Still getting a noticeable pause when making my first move on a new floor in Factory onwards.

I noticed that a lot in my most recent runs, too! It was definitely gone for a while after I optimized it, and now it's more or less back to what it was. Damn :/. I'll have to look into that again and see what's up... The last optimization isn't guaranteed to always resolve the issue, but it generally helped a lot more frequently than what we're seeing now.

* Recycling machine hacks seem overly difficult for the reward.

Maybe. The idea there was that only dedicated hacking builds should be able to benefit from them, and a 30% hacking bonus at a tier 1 Recycling Unit pretty much guarantees that you'll be able to get all the parts inside (50% chance per hack), which isn't too bad. But seeing as you can't control what they are, and they're likely messed up anyway, yeah I'll up the base chances.

[zxc]

You skipped my first point. Also, it would be cool to have an X mode (like DCSS) which allows panning around the map with the keyboard without having to hold Alt. This would be faster (you could allow shift + direction to skip many squares) and safer for Windows users (avoids that awkward Alt problem after using alt-tab). On the same train of thought, a mini-map could be cool. OK OK I'll stop now

[Kyzrati]

I skipped your first point because it felt like the most minor thing on the list, which incidentally is already planned. It's one of those tiny QoL features that requires a disproportionately large amount of work to full implement and test.

I'll see about adding another map panning mode (we already have three panning methods...), though Alt-Shift already does that, and adding a new method wouldn't solve the Windows Alt problem in all its other forms, so it's still always going to be safest to hit Alt and or Alt-Shift when Alt-Tabbing back to be safe anyway.

A minimap I was thinking about just yesterday, because I'd love to have it if possible, but the only implementation possibilities for that given the technical limitations of the display aren't going to be useful in all cases due to the large size of some maps. I might try to play with the idea some more, but the layout wouldn't be able to show more than a 100x100 area at a time (so only a quarter of a Factory), and there would be zero detail whatsoever--pure layout info in a solid color. Not that great...

[zxc]

I skipped your first point because it felt like the most minor thing on the list, which incidentally is already planned. It's one of those tiny QoL features that requires a disproportionately large amount of work to full implement and test.

It was the only one missing so it looked like you overlooked it . Indeed a lot of these are tiny yet require disproportionate work.

I'll see about adding another map panning mode (we already have three panning methods...), though Alt-Shift already does that, and adding a new method wouldn't solve the Windows Alt problem in all its other forms, so it's still always going to be safest to hit Alt and or Alt-Shift when Alt-Tabbing back to be safe anyway.

Yeah, I tap alt a bunch of times before using alt to look around, pretty much regardless of whether or not I alt-tabbed recently. I hadn't tried alt-shift-direction because I have learnt not to fool around holding multiple keys, especially since the times when I pan around the map are usually when I'm in major trouble and a wrong move can mean a lot of harm done.

A minimap I was thinking about just yesterday, because I'd love to have it if possible, but the only implementation possibilities for that given the technical limitations of the display aren't going to be useful in all cases due to the large size of some maps. I might try to play with the idea some more, but the layout wouldn't be able to show more than a 100x100 area at a time (so only a quarter of a Factory), and there would be zero detail whatsoever--pure layout info in a solid color. Not that great...

This was half a joke. I'm surprised you've even considered this! That said, a map of the current level would really give a sense of the massive scale of some of the levels. The sprawling levels are one of my favourite aspects of the game.

[Kyzrati]

Yeah, I tap alt a bunch of times before using alt to look around, pretty much regardless of whether or not I alt-tabbed recently. I hadn't tried alt-shift-direction because I have learnt not to fool around holding multiple keys, especially since the times when I pan around the map are usually when I'm in major trouble and a wrong move can mean a lot of harm done.

Oh you haven't been using Alt-Shift direction? I always play via pure keyboard and use that a lot. Easy to jump around the map. There is also a new undocumented command, '`', which colors in the entire map area you've discovered, even if outside your FOV. You might find that useful.

