Factorio Forums

FuryoftheStars wrote: Sat Jun 01, 2019 6:32 amthis would be nothing more than another fuel source for boilers.[*]

* Steam engines/turbines

I think solar boilers would be different because you could make things more complex by having the temperature from solar boiler steam change over the course of the day, which would make for some very interesting new mechanics for steam distribution and energy balancing. If the temperature peaked at 200℃ or so, for instance, you'd have to either use turbines (expensive!) or include mixing buffers in your design to avoid energy wastage, since steam engines can't utilize energy above 165℃.

Theikkru wrote: Fri May 31, 2019 7:25 pm I think I should be allowed to take exception when said quirks serially demean other people's ideas and methods with unsupported hyperbole or double standards, BUT

No-one is exempt from the powers of Koub the mighty. Don't try him. And are you sure you should be the one pointing fingers here?

Unlike you, I point out the exact segment I take issue with. You and your friend fury are the ones who responds to a well-founded argument with variants of "No u":

Theikkru wrote: Fri May 31, 2019 1:53 pmSo you DO want everything served up on a platter then!

You claimed that automated space acquisition with recursive blueprints was trivial with artillery but you lash out when I ask for your design because you don't have it?

Theikkru wrote: Fri May 31, 2019 1:53 pm Then I believe I can conclude that you've never learned how logic works, since conditional statements say nothing about what is necessary to meet the condition, nor what happens if the condition is false.

Ad hominem. Why don't you tell us about your advanced mathematical education, since you look down on mine? And you didn't make a conditional statement:

Theikkru wrote: Thu May 30, 2019 2:06 pm With solar, the ONLY requirement is that you find enough space to throw down a blueprint.

Who said something about unsupported hyperbole again?

Theikkru wrote: Fri May 31, 2019 7:25 pm back on topic, as you say, I'd like to reiterate that at no point did I argue that nuclear somehow occupies more space, or requires more time, or is more of a chore.

Which no-one accused you of saying. That's a strawman defence. What I actually said:

Qon wrote: Fri May 31, 2019 9:12 am Are you seriously saying that you need to find space for nuclear so it's the same as solar, when nuclear gives orders of magnitudes more power/tile?

Stop twisting things.

You said:

Theikkru wrote: Fri May 31, 2019 1:36 am because even if you have a ready-made print on-hand, you either need to find a body of water big enough to accommodate it, or figure out how to run custom piping to it, which takes (you guessed it!) time and exploration, not to mention setting up a fuel line.

Theikkru wrote: Fri May 31, 2019 7:25 pmOn the other hand, nuclear power, which must also consider space and deployment requirements

Here you clearly state that nuclear requires time and exploration. And you say that it matters, otherwise there's no point in bringing it up. You are trying to equivalate it to solar by pointing out that nuclear "requires space too". Just because it isn't 0 space doesn't mean that it actually requires Qonsideration when the difference is a factor of orders of magnitude apart.

Theikkru wrote: Fri May 31, 2019 7:25 pm The reason I see weather or other unreliability in solar power as a potential solution is not because I think it will add more repetitive tasks, or make solar nonviable as a power source (because neither of those would involve more thinking), but because it will encourage players to find intelligent solutions to compensate for that unreliability without just brainlessly deploying 5× or so more accumulators.

Oh, Theikkru has a solution on how to use unreliable solar in a smart way that I could never come up with? And it's not the trivial solution of adding more accumulators? Oh boy, am I excited! Time to learn! Brain set to: Learning mode, maximum information intake possible. Remember all the things!

Theikkru wrote: Fri May 31, 2019 7:25 pm For instance, currently, a lot of people keep their boiler-based power plant when they start building solar, and convert that to backup power once their solar setup is sufficient for their needs. However, as has been pointed out plenty of times, that backup power rarely sees use under normal circumstances, is rarely thought of, almost never expanded, and is often ultimately destroyed as a result. Intermittent (not extended!) unreliability in solar power could make for far more interesting and relevant backup designs that are unique to the situation (since the design requirements for what is effectively a UPS are quite different from regular power supplies), all while avoiding the need to scale solar installations excessively.

Aye aye captain! Writes down note: How to use solar intelligently: Use coal instead!
Wow. That is truly advanced knowledge and deep exploitation of the finer details of solar/accumulator mechanisms.

