Forged Alliance Forever Forged Alliance Forever Forums 2014-05-06T10:39:28+02:00 /feed.php?f=42&t=3759 2014-05-06T10:39:28+02:00 2014-05-06T10:39:28+02:00 /viewtopic.php?t=3759&p=72335#p72335 <![CDATA[Re: TML/SML building bug]]>
Closing the topic, it's getting nowhere. You either know what is happening and knows how to fix it : Then fix it and post the fix, or you don't or don't believe it's a bug and do nothing.

No need to endlessly debate about it.

Statistics: Posted by Ze_PilOt — 06 May 2014, 10:39


]]>
2014-05-06T09:20:33+02:00 2014-05-06T09:20:33+02:00 /viewtopic.php?t=3759&p=72332#p72332 <![CDATA[Re: TML/SML building bug]]>
Anaryl wrote:
I avoid dignifying anything IceDreamer has to say as he enjoys bullying twelve year olds out of the game to make himself feel like a man.

Frankly I wouldn't piss on him if he was on fire.


Oh my.

Statistics: Posted by errorblankfield — 06 May 2014, 09:20


]]>
2014-05-06T08:41:09+02:00 2014-05-06T08:41:09+02:00 /viewtopic.php?t=3759&p=72329#p72329 <![CDATA[Re: TML/SML building bug]]>
Frankly I wouldn't piss on him if he was on fire.

Statistics: Posted by Anaryl — 06 May 2014, 08:41


]]>
2014-05-06T08:33:42+02:00 2014-05-06T08:33:42+02:00 /viewtopic.php?t=3759&p=72328#p72328 <![CDATA[Re: TML/SML building bug]]>
IceDreamer wrote:
I'm claiming the same thing I always have: The Engine throttles silos differently to everything else. If you can provide a video showing that it does this via step-down throttling, great, but I don't really care :p


Of course I can't! But that's why I agree with your observation -the engine throttling subjectively.
That said, I still think your wording is causing this tiff with Anaryl. Both of what you've said can be semantically correct at the same time -yet sound opposing.

I can't communicate that feeling any clearer, so hopefully you two sort this out without too must duress. Either way, I can't help further so I shall take leave.

Statistics: Posted by errorblankfield — 06 May 2014, 08:33


]]>
2014-05-06T03:51:51+02:00 2014-05-06T03:51:51+02:00 /viewtopic.php?t=3759&p=72320#p72320 <![CDATA[Re: TML/SML building bug]]>
Crotalus wrote:
While you guys were writing walls of text here I did some sandboxing and is now content I reached the correct assumption. I won't waste time making any proof about it though, I think I was pretty clear about the outcome of the empirical experiment.

Question is, have anyone of you spent time verifying / falsifying my claim or is that too much to ask? You guys sounded so sure you knew how / why the launcher stalled like they did (buildtime / buildpower / balancing blabla). Did you just theorycraft it or did you find that out by testing? In my view you're the ones claiming unicorns by just make up statements about how the stalling works... I tested your claims and they were wrong



What you descreibed where the expected results that where in my working. Some credit please.

What was I do instead of testing this? Important things that don't include staring at a mass indcator.

Again if any of you actually read

Statistics: Posted by Anaryl — 06 May 2014, 03:51


]]>
2014-05-06T00:01:12+02:00 2014-05-06T00:01:12+02:00 /viewtopic.php?t=3759&p=72316#p72316 <![CDATA[Re: TML/SML building bug]]> Statistics: Posted by IceDreamer — 06 May 2014, 00:01


]]>
2014-05-05T21:04:48+02:00 2014-05-05T21:04:48+02:00 /viewtopic.php?t=3759&p=72305#p72305 <![CDATA[Re: TML/SML building bug]]> Statistics: Posted by errorblankfield — 05 May 2014, 21:04


]]>
2014-05-05T19:37:49+02:00 2014-05-05T19:37:49+02:00 /viewtopic.php?t=3759&p=72303#p72303 <![CDATA[Re: TML/SML building bug]]>
Question is, have anyone of you spent time verifying / falsifying my claim or is that too much to ask? You guys sounded so sure you knew how / why the launcher stalled like they did (buildtime / buildpower / balancing blabla). Did you just theorycraft it or did you find that out by testing? In my view you're the ones claiming unicorns by just make up statements about how the stalling works... I tested your claims and they were wrong

Statistics: Posted by Crotalus — 05 May 2014, 19:37


]]>
2014-05-05T18:45:17+02:00 2014-05-05T18:45:17+02:00 /viewtopic.php?t=3759&p=72297#p72297 <![CDATA[Re: TML/SML building bug]]>
I made a statement, it's your job to prove it wrong.

See how that's backwards?
Not being rude here, but your missing a few key details about the scientific method. Namely the bit that puts the burden of proof on the accuser so we don't have to worry about nonsensical assertions such as mine.

Statistics: Posted by errorblankfield — 05 May 2014, 18:45


]]>
2014-05-05T12:41:28+02:00 2014-05-05T12:41:28+02:00 /viewtopic.php?t=3759&p=72273#p72273 <![CDATA[Re: TML/SML building bug]]>
But ok, get some T3 Pgens and a nuke with 1 mass income. See what happens with mass drain. You can also laborate with mass income / mass drainers to see the different throttle steps it will do. If you're using -10 gamespeed it's more easily spotted.