This was half a joke. I'm surprised you've even considered this! That said, a map of the current level would really give a sense of the massive scale of some of the levels. The sprawling levels are one of my favourite aspects of the game.

Hehe, I know you were joking, but I was seriously thinking about it just yesterday so I thought I'd throw my thoughts out there, too! I might do it as an experiment for the heck of it. Half the reason I want to do it is because it could appear with a cool animation

[biomatter]

Speaking of animations, the Data Miner level is beautiful. It's so awesome seeing that subtle pulse. Fluff-wise, you could have used that a million other places - power conduits, other data links, etc - but you didn't, and that makes it all the more memorable when you first see it. Less is more! It's so good

[Kyzrati]

There's code to use that effect anywhere, but it would also be distracting if found on a wider scale. It's nicer to give some areas their own unique feel, however possible with ASCII, and the Data Miner is a little place that is otherwise quiet and empty--perfect opportunity for something like that. It took a long time to get that whole conduit animating just right, though

It takes a lot more time of course, but I do like to go for the "awe players with something found nowhere else" approach whenever I can do it lore-wise (and technically). I had that feeling when creating Zion, for obvious reasons

* The best flight units are still an early-game prototype!

Huh, we'll definitely be fixing that before long then, especially now that fabrication is coming up! It's tough to balance them into the movement system with everything else--not enough room for common vs. prototype flexibility with respect to flying :/

So going back to this, I couldn't help but look at it now, and I'm not sure how the Imp. Flight Unit got that support value. It's pretty out of whack with everything, actually, at nearly double support for no extra costs. Oops. I'll go back through all the values and make some adjustments, but that one is definitely getting nerfed.

Edit: Hm, okay I take that back. I can't quite decide how to treat it because unlike other parts, what I did with flight is make it so that it's actually a fully two-tiered system: prototypes vs. common types, where high-rated forms of the latter can't stand up to even low-rated forms of the former in terms of efficiency. (Normal prototypes only average about rating+1 good.)

So there are better prototype flight propulsion outright better than what you were using, you just didn't find them. I did this because you can't get them off enemies, making them difficult to acquire, but with enough of them you could theoretically build a combat-capable flight build. Maybe I should give that up... (Also, that said some higher-tier prototypes do have the problem of fast movement = significant heat and energy cost increases even with a +1 increase on either, so they may be better in some ways, but are not always the most efficient if you want to run extremely light.)

Edit 2: Okay, I just made some rather significant stat changes to flight (and to a lesser extent hover). I'll be interested to see how things have changed for speed runs with Alpha 9!

[zxc]

It's definitely complicated and I certainly haven't looked deeply at the stats of the various flight units available. Late in the game I don't even equip new prototype flight units because I feel like I already have a game winning build and thus it would be taking a risk for very little gain. I'd love to try out all the prototypes but unless I have one of those special items that identifies them, it just isn't worth it unless I'm in a very poor position.

The main reason why I think the Imp Flight Units are the best (certainly the best of the flight units I've seen and used) is because of their energy/heat efficiency for their speed. I pay less attention to support capacity, and my last build would've functioned well enough with half the capacity it had.

[Kyzrati]

I noticed you were handling low support impressively well in your video. You won't mind that I've reduced all prototype flight mass support, then .

Probably the most meaningful change I made though is to reduce the speed modifier from -5 to -3, meaning you'll need more propulsion slots to reach the same speeds.

Also, to resolve the efficiency problem, which I found was otherwise insurmountable given the rough granularity of the system at even the reduced speeds, I completely set all heat and movement costs to 1!

This is a sudden collection of rather big changes, so more tweaks could follow, but let's see how this plays first. If together these changes can solve the original issue (which was kinda ridiculous) while still requiring skilled heat and energy management and keep speed builds fun and challenging, that would be, um, nice.

