aliens

Author Topic: [Suggestion] Research Progress based on actual project cost  (Read 1417 times)

Offline Delian

  • Colonel
  • ****
  • Posts: 228
    • View Profile
Re: [Suggestion] Research Progress based on actual project cost
« Reply #15 on: August 17, 2022, 09:57:05 pm »
Genuine, ground-breaking research doesn't really work that way.
What? No? Research isn't done by shooting in the dark lol, only kids think that way. Sure, sometimes a rare eureka comes flying by coincidentally, but those are exceptions.
Normally, science is done by obtaining data, analyzing data, and then interpreting the data to come to some conclusions or discoveries. Sometimes it's just interpreting existing data. But almost always, it's done with a set goal. If you do research for a living, then you're able to make very good predictions (again, because you do research for a living) for how long each step of the scientific method will take (it's more of a science than art). Not all scientific work yields useful results, but in xcom it does heh.
You can check out the history of semiconductor manufacturing. Process nodes shrink every 2-3 years like a clockwork. Hardly any randomness there.

Anyway, it's fine.

499 scientists collectively managed to come up with the bright idea of blasting alien alloys with EM radiation, and then the other 335 went back in time, saw that they weren't needed to finish the project and decided to redo their day and study laser pistols instead? :P
That's also not... how anything works at all. You don't just... think hard for 24 hours and then at the end of day come up with "an idea", as if you had to wait until the end of the day before you were able to put the idea onto the paper.
No. You work on something, and when you finish it, you start working on the next thing. Even if it's *gasp* in the middle of the day. That's realistic. The scientists are paid to do research, not to sit around.

I've searched the forums for a research roll over feature, but couldn't find any discussion on it, so perhaps this would be a good solution to the problem then.
When a research project is done and there's overflow, you get a popup window to select which other project to overflow into. The overflow would be added to that project's mandays spent.

Offline Meridian

  • Global Moderator
  • Commander
  • ***
  • Posts: 7794
    • View Profile
Re: [Suggestion] Research Progress based on actual project cost
« Reply #16 on: August 17, 2022, 10:07:10 pm »
Dude, this borders on trolling.

You can't be serious.

There are no hard rules against trolling on this forum, so I can't give you a warning, but if you want to be taken seriously in the future, you'll need to stop grasping at straws.

Offline Delian

  • Colonel
  • ****
  • Posts: 228
    • View Profile
Re: [Suggestion] Research Progress based on actual project cost
« Reply #17 on: August 17, 2022, 10:34:42 pm »
Just to be clear, I've given up on the "show progress based on remaining effort" and "show progress based on actual project cost" features, because your explanation made sense.
However, the research roll over is an entirely different issue.

Do you remember anything said about a... research rollover feature?

I've asked you about it, but you didn't reply, so what am I to assume?

Offline Meridian

  • Global Moderator
  • Commander
  • ***
  • Posts: 7794
    • View Profile
Re: [Suggestion] Research Progress based on actual project cost
« Reply #18 on: August 17, 2022, 11:05:45 pm »
You can follow the scientific method rigorously, obtain and analyze a crapton of data, even come to conclusions... but guess what, the conclusion is usually that you didn't discover "cold fusion" yet (replace "cold fusion" with your favourite research goal).

Nobody can estimate that cold fusion will be researched in 2 years, 4 months and 13 days.

And certainly nobody will work on the cold fusion for 2 years, 4 months, 12 days and when finally researched at 3:14pm the next day... immediately switch to researching the cure for cancer or something.

Is that enough of an answer to your question?

Offline Meridian

  • Global Moderator
  • Commander
  • ***
  • Posts: 7794
    • View Profile
Re: [Suggestion] Research Progress based on actual project cost
« Reply #19 on: August 17, 2022, 11:16:21 pm »
If you want a more practical answer: we have already established that the player doesn't know and shouldn't know the true research cost of a project.
That means the player cannot possibly know if any research time was "wasted" or not.
If it helps ease your mind, just convince yourself that the game rolled a higher project cost (higher exactly by the amount of the theoretically "wasted" time).

