Master of Orion 1 unofficial patch

[Ingos]

I think i found a bug in the patch

Every time i invest reserves on a colony, i double its production value, but the change is permanent

Like right now, i invested 40bc on my homeworld, and jumped from about 250+- to 500+- production, 20 turns later, i still have 500+- production

since this happened on 2 different games i think its a reproducible bug.

[RefSteel]

Okay, finished up my Impossible Klackons game. This time, I had Bulrathi neighbors, among other exciting things ... but I never did see them send transports without a fleet. I scouted aggressively enough to keep them from scouting out any of my colonies, and by the time I had a world they had scouted ... it was one of their worlds, which I took by storm, with so many toys I had a gropo advantage over them. (I won by conquest again, this time in 2450). I did test the diplo a little, and was not able to get a DoW-against-an-ally from a race that wouldn't break their treaty, so that's a good sign, but I still haven't tested it enough to be 100% sure. I also again confirmed that "C" continues in lieu of "SPACE" when enemies sabotage our factories, but neglected to confirm whether the same thing applies to sabotage by the player. I've also been keeping track of all the leaders I've encountered in games under this patch. Thus far:

Alkari: hon/exp, hon/diplo
Silicoids: agg/exp
Humans: hon/tech, pac/diplo
Mrrshans: ruth/tech, xen/mil, xen/tech
Bulrathi: err/eco, agg/eco
Psilons: err/eco, err/tech, pac/diplo
Klackons: agg/eco
Meklar: err/eco, xen/diplo
Darloks: agg/exp

I can't wait to see my first non-pacifist Sakkra!

In case I end up having time for additional playtesting soonish (not sure how that'll go) is there anything in particular you'd like me to check/review? (Also, would any more savefiles be useful, e.g. demonstrating the bug of which Ingos just provided additional confirmation, or from my Sakkra game with the miserable Psilons?)

[kyrub]

Miserable Psilons is the real deal, If I may ask. The transfer thing is easy to do, I have very precise idea where to go. Miserable Psilons is a mystery, and it's kinda threatening the whole patch. It's worrying because I did not exactly touch those routines, I did some things in the vicinity but what can create such a mess, is beyond me.

You may also try to fiddle with your previous game (if you find some space in your timetable), e.g. try to find if it is race specific (could be) - change the race in the save file and see if they find any techno. Or something else, player number specific. Or put a new planet real close to them and see. I really don't know.

------------------

I will have some time this week and I will try to put things in motion again. I expect making some test build and putting them in the AI enhancement thread.

[RefSteel]

Miserable Psilons is the real deal, If I may ask.

Does this mean you'd like to see more/all of the savefiles I have for the game? (I still have the every-turn-autosave files the patch generates, from 2362 through the end of the game.) Or were you just asking for more help isolating the problem? I'll have another look at the game when I have a chance, but a few thoughts in the meantime:

1) I agree that it's a mystery; I couldn't fathom why it would happen at first. It does occur to me however that it might not be a problem with your patch; it might be an undetected bug in the original game. Usually, when I see a hopeless AI in a normal game, my assumption is that other AIs did it in; this situation was kind of a perfect storm: A totally isolated AI in a small enough galaxy that I was able to be sure I was the first to meet them, in a game where I was specifically on the lookout for bugs and anomalies. I remember being mystified in OSG 21 by the Mrrshans' incredibly slow factory building; they won the game, of course, but only thanks to massive assistance from their Psilon Big Brothers (played by us crazy RB'ers, of course, on a deliberate mission to make the Mrrshans win!) It's imaginable that those kitties were suffering a version of the same problem the Psilons had in my test game. Conveniently, there are also periodic savefiles posted of that game, if you're interested in looking at them. (Note that if the same problem does exist in the normal version, and is too knotty a problem to untangle in the patch, the patch with that rare bug still extant would still be an improvement on the standard game!)

2) It's definitely not an "All Psilons are Lame" kind of problem, though it's still possible it doesn't affect other races (or affects them less). The Psilons in my first test did okay for themselves (until I exterminated them; it was Average!) even on Average difficulty.