Statistics: Posted by Crotalus — 05 May 2014, 12:41


]]>
2014-05-05T12:39:23+02:00 2014-05-05T12:39:23+02:00 /viewtopic.php?t=3759&p=72272#p72272 <![CDATA[Re: TML/SML building bug]]>
Crotalus wrote:
The thing you all overlook is that nuclear silos throttle themselves if they dont get enough mass. For instance, if a silo requests 100 mass per second but only gets 10 mass it will lower it's mass drain from 100 to 90 next tick. This will continue until it fills it's 100 mass, then progress bar is increased one step and the cycle repeats. But if doesn't fill it's share before it reaches 0 mass drain this means it will stall indefinetly until there's mass in store.


It makes sense, however... Proof?

Statistics: Posted by IceDreamer — 05 May 2014, 12:39


]]>
2014-05-05T12:17:42+02:00 2014-05-05T12:17:42+02:00 /viewtopic.php?t=3759&p=72271#p72271 <![CDATA[Re: TML/SML building bug]]>
DS = 1 / VDS

Now, when it comes to structure builders/upgraders like silos. and shields.

The mass is distributed evenly depending on how much mass a unit is requesting. If there is 100 mass available and one unit is requesting 200 and the other 100 the distribution will be ~66.6/33.3.

The thing you all overlook is that nuclear silos throttle themselves if they dont get enough mass. For instance, if a silo requests 100 mass per second but only gets 10 mass it will lower it's mass drain from 100 to 90 next tick. This will continue until it fills it's 100 mass, then progress bar is increased one step and the cycle repeats. But if doesn't fill it's share before it reaches 0 mass drain this means it will stall indefinetly until there's mass in store.

Lets see if one of you rocket scientists can come up with the solution, hint: This is probably same issue as the infamous mex upgrade engine bug which occured at mass stalls

Statistics: Posted by Crotalus — 05 May 2014, 12:17


]]>
2014-05-05T11:00:59+02:00 2014-05-05T11:00:59+02:00 /viewtopic.php?t=3759&p=72267#p72267 <![CDATA[Re: TML/SML building bug]]> Statistics: Posted by IceDreamer — 05 May 2014, 11:00


]]>
2014-05-05T06:23:52+02:00 2014-05-05T06:23:52+02:00 /viewtopic.php?t=3759&p=72260#p72260 <![CDATA[Re: TML/SML building bug]]>
Second, lol for yelling at zep and wtf for zep not even caring.

Thirdly, I think I handled the semantics of 'priority' well enough already. But I'll summarize it here...

The game doesn't prioritize anything. It has a function that deals with who get's what mass somehow based off build rate(power), build time, and current mass/energy.

Different combinations of build rate and time yield better results for getting built under a shortage -generally high build rate with short build time. In this sense, there is priority but that's not semantically how we think of priority. This is bottom-driven, what we are currently building decides who gets what. To say there is priority implies the opposite, a top-driven approach. Where the game says 'oh, factories need mass first, screw TML/Nukes' or w/e heuristic you prefer.

Now resume squabbling over the uncertainty. I'm working on reverse engineering the code myself (by making a working model). I just wonder if it will be faster to guess a few logical approaches or try and incorporate my limited knowledge from my signals and systems classes to calculate the impulse response of the function mentioned above. Not sure if you even can do this seeing as we have 4 inputs and only one output.
Bleh.

Statistics: Posted by errorblankfield — 05 May 2014, 06:23


]]>
2014-05-05T02:03:35+02:00 2014-05-05T02:03:35+02:00 /viewtopic.php?t=3759&p=72257#p72257 <![CDATA[Re: TML/SML building bug]]>
Anaryl wrote:
How the engine prioritize the resources is in the engine.

You don't have access to it, so you are assuming that it's using the buildcost to do so.

I'm not saying that you are right or wrong, I'm just saying that copy/pasting the BP proves nothing as the resource system is not in LUA.


I'm not assuming anything. If you had bothered to read the post you would see that the numbers I quoted were functions of existing build characteristics with which players could test for results in sandbox.

As you do not have the source code either, how are you sure there is any prioritisation going on? You're saying I'm only assuming - but you're assuming that there is some kind of prioritisation in-engine. To which I say - well if you know something we don't - please share it with us.

All my numbers do is give you some tools to infer from your results *how* the mass is being prioritised or at least a reasonable approximation thereof. From there players can work out from testing what is actually happening behind the scenes.


Except that all the relevant code was not given, because we don't have access to a massive amount of the code which makes this game tick, hence PilOt's responses. Thanks to this fact, testing top trumps code in every single aspect of SupCom. The code is irrelevant. If something is happening in game, that's what happens; not being able to understand which code is causing something is beside the point, because we can't see it all.

Testing > Code


So you're arguing that because there isn't code that supports your assertion it must simply be missing? Not that it doesn't exist, it's just yet to be found, and exists in a place where it's existence cannot be either proven or disproven. Say, you wouldn't be a Christian would you?


Testing > Code


Come back when you've done something more technically proficient than updating the code comments.

Research>Code>test>Research>test>Code>test>research>code - it's kinda obvious when you say test first code after that you have never programmed before.


Lol. Calm down Anaryl. It's too funny that the mere insinuation that you don't know something has you completely unhinged. At what point did I "ask you to educate me" or call anyone's argument "invalid"? I'm stating what occurs while actually playing the game. You know, the one I dominate you in. :twisted:

Being a forum queen is all well and good but in this instance it appears you don't have the information available to do even that; you're merely making your own assumptions. Let me know when you have a replay that contradicts anything I've said.


If you'd read my posts you'd see it's all there. You sya "OMG you don't know anything and are making assupmtions" yet all the data is there from which you can infer res distribution. It's not my fault if you TL;DR'd it. But don't complain there are no answers - they're right there

Statistics: Posted by Anaryl — 05 May 2014, 02:03


]]>