You know, instead of buffering steam in lots of tanks, you could replace the space used for steam tanks in your backup power with boilers. And then you could run it all the time. Most of the space is used by the steam engines anyways, which you can't get rid in your backup power if you want to keep the same output. And then you could skip building solar completely because why have 1GW of solar and 1GW of steam for your 1GW Qonsuming factory when you can just use steam all the time? Saves a lot of time building solar panels.

"But coal runs out :<"
Research mining productivity and get a faraway coal patch that lasts.

"But coal is polluting :{"
If you care about pollution then you are not using productivity modules so you have slots free for efficiency modules also. You get rid of pollution in machines and reduces electrical use so much that boilers don't really pollute either. And your coal lasts another 5x longer.

You can find coal patches (30M) that give you more than 300 MW for a hundred hours (per patch) just 6k~ tiles from spawn. That is without any productivity research or any modules. If you use speed modules (yes patches last not as long then but still) just the first 4 mining prod researches then you can get 1.2 GW from it. And all this is from a single patch at default settings at a reasonable distance from spawn.

It has been four years at least we discuss this topic, and we have become extremely efficient at this.

Adding randomization is a divisive decision. Some people will like it, some won't. I think (but I might be wrong) that the majority of the player base prefers predictability over unpredictability. If I'm right, adding some way to make the solar panels inadequate would be a bad move from a design standpoint.

I dunno about that. Predictable situations force one kind of solution, while unpredictable situations force a different kind of solution. Solving either type of problem offers a good exercise for players.

Solar power is very annoying to reach GW demand. If solar was both difficult and unpredictable, there'd be very little reason to use them. There needs to be a good payout before throwing uncertainty into the equation.

Qon wrote: Sat Jun 01, 2019 1:40 pm No-one is exempt from the powers of Koub the mighty. Don't try him. And are you sure you should be the one pointing fingers here?

Unlike you, I point out the exact segment I take issue with. You and your friend fury are the ones who responds to a well-founded argument with variants of "No u":

Really? We're doing this? Fine, if you want me to spell things out then, let's have a look at these "well founded arguments" as you put it. Here are my quotes of you from my earlier post:

Qon wrote: Fri May 31, 2019 12:54 amIf you are exhausting ore fields in megabase stage then you are doing it wrong. Resources are effectivly infinite in 0.17

To start, this is incorrect. Ore fields are not infinite, and megabases can run for long periods of time at high consumption rates, so ore fields can and will expire if you have a big enough base and play long enough. You also posit that Fury is doing this, and you are directly calling anyone (including Fury, given your assumption) who does this wrong, which is narrow-minded and insulting.

Qon wrote: Fri May 31, 2019 12:54 am

FuryoftheStars wrote: Fri May 31, 2019 12:03 amAdding an unreliability factor, even if only 10% of the time, would drastically increase the space requirements as a primary power source, while not significantly nerfing their actual power generation.

But there's no point to having a "secondary" system.[...]Your suggestion is to make solar completely useless.

You state here that there is no point to a secondary system, assume (without basis) that "secondary" translates to "not used 90% of the time" and then imply that anything unreliable is useless without supporting evidence. Uranium processing, an inherently very unreliable recipe already in vanilla, produces enough 235 to start Kovarex in every map, and while people hem and haw about how long it sometimes takes, it always works in the end. Unreliability, while limiting, does not make things "useless", and Fury even qualifies that it would not significantly affect overall power generation. Ergo, this response is hyperbole, used to denigrate Fury's suggestion.

Qon wrote: Fri May 31, 2019 8:56 am

FuryoftheStars wrote: Fri May 31, 2019 2:03 am

Qon wrote: Fri May 31, 2019 12:54 am If you are exhausting ore fields

Nope, wasn’t saying that.

Qon wrote: Fri May 31, 2019 12:54 am no power source requires maintenance. Initial investment with no maintenance cost is a common factor for all of them.

Not strictly true. All ore fields have limits and eventually you do need to find another.

I don't think this level contradictory beliefs are conducive to a fruitful discussion. Maybe take a break until the cognitive dissonance settles down?

Fury made 2 statements, the latter of which can be verified by anyone who has played the game, and the former of which you cannot verify from a 3rd person standpoint (conjecture is not veracity), and yet you suggest that Fury's beliefs are contradictory and level an accusation of cognitive dissonance. This is libel.