Offline Delian

  • Colonel
  • ****
  • Posts: 228
    • View Profile
Re: [Suggestion] Research Progress based on actual project cost
« Reply #20 on: August 18, 2022, 12:34:37 am »
Right. So in conclusion, players have to choose to either be effective or efficient.
Effective - doing the right thing - focus on a single research topic you need, but waste research by overshooting
Efficient - doing it the right way - spread research over multiple topics, but not be able to quickly research the thing you need

For players who wanted to have both, well, they will have to live with the frustration.
« Last Edit: August 18, 2022, 12:38:52 am by Delian »

Offline Juku121

  • Commander
  • *****
  • Posts: 713
  • We're all mad here.
    • View Profile
Re: [Suggestion] Research Progress based on actual project cost
« Reply #21 on: August 18, 2022, 01:16:22 am »
You confusing the R and the D in R&D. Doing all the menial parts of a scientific project is no guarantee of success. Unless you're after a specific, well-founded application of existing theory, in which case you're already in the D camp.

X-Com (the game) doesn't really distinguish between these two, so I suppose you're right to a degree.


As to there being two substantially different choices you can make regarding research, that's what games are all about, are they not? Making interesting choices.


Anyway, if your quality of life suffers so much from this, just peek into the save.

Offline Meridian

  • Global Moderator
  • Commander
  • ***
  • Posts: 7794
    • View Profile
Re: [Suggestion] Research Progress based on actual project cost
« Reply #22 on: August 18, 2022, 09:14:15 am »
Right. So in conclusion, players have to choose to either be effective or efficient.
Effective - doing the right thing - focus on a single research topic you need, but waste research by overshooting
Efficient - doing it the right way - spread research over multiple topics, but not be able to quickly research the thing you need

For players who wanted to have both, well, they will have to live with the frustration.

Yes, absolutely.

(Infinitely) scalable efficiency is anyway just theoretical. If 1 scientist is able to discover a COVID vaccine in 200 days, 400 scientists are not going to do it in half a day (not even a million scientists are going to do it in half a day).

And I repeat again, the frustration is SELF-INFLICTED. If you couldn't look into the save file (which is cheating = looking into the future) and if you couldn't reload older saves (which is also cheating = time traveling into the past), you wouldn't be able to know that in the background you're overshooting some arbitrary number you're not even supposed to know about. As they say, you don't wanna know how the sausage is made.

Offline nooblord

  • Sergeant
  • **
  • Posts: 30
    • View Profile
Re: [Suggestion] Research Progress based on actual project cost
« Reply #23 on: August 18, 2022, 10:27:02 am »
Jesus Christ, Meridian,  the amount of time you spend in replying of these posts are astronomical.  I wonder if your time could be better utilised elsewhere than to catch up in these troll fights,  if I might classify them from my observation.
The usage of your personal time efficiently is very important to you? If I am not mistaken?

Offline Delian

  • Colonel
  • ****
  • Posts: 228
    • View Profile
Re: [Suggestion] Research Progress display improvement
« Reply #24 on: August 18, 2022, 12:45:06 pm »
you don't wanna know how the sausage is made.
Right. To keep things realistic, information on the actual research project cost should never be revealed to the player.

Let me, umm, circle back a bit to the topic of research progress ratings.
I think most players would agree that the existing research progress ratings are... lackluster to say the least. But how to improve them without giving the player any new information?

Currently, after a research project passes the unknown phase, the player can use the progress ratings to estimate the project's base cost. Just the base cost, not the actual cost. The base cost doesn't tell the player much, but it's still useful information. So let's assume that, if a project passes the unknown phase, it's not considered cheating for the player to know the project's base cost at that point, or for the game to use the base cost information.

With that in mind, I propose the following QoL feature:
In the Current Research window, add a new column (or modify the PROGRESS column) to include information on "chance to finish the project".

A few examples:
Base cost=100, spent=0, allocated=50: Chance to finish shows: UNKNOWN
Base cost=100, spent=34 allocated=10: Chance to finish shows: 0% - POOR
Base cost=100, spent=50, allocated=10: Chance to finish shows: 10% - AVERAGE
Base cost=100, spent=50, allocated=50: Chance to finish shows: 50% - EXCELLENT
Base cost=100, spent=100, allocated=10: Chance to finish shows: 20% - GOOD
Base cost=100, spent=100, allocated=50: Chance to finish shows: 100% - EXCELLENT

This is something that the player can already calculate on their own, no new information is revealed to them. It would simply make it more convenient due to the calculations happening automatically.