[zxc]

Speed builds are most fun when you don't need to worry about heat, energy and mass support

Those are some big changes. No idea what effect they'll have. I've been telling you about flight propulsion efficiency for so long now and so you go ahead with such a hasty balance change 

[Kyzrati]

You single-handedly derailed my evening work schedule today (Sure I was going to be working on Cogmind, as usual, but not this...)

[zxc]

Goddamn... sorry.

My intention was just to offer some thoughts and get some discussion going. Flight propulsion seems to be very delicate to balance.

[Decker]

FWIW I did run a combat build for a while using Imp VTOL (IIRC). The low support penalty means you can stack it 3x and still come up ahead of hover, legs and treads. Of course they have crappy integrity, limiting their interest for combat.

I also think that late-game flight units are crap, with their high heat generation.

Heat and energy requirements balloon up quickly due to the double effect of faster movement and increased resource requirements per tile. In effect, it's impossible to lift a high mass because stacking flight units doesn't scale linearly with respect to heat and power. IMO it should. If the player wants to dedicate so many slots to propulsion, power and heat sinks, why not allow it? I would favor a per tile requirement of 0 balanced with increased base costs.

While we're at it, I would also remove the quantum effect of the overweight penalty. Your speed should be just a linear function of (total propulsion)/weight.

Sorry for your work schedule

[Kyzrati]

Goddamn... sorry.

My intention was just to offer some thoughts and get some discussion going. Flight propulsion seems to be very delicate to balance.

Haha, nah, it's fine. Obviously still my choice to actually go through with it!

I also think that late-game flight units are crap, with their high heat generation.

Heat and energy requirements balloon up quickly due to the double effect of faster movement and increased resource requirements per tile. In effect, it's impossible to lift a high mass because stacking flight units doesn't scale linearly with respect to heat and power.

Well now it does, so it's no longer an issue . And you will now pretty much always want to upgrade flight units when you can. If you need the support, anyway, because obviously the experts don't need quite so much, BUT you'll probably be wanting more either way due to the speed reduction.

I also nerfed mass support utilities.

IMO it should. If the player wants to dedicate so many slots to propulsion, power and heat sinks, why not allow it? I would favor a per tile requirement of 0 balanced with increased base costs.

Increased per-turn costs would remove the challenge of getting enough energy and dissipation to move as much as you want to. I like that the cost varies directly proportionally to the amount you move. Every step means something.

While we're at it, I would also remove the quantum effect of the overweight penalty. Your speed should be just a linear function of (total propulsion)/weight.

Having a binary effect is clearer and also makes some sense as these things are more sensitive to weight restrictions and quickly become inefficient once you pass the threshold they're designed for.

[zxc]

I also nerfed mass support utilities.

Wish I never said anything

[Decker]

Well now it does, so it's no longer an issue.

The balloon effect is still present, just less pronounced (you need either zero per tile requirements or zero speed increments to avoid it entirely). It does make the game more challenging though, so it's fine. I'm eager to test the changes!

I'm not a fan of the binary overweight threshold because it's annoying to optimize for and it doesn't really make sense. Add 1 more mass => no effect in general, or dramatic slow down. You could keep the system as it is, but scale the slowdown to the amount of overweight. E.g. if you're 20% overweight you get 20% of the overweight penalty.

[Kyzrati]

I also nerfed mass support utilities.

Wish I never said anything

Don't worry, they're still usable, and I mostly nerfed the high-level ones, which were kinda ridiculously good, yeah? The entry-level one is unchanged, and Adv. went from 10->8. Some of the best could originally give you 40-50 support at a modest energy cost.

They still do their job of basically using energy to turn a utility slot into propulsion (albeit without a speed bonus), and cost a good bit less energy than the power source required to generate it which they themselves can support.

Depending on how the propulsion changes turn out, maybe this could be readjusted. In fact, I may as well go over it here to ask for input from anyone who wants to examine the numbers. Weight redistribution components:

Rating /	Upkeep /	Support (Old) /	Support (New)
2	4	5	5
3	6	10	8
5	8	20	12
7	10	30	16
8	12	40	20
9	15	50	25

Well now it does, so it's no longer an issue.