Qon wrote: Fri May 31, 2019 9:12 am

Theikkru wrote: Fri May 31, 2019 1:36 am It would help if you didn't make erroneous assumptions about how other people play, for a start.

I didn't. You told me that yourself:

Theikkru wrote: Thu May 30, 2019 2:06 pm With solar, the ONLY requirement is that you find enough space to throw down a blueprint.

You can't say it's the only requirement on a world where there's no space to find. If you need to do other things first then it's a lie.

This second statement of mine appears repeatedly in this conversation, so I see fit to elaborate here. First, this IS in fact a conditional statement, as anyone studied in logic would recognize:

With [solar], the ONLY requirement is that [you find enough space to throw down a blueprint].
⇔[you find enough space to throw down a blueprint] is sufficient to satisfy [solar]
⇔If [you find enough space to throw down a blueprint] then [solar].

Paraphrasing does not change its conditional nature. Logical rules dictate that such a statement has no bearing on the prerequisites required to meet the condition, or what may or may not be true if the condition is not. The content of a conditional statement concerns only the case where the condition is true.

Further, similar rules hold in plain English. If someone says that "the only thing you need to do to quench your thirst is go to a sink and pour yourself a glass of water", that statement is in no way invalidated if "going to a sink" involves intermediate steps such as "lifting the left leg, followed by the right", or "climbing a flight of stairs". The statement is also no less true if there is no sink; it only becomes inapplicable. Thus, returning to my original statement, it is not invalidated simply because "finding space" may involve other intermediate steps, such as exploring or clearing biters. The statement also remains true on a map where space is unavailable; it simply becomes unsatisfiable. Unsatisfiable and inapplicable do not equate to false.

Ergo, your "Qonclusion", as you put it, is erroneous. What's more, you claim that I "told" you that I played without biters or water, which is patently false, as my statement contained no such information. Inductions and deductions based based on what I DID say, correct or otherwise, are not sufficiently direct to constitute me telling you. You also tried to insinuate that I was lying, which is insulting.

Qon wrote: Fri May 31, 2019 9:12 am

Theikkru wrote: Fri May 31, 2019 1:36 am You pointed out the solution to the blueprint placing problem yourself in the linked post: multi-stage prints.

Multiple stages mean you just spend more of your time.

You imply here, without evidence, that placing large, multi-stage blueprints takes more user time than placing multiple smaller blueprints. While larger blueprints may take longer to complete than smaller ones, it is not necessary to watch over the whole process, so most of the time taken does not qualify as player time. This is hyperbole.

Qon wrote: Fri May 31, 2019 12:23 pm

FuryoftheStars wrote: Fri May 31, 2019 10:58 am I think you may need to take a step back and reread your own replies as an outside observer and maybe take you’re own advice.

No u.

Good one. Really Qonvinced me. No Qontext. Just because you say that I'm wrong, I am? No need to demonstrate anything? This is silly.

Both Fury and I tried to be tactful, and requested that you check your own words without being too specific, but you insisted, so here we are. Above, not only did I demonstrate multiple instances of errors in reasoning, exaggeration, baseless statements, and insults of varying directness, you have since accused us of twisting your words, whereas in this last line here alone, you claim Fury said that you're wrong while quoting a reply with no such statement. In context, you accused me of twisting things where I referenced time and exploration, which are measures of player action that you subsequently repeatedly interpreted as "space", irrespective of Fury's correction (which I also quoted for clarity), and where I later made a general clarifying statement that DID NOT reference anything you said (I was replying more to Koub). These examples (and ones in following replies) show hypocrisy, also immediately apparent with a page search for the words "wrong" and "No u".

Nor is this an isolated case. I distinctly remember replies in one other topics that started with a rather rude "Wrong.", or summarily dismissed a case I provided as "a toy example".

As such, I must declare an unusually low opinion of you, and I hold it as grounds to refuse you my artillery designs. Regardless, I feel no need to provide a specific example to prove the point under contention. It should be sufficient to assert that a recursive blueprint for artillery exists, and that said blueprint is no harder to deploy than any other recursive blueprint. Since such a blueprint lacks a preparatory step requisite of other recursive blueprints, that is, clearing biters from the space into which it is to be deployed, it is therefore categorically simpler to deploy than other recursive blueprints, and ergo trivial. If reasonable doubt cannot be cast upon these assertions, I consider the point made.