If, for instance, the player knows that the base cost is 20, then they can calculate that the actual cost is between 10 and 30. So if they spent 10, then each subsequently allocated scientist would give them a 5% chance to finish the project. If they spent 20, but the project hasn't finished yet, then they can calculate that the actual cost must be between 21 and 30, and each allocated scientist at that point would provide a 10% chance to finish the project. Again, any player can already calculate this, but it would be nice if it was more convenient.
« Last Edit: August 18, 2022, 12:50:21 pm by Delian »

Offline Juku121

  • Commander
  • *****
  • Posts: 713
  • We're all mad here.
    • View Profile
Re: [Suggestion] Research Progress based on actual project cost
« Reply #25 on: August 18, 2022, 01:30:02 pm »
While that is all true, this '%' is not an actual probability (since the actual project cost is already fixed) and will mislead players into thinking it is. We already have accuracy percentages that aren't, no need to double down on that.

Also, what's the supposed benefit to the player here? Are they going to try gaming these 'odds' to reduce wasted scientist-days? That's basically enticing them to gamble while the cards are rigged.

Jesus Christ, Meridian,  the amount of time you spend in replying of these posts are astronomical.
Meridian secretly likes to be wall-of-texted by argumentative internetizens. :P

Offline Delian

  • Colonel
  • ****
  • Posts: 228
    • View Profile
Re: [Suggestion] Research Progress display improvement
« Reply #26 on: August 18, 2022, 02:21:30 pm »
this '%' is not an actual probability

The % is the actual probability, both theoretical and practical.

In case base cost=100, spent=50, allocated=50, the chance to finish is 50% - what else would it be? There's exactly a 50% chance that the actual cost was rolled between 51 and 100, and a 50% chance that the actual cost was rolled 101-150.
If the cost was rolled to be 100 or under, then the project is finished.
If the cost was rolled to be above 100, then the project doesn't finish. In that case, the next day you'd be shown a 100% chance to finish with 50 scientists allocated.

The benefits to the player are:
- practicality in a sense that they wouldn't need to do the calculations manually
- it would allow the player to easily spot any projects with too many assigned scientists (spent + assigned > 150% base cost)
- it would make the progress ratings more informative

Offline R1dO

  • Colonel
  • ****
  • Posts: 412
    • View Profile
    • Email
Re: [Suggestion] Research Progress display improvement
« Reply #27 on: August 18, 2022, 02:56:27 pm »
Right. To keep things realistic, information on the actual research project cost should never be revealed to the player.

I believe you've misinterpreted the paragraph the quote (.. sausage ..) originated from. This is not about having realistic changes to that screen.
It is about the lack of desire to deviate from vanilla representation/behavior just to cater to a subset of players that want absolute efficiency when it comes to scientist resource allocation.
Especially since there already is an information source available indicating to a player it would be wise to assign lots (or as few as possible) scientists for a given project: The tech viewer difficulty indicator upon middle-mouse click on a (allocated) topic, just in case you were not aware.

...
I think most players would agree that the existing research progress ratings are... lackluster to say the least. But how to improve them without giving the player any new information?
...
This is probably the core of frustration for this topic. You're so invested in this idea that you're even assuming this is something everybody struggles with. The first (and 5th) reaction on you're opening post already stated this is not the case (or at least not to the level you are assuming).

Luckily in the opening post you've already indicated being capable of altering the source code to change behavior to your liking.
There is no shame in compiling your own OXCE flavor that includes those changes (on the contrary: it shows programming skills). I would recommend to keep a separate branch around though, for easy rebase (or merge, depending on your preferred git branch management style) when a new OXCE version comes around. That way even if this topic leads to a OXCE inclusion rejection you can still enjoy your personal play style.

Offline Juku121

  • Commander
  • *****
  • Posts: 713
  • We're all mad here.
    • View Profile
Re: [Suggestion] Research Progress display improvement
« Reply #28 on: August 18, 2022, 07:34:33 pm »
The % is the actual probability, both theoretical and practical.
It's the probability of you being in a situation described by the estimate. But your situation will not vary no matter how you allocate your scientists. Many (perhaps most) players will think the probability is the probability of finishing next day, which it is not. Since the actual cost is already fixed.

It doesn't help your case that you're calling it 'chance to finish' yourself.

Benefits 1 and 3 are still in the 'self-inflicted' box for me. Benefit 2 is something I could get behind if everything else wasn't stacked against the feature request.