[0.15.1] [kovarex] Non-Conducting heat pipes after 60 tile

This subforum contains all the issues which we already resolved.
saturn7
Long Handed Inserter
Long Handed Inserter
Posts: 89
Joined: Wed Apr 27, 2016 6:14 pm
Contact:

Re: [0.15.1] Non-Conducting heat pipes after 60 tile

Post by saturn7 »

You can still use robots, but the maximum reliable heat pipe length is limited to 30 or so.

User avatar
MeduSalem
Smart Inserter
Smart Inserter
Posts: 1486
Joined: Sun Jun 08, 2014 8:13 pm
Contact:

Re: [0.15.1] Non-Conducting heat pipes after 60 tile

Post by MeduSalem »

saturn7 wrote:You can still use robots, but the maximum reliable heat pipe length is limited to 30 or so.
Probably less. I used Robots for placement when I didn't know about the issue and it already caused efficiency issues with the Exchangers closest to the Reactor... with maybe 5-10 tiles distance. So at a distance of 30 tiles they are probably just purely cosmetic.

Vader
Burner Inserter
Burner Inserter
Posts: 6
Joined: Wed Apr 26, 2017 2:20 am
Contact:

Re: [0.15.1] Non-Conducting heat pipes after 60 tile

Post by Vader »

MeduSalem wrote:
saturn7 wrote:You can still use robots, but the maximum reliable heat pipe length is limited to 30 or so.
Probably less. I used Robots for placement when I didn't know about the issue and it already caused efficiency issues with the Exchangers closest to the Reactor... with maybe 5-10 tiles distance. So at a distance of 30 tiles they are probably just purely cosmetic.
The maximum reliable heat pipe length remains 30.
And by reliable I mean it can transfer heat above 500 C which is the minimum for steam production. One tile further and the heat pipe may no longer heat a heat-exchanger above 500 C. But until 30, it will remain usable.
(I made experiments the absolute maximum with 30 tile limit is a 2x6 reactor setup with 1760 MW but that build is extremely difficult to get right and not even sure it can be built)

mOoEyThEcOw
Inserter
Inserter
Posts: 22
Joined: Fri Mar 17, 2017 11:27 pm
Contact:

Re: [0.15.1] Non-Conducting heat pipes after 60 tile

Post by mOoEyThEcOw »

Klonan wrote:
There is something wrong with the design of the system, we are aware of this, the mechanic of the heat transfer isn't clear, from a gameplay perspective it isn't working well

However the actual implementation is working as intended, it was written to act as it is acting, and thus there is no bug.

We have had some discussion about changing the way they work, but as it stands, it is working as it was designed
Then this would appear to be a report on the flaw of the design. When your design is so egregiously wrong a large number of people think it is a bug, and no one else thinks it's a feature (as the "Not a bug" forum labels it), it's probably a very large bug. Let alone the fact that this feature breaks (more important) orthogonal features: drone placement, blueprints, auto replacement after destruction. It also is just inconsistent.

I just don't understand how you can implement complex train systems, conveyors, and fluid pipes that can be hooked up in any order and be consistent, but not a simple heat pipe... and still call it a feature!

malventano
Filter Inserter
Filter Inserter
Posts: 340
Joined: Thu Apr 27, 2017 4:31 pm
Contact:

Re: [0.15.1] Non-Conducting heat pipes after 60 tile

Post by malventano »

Klonan wrote:
malventano wrote:
Rseding91 wrote:This is currently how heat pipes work. The order you build the pipes will change how they behave.

It's not really desired to work that way but it's not a bug. We may change it in the future.
I'm sorry guys but this is most definitely a bug - in more ways than one. Something is absolutely broken with the mechanic as it is currently implemented. You guys are the best / most active developers I have seen across any software project ever, but it baffles me that you can look at what heat pipes are doing and say there is nothing wrong with the implementation. You might as well be saying that 2+2=5 is not a bug.
There is something wrong with the design of the system, we are aware of this, the mechanic of the heat transfer isn't clear, from a gameplay perspective it isn't working well

However the actual implementation is working as intended, it was written to act as it is acting, and thus there is no bug.