If you think that space in nuclear setups is too insignificant to be considered a main design factor, then fine, I can concede that point for the sake of argument, but my case there still holds: nuclear is more complex because it has many more factors to balance.

As for the design considerations of backup power, I should point out that you don't need to supply an entire base's worth of backup power, only as much as is necessary to account for the amount lost by intermittent behavior, especially since this is assuming the presence of a solar setup and its accumulators, which will provide buffering for large power spikes or dropouts anyways. The requirements for a backup with 5~10% total power capability are very low.

Theikkru wrote: Sat Jun 01, 2019 1:33 pm

FuryoftheStars wrote: Sat Jun 01, 2019 6:32 amthis would be nothing more than another fuel source for boilers.[*]

* Steam engines/turbines

I think solar boilers would be different because you could make things more complex by having the temperature from solar boiler steam change over the course of the day, which would make for some very interesting new mechanics for steam distribution and energy balancing. If the temperature peaked at 200℃ or so, for instance, you'd have to either use turbines (expensive!) or include mixing buffers in your design to avoid energy wastage, since steam engines can't utilize energy above 165℃.

I think what I meant with another fuel source for boilers is that you could use coal, solid fuel, etc to run the boiler. This just adds solar as one of the options. In my mind, anyway.

I’m not sure about the rest. I mean, technically speaking, you could achieve the same thing with the other fuel sources. I just always figured there were automatic controls in the boiler to reduce the burn rate / vent the steam out the steam pipe once it was to temp. What you’re talking about with overheating and needing to cool it back down I think is something that could happen/be done with all the boilers? Just don’t know how I feel on that.

Theikkru wrote: Sat Jun 01, 2019 5:01 pm Really? We're doing this? Fine, if you want me to spell things out then, let's have a look at these "well founded arguments" as you put it. Here are my quotes of you from my earlier post:

Qon wrote: Fri May 31, 2019 12:54 amIf you are exhausting ore fields in megabase stage then you are doing it wrong. Resources are effectivly infinite in 0.17

To start, this is incorrect. Ore fields are not infinite,

Did you not read what you quoted or don't you understand the difference between "effectively infinite" and "infinite"? If you don't understand words then you can just duckduckgo them. Is there even a point in reading the rest of your post when it begins like this? You can't just keep attacking strawmen like this. It's not working.

FuryoftheStars wrote: Sat Jun 01, 2019 5:03 pm I think what I meant with another fuel source for boilers is that you could use coal, solid fuel, etc to run the boiler. This just adds solar as one of the options. In my mind, anyway.

I’m not sure about the rest. I mean, technically speaking, you could achieve the same thing with the other fuel sources. I just always figured there were automatic controls in the boiler to reduce the burn rate / vent the steam out the steam pipe once it was to temp. What you’re talking about with overheating and needing to cool it back down I think is something that could happen/be done with all the boilers? Just don’t know how I feel on that.

I just see it as a lot more thematic with solar, since not only do we already have a day-night cycle with them, it makes sense: solar power of any kind gets more energy the more directly light is shining on it (with maximum at midday). At the levels I suggested, it doesn't even need to be a critical consideration, just a potential optimization. Even now, a lot of people will throw together an uncontrolled initial nuclear setup just to get it working, and not worry about fuel wastage, heat loss, or circuit control until later down the line (if ever). Steam that's only a bit over the maximum usable temperature would put it in that category, where people just wanting to get something up and running could ignore the small losses and throw together some direct connections, whereas optimization and efficiency could be handled later.

Theikkru wrote: Sat Jun 01, 2019 5:19 pm

FuryoftheStars wrote: Sat Jun 01, 2019 5:03 pm I think what I meant with another fuel source for boilers is that you could use coal, solid fuel, etc to run the boiler. This just adds solar as one of the options. In my mind, anyway.

I’m not sure about the rest. I mean, technically speaking, you could achieve the same thing with the other fuel sources. I just always figured there were automatic controls in the boiler to reduce the burn rate / vent the steam out the steam pipe once it was to temp. What you’re talking about with overheating and needing to cool it back down I think is something that could happen/be done with all the boilers? Just don’t know how I feel on that.