3) This is the most promising possibility I can see for an explanation so far:

they were limited to Mentar itself ... Mentar was at 100 pop, 66 facs, no missile bases(!), 1 destroyer, 303 fighters, and two colony ships.

Okay, they've got no tech, and no habitable planets within 5 parsecs. What are they doing with two colony ships?! By 2366, they had a third ... and a fourth by 2368! My understanding is that AIs do not build colships, but (randomly, at times) get them for free when they have a colonizable planet in range (which is not the case here!) ... and do pay maintenance on them! I wonder if the Psilons, first stunted by their lack of secondary worlds, ended up getting crushed under the weight of fleet maintenance. I'm not sure how long that second colony ship had been there, unfortunately, but 300 fighters is also awfully costly for an under-developed single world....

4) That's all I've got for now; I may have a chance to look at the game again later today though. I unfortunately haven't developed the ability to read data from the savefile (I've leaned heavily on Sargon's help to even do the simple edits I did for Imperium 21) but I'll have a look in-game if I have time.

[RefSteel]

Apologies for the double post, but I got some additional data:

First, relating to the Not-So-Brainy Brains: There were isolated (Pacifistic Diplomat; added to list above) Psilons in my Impossible Klackons game too - in fact the only star within six parsecs of Mentar (at 6 exactly) was ... Ursa! Nevertheless, the Brains weren't nearly as hopeless as those in my previous game: By 2377, they had 28 missile bases (other AIs had 26, 33, and 34), 12 destroyers, and, yeeeeup, four Colony Ships, with literally nowhere to send them! Note that on Impossible difficulty, they had twice their Hard production, and so could more easily address whatever maintenance they may have been facing, so that might still have been the problem. By the way, how come I keep seeing isolated Psilons lately? (Starting in Imperium 10, with Imperium 19 and my Imperium 12 Shadow among the exceptions...) Pure luck, obviously, but it's kind of amusing....

Interesting side note on AI programming: They apparently do not send scouts to one another's homeworlds at the start of the game, as Mentar's are sitting at home, neither one en route to Ursa, when I use ALT-GALAXY in an autosave from early in the game. This is a good thing, at least in the unpatched game, as a Scout-in-orbit reflexive transport launch between homeworlds at that stage would ruin the balance of the game.

Other things I forgot to mention from the Impossible Klackon game: <RETURN> already does what I wanted <SPACE> to do on the election screens, so feel free to ignore that bit (unless that actually makes it super easy). I did spot an important (when it happens, which is fortunately rare) problem with one of the interface changes though. With the newly reduced clickable area for stars, it can be difficult to find a place where you can click a star at all if a fleet is "passing over" it on the main screen. There are usually other ways to look at the planet (click on it from the Map, F2 or F3 around to it if it's one of yours, etc.) but on one turn (2438, I think...) in my Klackon game, I was unable to send a fleet or transports to the Darlok homeworld! (I just tested it again, and found a space off to the upper left where I could have clicked to make this work, but it's small and offset enough that I wasn't able to find it at the time, and I'm not sure it would have worked had there been a fleet retreating from the world). Fortunately, on the turn this happened, it was immaterial; I was just trying to check range and get a look at the planet ... while my SoD and transports were already en route! It should also be noted that you can't get to the planet screen for an enemy or neutral star by clicking on its picture in the upper right the way you can with planets you own. My sales pitch for increasing the planet's clickable radius again would be that inability to click on and learn about a fleet in transit is going to crop up here and there anyway - for instance, when two fleets are at the same coordinates (usually when both are traveling between the same two stars) whereas the ability to click on a planet is much more important to the game. (End of sales pitch.)

Again, thanks for all your work on this; let me know if these data help, or what other specific things I should look for. I'll look into the ones you mentioned above when/if I have a chance, but it'll take some time for me to figure out (or ask Sargon to please help me with) a way to edit the saves properly.