We have had some discussion about changing the way they work, but as it stands, it is working as it was designed
I recommend implementing it (internally, code wise) parallel to how pipes / flow functions. Treat the reactor / pipes as volumes of water, where the fluid level (heat) 'flows' from high (hot) to cold (low). This should prevent reinventing the wheel, which appears to not be working as intended in current form. Treating heat similar to water level will also prevent the odd mismatches we are seeing (where sections of heat pipes with no heat flow show temperature differences as high as 20C). Further, your intended 'transfer rate' would actually work as expected, since it currently works for pipes. You would naturally see temperature deltas across pipes as you increased the thermal load (flow) on them, which is exactly what you would expect with real heat pipes.
Allyn Malventano
---
Want to improve fluid flow between pumps / across longer distances? Try my Manifolds mod.

saturn7
Long Handed Inserter
Long Handed Inserter
Posts: 89
Joined: Wed Apr 27, 2016 6:14 pm
Contact:

Re: [0.15.1] Non-Conducting heat pipes after 60 tile

Post by saturn7 »

mOoEyThEcOw wrote:
Klonan wrote:
There is something wrong with the design of the system, we are aware of this, the mechanic of the heat transfer isn't clear, from a gameplay perspective it isn't working well

However the actual implementation is working as intended, it was written to act as it is acting, and thus there is no bug.

We have had some discussion about changing the way they work, but as it stands, it is working as it was designed
Then this would appear to be a report on the flaw of the design. When your design is so egregiously wrong a large number of people think it is a bug, and no one else thinks it's a feature (as the "Not a bug" forum labels it), it's probably a very large bug. Let alone the fact that this feature breaks (more important) orthogonal features: drone placement, blueprints, auto replacement after destruction. It also is just inconsistent.

I just don't understand how you can implement complex train systems, conveyors, and fluid pipes that can be hooked up in any order and be consistent, but not a simple heat pipe... and still call it a feature!
I'm sure the devs are aware of this, but for reason we don't know they have decided to postpone addressing that problem for now. I guess there is some deeper problem with the design of heat pipes that would need considerable effort to fix; and either the devs are not sure yet how to fix it properly, or they have other priorities for the time being (fixing more critical bugs).

For the time being, while not ideal, hand-placing heat pipes, or limiting the length to 30 (and accepting the heat loss in between), will have to do.

Aeternus
Filter Inserter
Filter Inserter
Posts: 835
Joined: Wed Mar 29, 2017 2:10 am
Contact:

Re: [0.15.1] Non-Conducting heat pipes after 60 tile

Post by Aeternus »

If placement order is important then there should be a visual indicator for it, that way a player can at least spot when there's a problem when 2 heat pipes aren't properly connecting or the heat transfer direction is faulty. I get the sense that the heat pipes act similar to conveyors rather then fluid pipes in that they move heat directionally. If this is the case, simply give a visual indication of which direction they are going in, and allow players to change the orientation of heatpipes as needed.

The system isn't neccesarily wrong, but the lack of tools to diagnose and fix unintended configurations is. As it stands, placing heatpipes from blueprints is asking for problems - and that can't be an intended situation.

User avatar
Optera
Smart Inserter
Smart Inserter
Posts: 2916
Joined: Sat Jun 11, 2016 6:41 am
Contact:

Re: [0.15.1] Non-Conducting heat pipes after 60 tile

Post by Optera »

Klonan wrote: There is something wrong with the design of the system, we are aware of this, the mechanic of the heat transfer isn't clear, from a gameplay perspective it isn't working well

However the actual implementation is working as intended, it was written to act as it is acting, and thus there is no bug.