I just see it as a lot more thematic with solar, since not only do we already have a day-night cycle with them, it makes sense: solar power of any kind gets more energy the more directly light is shining on it (with maximum at midday). At the levels I suggested, it doesn't even need to be a critical consideration, just a potential optimization. Even now, a lot of people will throw together an uncontrolled initial nuclear setup just to get it working, and not worry about fuel wastage, heat loss, or circuit control until later down the line (if ever). Steam that's only a bit over the maximum usable temperature would put it in that category, where people just wanting to get something up and running could ignore the small losses and throw together some direct connections, whereas optimization and efficiency could be handled later.

Ok, I think I understand where you’re coming from. Not sure if I’d like it still, but I understand.

Are any of some of these suggestions moddable, or are some of these ideas divergent enough that it’d be too hard to do with the current game mechanics?

I seem to remember some mods that had solar panels output heat on heat pipes, but I don't remember any outputting steam directly, or varying its temperature.

Qon wrote: Sat Jun 01, 2019 5:12 pm

Theikkru wrote: Sat Jun 01, 2019 5:01 pm Really? We're doing this? Fine, if you want me to spell things out then, let's have a look at these "well founded arguments" as you put it. Here are my quotes of you from my earlier post:

Qon wrote: Fri May 31, 2019 12:54 amIf you are exhausting ore fields in megabase stage then you are doing it wrong. Resources are effectivly infinite in 0.17

To start, this is incorrect. Ore fields are not infinite,

Did you not read what you quoted or don't you understand the difference between "effectively infinite" and "infinite"? If you don't understand words then you can just duckduckgo them. Is there even a point in reading the rest of your post when it begins like this? You can't just keep attacking strawmen like this. It's not working.

I wasn't going to address this, but then it struck me how hilariously ironic this is. You truncated this quote of me, excluding the half-sentence that addressed the "effectively" part of your "infinite", so that you could accuse me of using a straw man argument, and then used that to dismiss the rest of my post.

You actually used a straw man argument, to accuse me of using a straw man argument! Incredible!

Theikkru wrote: Sun Jun 02, 2019 1:11 am I wasn't going to address this, but then it struck me how hilariously ironic this is. You truncated this quote of me, excluding the half-sentence that addressed the "effectively" part of your "infinite", so that you could accuse me of using a straw man argument, and then used that to dismiss the rest of my post.

You actually used a straw man argument, to accuse me of using a straw man argument! Incredible!

You didn't address it.

Theikkru wrote: Sat Jun 01, 2019 5:01 pm To start, this is incorrect. Ore fields are not infinite, and megabases can run for long periods of time at high consumption rates, so ore fields can and will expire if you have a big enough base and play long enough.

At the edge of the map you get another 100 times richer ore fields and a megabase can get mining productivity of 5x output. You can have ore fields that last for 100 000 hours at 5k/min then. So your claim is that megabases run non-stop 24/7 for 11 years and have to replace their mining drills then? Ok, you are right. They are not effectively infinite for your kind of long lasting base. Too bad the map is so small that you can't find ore fields that last 640 years at that rate (unless you crank up the settings and get just that).

Even if you use mining productivity to just boost your output speed and use speed modules to increase the mining rate to 71k/minute (drained, you get much more output with productivity research, and you need max settings to get fields big enough to support that rate) a 348G field lasts 221h. Or the default settings 6.2G drained at 16k/min with speed modules lasts a while too.

Of course, you don't need to go that far, you just go far enough that the resource fields will not run out for you.

I did very well address it. Megabases do not start at the "edge of the map" (not sure where that's supposed to be), unless you go out of your way to put them there. By default, they start right near your starting point, where richness is at or near base value. Until you go far enough to find those ore fields rich enough to last the span of your play, you will be tapping ore fields that will run out. Your statement did not make any qualifications about the nature of the ore fields, it simply declared that anyone exhausting any ore fields is wrong. (For pedantry's sake I will include that "megabase stage" only qualifies the state of the player's base, and says nothing about the ore fields.) That is accusatory hyperbole.

Sure, be like that.
You have a megabase but you still build everything around the starting puddle of water. You make sure to only tap the smallest resource patches on a 2Mx2M sized map. And you never make a train because you never leave the starting area anyways. If that's true then I'll grant you that resources are limited with that playstyle.