[sargon0]

Sad Psilons
The savefiles show the following:

Mentar (fields are pop/pop tenths/fact/waste/sliders ship->tech/net prod/ number of bases)
date pop pt fac wst Slider%......net bas
2362 100 -7 .68 .14 4/4/.8/51/33 .81 ..0
2370 100 .0 .69 ..0 5/5/10/59/21 122 ..0
2409 100 .9 137 ..0 5/5/10/30/50 237 ..2

Empire (fields are [security%] then bc for trade/ships/bases/spies/[net prod%]/maint total/tech total/planets total/reserve/[tech focus])
date sec% trd shp bas spy net mnt tec pla rsv .focus
2362 2.4% ..0 .92 ..0 ..0 45% .99 .52 181 .451 FF
2370 2.4% ..2 .55 ..0 ..0 67% .60 .79 183 1053 Cons
2409 2.4% -16 .50 ..6 ..3 73% .87 177 325 1081 Prop

Issues
2362 save has negative pop tenths, a bug I have not seen before
ECO is minimum in 2362 but overallocated in 2370 (heavily) & 2409 (slightly) despite being full pop (all saves have no waste reduction)
2362 save has few factories caused by too many ships (free ships cause waste also) meaning 6 BC for factories and 52 BC between 6 tech fields (2 BC to most)
a large reserve has been created (from ship scrapping) but is not used, maybe saved for future colonies
tech bc in empire data is correct in 2409 but too high in 2362 & 2370. If you hit next turn & save with original game code it is 'corrected' to 39 & 37 as expected.

Some of this analysis is still speculative. Consecutive saves would be needed to confirm the situation in more detail.

The AI does not cope well with isolated starts. I speculated free colony ships might be delivered by need but this was a guess having seen no pattern. There may be some form of regular supply expecting the AI to be using them. Even so, the 3 free colony ships in 2362 save cost the Psilons 37 maint, the 303 small fighters they built cost them 54 maint!

Poor performance from the brains. They just needed to research R5, IT+10 & Dead to colonise Tundra Draconis but the AI (the AU?) is not up to it.

[RefSteel]

Issues
2362 save has negative pop tenths, a bug I have not seen before

It also looks like the Brains have pollution at their world that year; could this be related? I'll attach the save files I have for years before the Brains scrapped their massive fighter fleet in case this helps identify (or at least confirm) some correlation.

Some of this analysis is still speculative. Consecutive saves would be needed to confirm the situation in more detail.

Ask (or imply interest at least) and you shall receive! The saves attached, as noted above, are for all years from 2363 through 2367. Let me know if more would be helpful/needed.

[kyrub]

That is much more specific, pop tenths and the focus being FF. Thanks, RefSteel & Sargon!
Tonight and tomorrow I will try to look at the bug.

I will maybe pop-up with a question or two.

[LordLydon]

That's some detailed help RefSteel and Sargon0! Very interesting findings.

I did some poking around the game files myself and became obsessed with figuring out why the fleet maintenance reverted back after the first turn in Kyrub's patch.

I figured it out and I've PM'ed Kyrub the details on fixing it. Also a fix for the Meklar's maximum factory build bug. (original MOO bug) It was 2500 factories and should have been 2700.

I'm not much of a hex-editor (don't have the patience!!) but I hope this helps because I'm also a long time MOO fan!

Keep up the great work Kyrub and everyone else who's helping!

[sargon0]

Sad Psilons2
I did not comment earlier that Psilons in 2362 had 0.75% tax, changed to 0.4% then 0.7% in later saves. 2363 save shows reserve spending of 7 BC at Mentar but reserve funds dropping from 461 (not 451 from earlier post) to 370. But the big problem is from 2363-2367 where spending is in ECO and SHIPs only. Tech focus changes to Construction in 2364 and spending and ships are reallocated after 2367 due to counters reaching 0. Note Waste happens when AI is given free colony ships (ECO% is already calculated before maint suddenly rises) so I assume Psilons were given third free colony ship in 2361.

Major Issues
Ship concentrated spending - why?
tech spending withdrawal - AI cannot aford this since all 6 fields always in progress
ECO overallocation - inefficient unless unworked factories, just wasteful when full

Minor Issues
negative pop tenths - a bug
reserve spending - insufficient and reserves reduced too much
taxation - a waste