We have had some discussion about changing the way they work, but as it stands, it is working as it was designed
Since you know heat pipes are not working as intended regardless if it's by design or implementation this issue is a known bug and certainly not not a bug.

mOoEyThEcOw
Inserter
Inserter
Posts: 22
Joined: Fri Mar 17, 2017 11:27 pm
Contact:

Re: [0.15.1] Non-Conducting heat pipes after 60 tile

Post by mOoEyThEcOw »

saturn7 wrote: I'm sure the devs are aware of this, but for reason we don't know they have decided to postpone addressing that problem for now. I guess there is some deeper problem with the design of heat pipes that would need considerable effort to fix; and either the devs are not sure yet how to fix it properly, or they have other priorities for the time being (fixing more critical bugs).
I would agree with you except the devs marked it as "Not a bug" ("It's a feature") rather than "known issue". If this is a feature working as intended then it doesn't sound like they plan on fixing it anytime soon (like it doesn't even sound like they are planning on doing anything about it this major version; which means it will probably be in 1.0 or last for this entire year), which is super frustrating (I spent like 4 hours trying to figure out why my heat pipe setups weren't working, I got pretty frustrated from dealing with it, major pain point).

Vin
Fast Inserter
Fast Inserter
Posts: 106
Joined: Sat Jun 21, 2014 2:46 pm
Contact:

Re: [0.15.1] Non-Conducting heat pipes after 60 tile

Post by Vin »

I think changing this behavior should be a priority. I don't have anything to say that already hasn't been said, but consistent mechanics that don't have anti-synergy with other mechanics are important enough that I felt the need to say something.

SaurusRex
Burner Inserter
Burner Inserter
Posts: 5
Joined: Tue May 09, 2017 7:49 pm
Contact:

Re: [0.15.1] Non-Conducting heat pipes after 60 tile

Post by SaurusRex »

Signed up after spending a while utterly baffled and eventually finding this thread.

We set up a nuclear plant at our base, but noticed something odd about the heat pipes. Here's a screenshot of the heat exchanger layout for reference (yes, I'm aware of the many ways in which it could be improved).
Screenshot
There was a strange temperature gradient where the bottom left corner was barely 330 degrees, but the top right corner was nearly 1,000 degrees. We figured it would stabilise after a while, but over 30 hours later it remained like that. We got to the point where we weren't getting enough power, and set out to resolve the issue. After some experimentation got absolutely nowhere and only resulted in more confusion, it was time to do some googling, and I stumbled upon this thread and similar ones on reddit. Upon learning of this oddity we proceed to rip up all the heat pipes and carefully - slowly - place them down one at a time from the reactor outwards. As of now, they're all at least 970 degrees, with absolutely no changes to layout.

If this is "working as designed", then that design should be scrapped.

I've played this game for hundreds of hours, and am consistently impressed with the level of detail and care that has gone into this game, which makes it all the more disappointing that such a fundamentally broken part of an exciting new feature is apparently working as intended and that it MIGHT be changed in the future. The behaviour as it stands is unintuitive, undocumented, and quite frankly bizarre. Please don't let this ugly stain on an otherwise outstanding game remain for long.

factoriouzr
Filter Inserter
Filter Inserter
Posts: 660
Joined: Sat Jun 06, 2015 2:23 am
Contact:

Re: [0.15.1] Non-Conducting heat pipes after 60 tile

Post by factoriouzr »

SaurusRex wrote:Signed up after spending a while utterly baffled and eventually finding this thread.

We set up a nuclear plant at our base, but noticed something odd about the heat pipes. Here's a screenshot of the heat exchanger layout for reference (yes, I'm aware of the many ways in which it could be improved).
Screenshot
There was a strange temperature gradient where the bottom left corner was barely 330 degrees, but the top right corner was nearly 1,000 degrees. We figured it would stabilise after a while, but over 30 hours later it remained like that. We got to the point where we weren't getting enough power, and set out to resolve the issue. After some experimentation got absolutely nowhere and only resulted in more confusion, it was time to do some googling, and I stumbled upon this thread and similar ones on reddit. Upon learning of this oddity we proceed to rip up all the heat pipes and carefully - slowly - place them down one at a time from the reactor outwards. As of now, they're all at least 970 degrees, with absolutely no changes to layout.