Qon wrote: Fri May 31, 2019 12:54 am If you are exhausting ore fields in megabase stage then you are doing it wrong.

Maybe don't play in the worst possible way and then complain about how your resources run out. Adjust your resource settings to max and install mods that make them last forever. Don't come here complaining about the balance saying that your resources run out at megabase stage when you avoid the rich ones on purpose.

If you are insisting on playing like that I can claim that solar panels don't last forever either. Since they take so much space they are annoying to walk around so my playstyle is to hop into my tank and drive right through them. Or run with nukes through them.
I think we need to buff solar panels because they cost so much compared to how short they last.

You implied that megabase players should only use inexhaustible ore fields.
I simply refuted that use of exhaustible ore fields is normal.
Then you try to make it sound like I advocated the EXCLUSIVE use of exhaustible ore fields.
Who's twisting words and making straw man arguments now?

Theikkru wrote: Sun Jun 02, 2019 5:25 pm You implied that megabase players should only use inexhaustible ore fields.
I simply refuted that use of exhaustible ore fields is normal.
Then you try to make it sound like I advocated the EXCLUSIVE use of exhaustible ore fields.
Who's twisting words and making straw man arguments now?

No. I'm talking about how silly it is to consider solar panels OP simply because they don't run out. They are just simpler to blueprint. And coal can run out in early game if you are working with the starting resources. You can use exhaustible patches if you feel like it, but then don't use that as a basis for balancing optimal megabase building when you aren't playing smart.
Don't forget the topic.

My case isn't that they're just OP, it's that they're too simple and boring to set up and run for the amount of power they generate. If their setup were more involved, I'd be perfectly fine with an appropriate buff to their power delivery.

Qon wrote: Sun Jun 02, 2019 8:37 pm

Theikkru wrote: Sun Jun 02, 2019 5:25 pm You implied that megabase players should only use inexhaustible ore fields.
I simply refuted that use of exhaustible ore fields is normal.
Then you try to make it sound like I advocated the EXCLUSIVE use of exhaustible ore fields.
Who's twisting words and making straw man arguments now?

No. I'm talking about how silly it is to consider solar panels OP simply because they don't run out. They are just simpler to blueprint. And coal can run out in early game if you are working with the starting resources. You can use exhaustible patches if you feel like it, but then don't use that as a basis for balancing optimal megabase building when you aren't playing smart.
Don't forget the topic.

Sorry to say but you are wrong there. Ore fields do run out. Solar does not. Once you put down a solar cell it works forever. All other energy sources require to move your miners from time to time. That time might be hours or days but it happens. And usually at the worst possible moment. And yes, there are more ore fields than you can reasonably exhaust and they grow with distance. But they still always run out at some point and you have to take action to mine the next patch.

Individual ore fields are short lived.

Theikkru wrote: Sat Jun 01, 2019 5:19 pm

FuryoftheStars wrote: Sat Jun 01, 2019 5:03 pm I think what I meant with another fuel source for boilers is that you could use coal, solid fuel, etc to run the boiler. This just adds solar as one of the options. In my mind, anyway.

I’m not sure about the rest. I mean, technically speaking, you could achieve the same thing with the other fuel sources. I just always figured there were automatic controls in the boiler to reduce the burn rate / vent the steam out the steam pipe once it was to temp. What you’re talking about with overheating and needing to cool it back down I think is something that could happen/be done with all the boilers? Just don’t know how I feel on that.

I just see it as a lot more thematic with solar, since not only do we already have a day-night cycle with them, it makes sense: solar power of any kind gets more energy the more directly light is shining on it (with maximum at midday). At the levels I suggested, it doesn't even need to be a critical consideration, just a potential optimization. Even now, a lot of people will throw together an uncontrolled initial nuclear setup just to get it working, and not worry about fuel wastage, heat loss, or circuit control until later down the line (if ever). Steam that's only a bit over the maximum usable temperature would put it in that category, where people just wanting to get something up and running could ignore the small losses and throw together some direct connections, whereas optimization and efficiency could be handled later.

I think such a setup would be simple to control. Add a tank after the boilers (or many) and a pump set to activate at "steam > 10000" before the steam engine. Then during the day the combined steam in the tank heats up and at dawn/dust it cools down. At night the tank simply drains to 10000. Keeping that much steam in the tank should give enough buffering to average the temp somewhat.