The balloon effect is still present, just less pronounced (you need either zero per tile requirements or zero speed increments to avoid it entirely).

Ah, I wasn't sure what you were referring to, more movement causing the effect, or what zxc and I were talking about (the thing I was trying to fix): high-tier flight costs rapidly expanding due to a single increment in the per-move cost, which is no longer an issue.

A system where costs are derived entirely from upkeep wouldn't be fun, because chunks of heat and energy cost would suddenly be applied at turn intervals, and you'd start trying to game when you could take your "free" moves, and when those costs would come into play.  Then we'd need some kind of countdown for it, and it would become a focus of the game for some players. It would also mean you'd have a much harder time moving when low on power for whatever reason, or you'd at least be stuck in a single position for longer.

I'm not a fan of the binary overweight threshold because it's annoying to optimize for and it doesn't really make sense. Add 1 more mass => no effect in general, or dramatic slow down. You could keep the system as it is, but scale the slowdown to the amount of overweight. E.g. if you're 20% overweight you get 20% of the overweight penalty.

From a gameplay standpoint, a system with concrete thresholds leads to more interesting decisions, because it makes the results of those decisions much more extreme here, in an area where a linear model would add to the uncertainty of the true long-term impact of carrying a little extra weight.

Probably more importantly, it makes it much clearer when one form of propulsion is no longer ideal for what you're aiming to do.

I do wonder how a linear model would impact the game in specific ways, but it isn't something to spend time on now.

[zxc]

FWIW I'm fine with the way overweight works currently. It's very punishing to go over the support limit even slightly with flight, but it doesn't prevent moving outright. Going overweight generally happens when propulsion gets destroyed, which is always very harsh with a flight build. Maybe it's overcomplicated as a mechanic but the end result is you should stay beneath your support limit with the lighter propulsion, which is simple enough of a rule to follow.

The new weight redistribution values seem OK. The really high rating ones were insane, but also really rare. It was always such a treat to find one. I guess it will still be a treat with these values too. You've probably seen one or two of my morgues with far more mass support than I needed (maybe over 100 with flight?).

[biomatter]

Just chiming in that I am also ok with how the weight system works currently. I've been running mostly with legs and treads lately, and it's very easy to tell when I'm underweight, overweight x1, x2, etc, and what the penalty for each level is.

[Kyzrati]

Thanks for the feedback guys.

You've probably seen one or two of my morgues with far more mass support than I needed (maybe over 100 with flight?).

Yeah those support utils are crazy good . And really the essential parts for a speed run are mostly going to be lightweight.

The idea was that if you managed to find one of the good support utils you could theoretically be a combat flier, but I got worried about them while working on this latest flight update, since we have the upcoming fabrication changes and you might be able to easily build one. Maybe I overreacted a bit and a better solution would be to rebuff them and just make them even more rare. (Their distribution setting before was just "uncommon.")

Depends on what they're for, really--to improve a flight build, or is the goal to enable flying death machines, and would anyone even want to do that, anyway...

[Kyzrati]

* One-time hacks on terminals like Locate Traps should gray out if you complete the hack, exit the terminal and re-open it (likewise if you take an operator's key and discover where the traps are).

I've just add a feature that remembers all previously hacked targets on a given machine and grays them out if you reconnect and it makes no sense for those to be hacked again (about a third of them).

The Locate Traps / Operator Data change can't be made, however, at least not in blanket fashion, because that's information you're not necessarily aware of in all cases--a nearby terminal might have other traps within range that the core does not (because the core technically reveals those near where you acquire it, rather than based out of that Operator's terminal). An alternative approach would be possible, but is rather involved because it requires allowing cores to remember who their previous owner was, and trace that back to the original terminal to search from that point... not really worth it right now for the relatively small benefit.