If this is "working as designed", then that design should be scrapped.

I've played this game for hundreds of hours, and am consistently impressed with the level of detail and care that has gone into this game, which makes it all the more disappointing that such a fundamentally broken part of an exciting new feature is apparently working as intended and that it MIGHT be changed in the future. The behaviour as it stands is unintuitive, undocumented, and quite frankly bizarre. Please don't let this ugly stain on an otherwise outstanding game remain for long.

I completely agree. I spend Hours upon hours of building the perfect reactor and I have tried fixing issues that I later found out are because of the heat pipe mechanics. How is this not fixed yet? How is it ok to make heat pipes dependent on the order you lay the pipes. That completely goes in the face of using robots to build the reactor. Using robots is a fundamental and important feature of the game.

factoriouzr
Filter Inserter
Filter Inserter
Posts: 660
Joined: Sat Jun 06, 2015 2:23 am
Contact:

Re: [0.15.1] Non-Conducting heat pipes after 60 tile

Post by factoriouzr »

Rseding91 wrote:This is currently how heat pipes work. The order you build the pipes will change how they behave.

It's not really desired to work that way but it's not a bug. We may change it in the future.

You are kidding right? This is completely a bug. It completely makes using robots obsolete. Robots and automated construction is a big part of the game that is completely rendered obsolete by this mechanic because robots will put down items in a random order.

I can't believe this isn't fixed yet. This is day one patch worthy and we are 10 releases (9 if you don't count 15.0) patches in.

malventano
Filter Inserter
Filter Inserter
Posts: 340
Joined: Thu Apr 27, 2017 4:31 pm
Contact:

Re: [0.15.1] Non-Conducting heat pipes after 60 tile

Post by malventano »

Gergely wrote:This bug is still not fixed in 0.15.9 Why is this "Not a bug"?! Placement order matters now? :cry: What about blueprints? Robot construction? We all have to "live with this?"
I've proposed a possible solution that should clear up the confusion/other issues under the suggestion forum. It's an expansion on my earlier post in this thread, but with numbers added to make implementation as simple as possible.
Allyn Malventano
---
Want to improve fluid flow between pumps / across longer distances? Try my Manifolds mod.

factoriouzr
Filter Inserter
Filter Inserter
Posts: 660
Joined: Sat Jun 06, 2015 2:23 am
Contact:

Re: [0.15.1] Non-Conducting heat pipes after 60 tile

Post by factoriouzr »

mOoEyThEcOw wrote:
Klonan wrote:
There is something wrong with the design of the system, we are aware of this, the mechanic of the heat transfer isn't clear, from a gameplay perspective it isn't working well

However the actual implementation is working as intended, it was written to act as it is acting, and thus there is no bug.

We have had some discussion about changing the way they work, but as it stands, it is working as it was designed
Then this would appear to be a report on the flaw of the design. When your design is so egregiously wrong a large number of people think it is a bug, and no one else thinks it's a feature (as the "Not a bug" forum labels it), it's probably a very large bug. Let alone the fact that this feature breaks (more important) orthogonal features: drone placement, blueprints, auto replacement after destruction. It also is just inconsistent.

I just don't understand how you can implement complex train systems, conveyors, and fluid pipes that can be hooked up in any order and be consistent, but not a simple heat pipe... and still call it a feature!
Exactly this! +100

I already mentioned random placement, and bot construction issues. I also wanted to add, what you so eloquently already mentioned. I also can't understand how you can implement other complex path finding systems like trains, belts, and pipes, but not an even simpler and smaller range (typically) heat pipe. I mean you build train construction that finds paths around obstacles, yet heat pipes are directional. I can't believe you are saying this is as designed and intended behaviour.

This literally breaks other key features of the game and is in direct contradition to them. It's inconsistent with everything else in the game and breaks prominent game mechanics like automated (robot) construction and automated replacement if destroyed (as already mentioned).

I literally spent hours trying to figure out why my reactor (with the proper ratios) wasn't working with all turbines and that seemingly unrelated changes would make some parts work yet other fail randomly. When I figured out it was the heat pipes because of the order I laid them, I was very disappointed. I even used robots to construct and reconstruct and move each 1/4 of my reactor to make room for pumps and other things to try an increase liquid and steam flow during the many iterations to try and solve the performance issues of the reactor. I then discovered it was the order of the heat pipes. This is very frustrating and frankly unbelievable that you would design something like this that contradicts the way everything else works in the game and breaks existing game features.

Swift016
Manual Inserter
Manual Inserter
Posts: 1
Joined: Wed May 10, 2017 1:36 am
Contact:

Re: [0.15.1] Non-Conducting heat pipes after 60 tile

Post by Swift016 »

SaurusRex wrote:Signed up after spending a while utterly baffled and eventually finding this thread.

We set up a nuclear plant at our base, but noticed something odd about the heat pipes. Here's a screenshot of the heat exchanger layout for reference (yes, I'm aware of the many ways in which it could be improved).
Screenshot
There was a strange temperature gradient where the bottom left corner was barely 330 degrees, but the top right corner was nearly 1,000 degrees. We figured it would stabilise after a while, but over 30 hours later it remained like that. We got to the point where we weren't getting enough power, and set out to resolve the issue. After some experimentation got absolutely nowhere and only resulted in more confusion, it was time to do some googling, and I stumbled upon this thread and similar ones on reddit. Upon learning of this oddity we proceed to rip up all the heat pipes and carefully - slowly - place them down one at a time from the reactor outwards. As of now, they're all at least 970 degrees, with absolutely no changes to layout.

If this is "working as designed", then that design should be scrapped.

I've played this game for hundreds of hours, and am consistently impressed with the level of detail and care that has gone into this game, which makes it all the more disappointing that such a fundamentally broken part of an exciting new feature is apparently working as intended and that it MIGHT be changed in the future. The behaviour as it stands is unintuitive, undocumented, and quite frankly bizarre. Please don't let this ugly stain on an otherwise outstanding game remain for long.
Just registered to quote this.

This "design decision" is stupid and unintuitive, end of story.

User avatar
TruePikachu
Filter Inserter
Filter Inserter
Posts: 978
Joined: Sat Apr 09, 2016 8:39 pm
Contact:

Re: [0.15.1] Non-Conducting heat pipes after 60 tile

Post by TruePikachu »

Something which I feel might be missing from this thread is an explanation as to how the currently-implemented system works, so that it could be better understood as to why the "issue" is being considered "Not a bug".

Going by the behavior reported here, I believe the underlying implementation is as such: there is a linked list of all the game entities, which is iterated across every game tick so that entities can update themselves; when an entity is created, it is placed at the end of the list, and when an entity is destroyed, it is removed from the list the next time the tick loop reaches it. Heat pipe conductance is done, as such, in the order they were placed, considering their own temperature, their neighbor's temperatures, and the distance from the original heat source.

I feel that, if the above is correct, the best way to implement the heat pipe's update routine would be to ensure that a heat pipe's temperature's changes aren't considered inside of a world update. I think the fastest way to do this would be to have 3 distinct pairs of temperature and delta fields, where the "current", "previous", and "next" pairs cycle every tick (and can be determined by a global counter); the update routine would read the temperature+delta sums that were in effect the previous tick, set the delta fields for the current tick, and prepare the pair for the next tick by zeroing the delta and setting the temperature to the sum of the "previous" temperature+delta and the "current" temperature (not delta), and dividing by two. This would result in nothing considered for calculations being changed inside a tick (such that the build order does not influence the behavior of the pipes), and the two "temperature states" (which alternate each cycle) remaining somewhat identical.

EDIT: I just noticed that the above proposed algorithm can be reduced to using only two pairs of values; it would be acceptable for the entity to update its "previous" pair instead of the theoretical "next" pair, since the temperature+delta would remain the same, and that is what everything should be looking at to compute how the temperatures should end up changing.

malventano
Filter Inserter
Filter Inserter
Posts: 340
Joined: Thu Apr 27, 2017 4:31 pm
Contact:

Re: [0.15.1] Non-Conducting heat pipes after 60 tile

Post by malventano »

TruePikachu wrote:Something which I feel might be missing from this thread is an explanation as to how the currently-implemented system works, so that it could be better understood as to why the "issue" is being considered "Not a bug".

Going by the behavior reported here, I believe the underlying implementation is as such: there is a linked list of all the game entities, which is iterated across every game tick so that entities can update themselves; when an entity is created, it is placed at the end of the list, and when an entity is destroyed, it is removed from the list the next time the tick loop reaches it. Heat pipe conductance is done, as such, in the order they were placed, considering their own temperature, their neighbor's temperatures, and the distance from the original heat source.

I feel that, if the above is correct, the best way to implement the heat pipe's update routine would be to ensure that a heat pipe's temperature's changes aren't considered inside of a world update. I think the fastest way to do this would be to have 3 distinct pairs of temperature and delta fields, where the "current", "previous", and "next" pairs cycle every tick (and can be determined by a global counter); the update routine would read the temperature+delta sums that were in effect the previous tick, set the delta fields for the current tick, and prepare the pair for the next tick by zeroing the delta and setting the temperature to the sum of the "previous" temperature+delta and the "current" temperature (not delta), and dividing by two. This would result in nothing considered for calculations being changed inside a tick (such that the build order does not influence the behavior of the pipes), and the two "temperature states" (which alternate each cycle) remaining somewhat identical.

EDIT: I just noticed that the above proposed algorithm can be reduced to using only two pairs of values; it would be acceptable for the entity to update its "previous" pair instead of the theoretical "next" pair, since the temperature+delta would remain the same, and that is what everything should be looking at to compute how the temperatures should end up changing.
A form of this already happens with pipes and tanks, which is why I've suggested that the heat system function using a simple port of that code. If I was familiar enough with the mod API, I'd try and roll my own fix, but I'm not nearly experienced enough to pull that off just yet.
Allyn Malventano
---
Want to improve fluid flow between pumps / across longer distances? Try my Manifolds mod.

mOoEyThEcOw
Inserter
Inserter
Posts: 22
Joined: Fri Mar 17, 2017 11:27 pm
Contact:

Re: [0.15.1] Non-Conducting heat pipes after 60 tile

Post by mOoEyThEcOw »

TruePikachu wrote: ...
Or since there are only sources and sinks they could use some basic network flow algorithms, with a simple update step that involved only updating the connected buildings (or a pipe on demand when hovered). There are even versions that can be updated in place (although the algorithms are only O(V+E) in the first place).

Or they could use their existing pipe system and treat the heat as a fluid.

User avatar
TruePikachu
Filter Inserter
Filter Inserter
Posts: 978
Joined: Sat Apr 09, 2016 8:39 pm
Contact:

Re: [0.15.1] Non-Conducting heat pipes after 60 tile

Post by TruePikachu »

I was mainly describing something which would potentially take a minimal amount of effort to implement.

I'm not entirely sure if the fluid system would actually work plausably for this, since it is intended for there to be a distance limit for heat transfer; fluids don't have any concept of distance, though, nor could it be implemented sensibly on top of the existing logic.

Post Reply

Return to “Resolved Problems and Bugs”