 |
| Will they ever release me from this torment? |
 |
| Children, this is your fate if you try a life of crime |
Two times every day I had a chance to enter a canteen and take some water and bread rolls filled with sausages. Either I’ve encountered another bug or my character is very badass, since I got him to drink only one bottle of water and to eat only one bread roll during the whole time. He did munch all the sausages, so perhaps he is just a dedicated carnivore. At the end, my inventory was filled with uneaten bread and water.
Once when I was enjoying my meal, couple of the inmates started to argue with one another. “Plot point!”, the game was crying, so I clicked the new hotspot. I was taken to a holding cell for the rest of the day and I got no more food. I could chat with a fellow inmate and exchange a bread roll for a pain stick. After one day had passed, I was transferred back to my own cell.
Once every day I could enter the arena and do a single fight. At first, I was meant to battle with a robot to show my worth as a gladiator. The interface was pretty simple - I could go left and right and hit. After a few training sessions, I found a good tactic - go to one end of the screen, hit droid when it comes, quickly go to the other end and repeat.
 |
| Just one more hit and I’ve done it. |
After two victorious battles with droid, I received a right to go to real battles.
 |
| In a quite cool fashion, a digitised actor can be heard saying “Ave Caesar, morituri te salutant”, just like in real gladiator fights. The developers sure knew their classic history |
 |
| ...or then they just read Asterix |
The gladiator battles weren’t that different from the fights with the droid. There were just three different enemies, rotating in the same order - one with a sword, one with a spiked steel ball and one with a trident. The trident fighter was the toughest, but only because the animation with the trident coming down screwed the movement of my character and I couldn’t really dodge the hits. That problem I solved simply by lowering the DOSBox cycles, and after that, none of the fights were any problem.
 |
| I don’t really understand why these are called technoglads, when there’s nothing futuristic in their battles. |
But despite how many fights I went through, nothing happened.
I am quite embarrassed to say what the solution to this problem was, because on hindsight it was quite obvious. You see, after every successful battle I got to choose whether to spare my opponent or not. At first I hadn’t paid that much attention to the proceedings, but I should have followed the demands of the crowd.
 |
| The audience are putting their thumbs down |
 |
| No mercy it is then |
After I caught on to this, I needed only few victories to get the plot moving.
 |
| Deus ex machina |
It is always a bit of a letdown in a graphical adventure game when big plot developments are presented in a block of text. This is even worse, when the plot developments feel abrupt and unbelievable. And typos don’t really help here.
I was taken to the king of Shedishan, who was supposed to grant me mercy due to my success in arena. It had been hinted throughout the game that the king had been acting strangely for some time, and it was then no wonder that he didn’t follow the opinion of the crowd and denied mercy from me. A block of text told that in rage I stroke him with my pain stick. Surprise, he was just a robot planted by Koshan to get monopoly on deeds to Bedhin 6!
I was struck by a guard, and after a couple of days of unconsciousness I read another block of text telling me that Shedishan had almost fallen into an anarchy, if it hadn’t been for the religious leaders taking control of the day-to-day politics. Under Shedish laws I now had the right to become the king of Shedishan. Koshan family had managed to retain their position by blaming one of their underlings for the robot affair.
 |
| My final task |
Next, I was contacted by B.A.T., which had a new plan. My death was supposed to be faked and blame of it was to be laid upon Koshan Inc., which would then finally be kicked out of Shedishan. I would just first have to take a contract made with my hired killer to the vault of Koshan. This sounds really convoluted. If all B.A.T. needs is a controlling portion of the Bedhin 6 deeds, why couldn’t I do what my (fake and mechanical) predecessor did and create some new shares I could then gift to a representative of B.A.T.? Oh well, I’m supposed to get myself killed, so let’s get on with it.
 |
| A nice detail - everyone loves king, which makes stealing that much easier |
Getting in Koshan building had been a problem before, but now I had a fairly good idea how to get in.
 |
| All it takes is some acid |
 |
| And there’s the vault! Now I’ll just open it… |
 |
| ...or not |
 |
| OK, let’s first go to this other room and turn that switch to turn the alarm system off |
 |
| And THAT was the easy part! |
Why do the French have to put practically insolvable puzzles at the very end of their games? I mean, Maupiti Island did it, Fascination did it and now Koshan Conspiracy does it.
 |
| You are sure it’s all over |
I’ve tried to bang my head on this one, but I just don’t have any idea what is going on here. Even my Request for Assistance didn’t help, although Voltgloss bravely offered some assistance. I’ve thus decided to do the only possible thing and declare this game Lost. According to walkthroughs, a) this appears to be final puzzle in the game and b) no one has any idea what is actually going on here, although some have solved it by brute forcing. Alas, I am unable to do even that, since every button I press just kills me. It’s a pity, since I’d have loved to at least see the ending of the game. Oh well, I guess I have nothing else to do now but to rate the game.
Session time: 4 hours
Total time: 33 hours, 45 minutes
Congratulations, or commiserations, or whatever you say to someone who's lost despite a valiant effort.
ReplyDeleteThanks! I am not sure how valiant it was, but effort it did take.
DeleteI have no idea about the solution but I did notice something in one of those images:
ReplyDeleteVf gurer n oht(va-tnzr) va gung frpbaq-gb-ynfg vzntr? Ng gur obggbz bs gur 2aq pbyhza bs flzobyf, gurer ner gjb Oyhrf tbvat vagb na =, juvpu fubhyq lvryq na Benatr, evtug? Ohg vg lvryqf n oyhr.
Indeed, this is one of the two anomalies in that image (the other one occurs at the bottom of the first column on the left). At first I thought I was meant to correct the setup, so that even the bottom rows would work correctly, but that appears to have been a false lead.
DeleteVg ybbxf gb zr nf vs <> vf qvssrerapr (KBE va ovanel tngrf) naq = vf fnzr (AKBE) jvgu benatr ercerfragvat 1 sbe lrf naq oyhr 0 sbe Ab. Va guvf pnfr fryrpgvat rvgure 13 be 14 ba gur yrsg fubhyq pbeerpg gur jubyr frdhrapr nf vg jvyy znxr gur bhgchg sbe guvf tngr 1, juvpu jvyy gura pbeerpg gur jubyr frdhrapr. Vs jr pna unir n srj zber rknzcyrf nf jryy nf juvpu yvtugf pna or fryrpgrq vg zvtug or penpxnoyr. Sebz jung V unir ernq va gur jnyx-guebhtu gung jnf yvaxrq vg ybbxf yvxr n fhpprffvba bs n srj bs gurfr vf arprffnel
DeleteThe above is what I could figure out from binary logic, with the game considered lost I am unsure if ROT13 was still necessary but rather safe than sorry.
DeleteI think at this point ROT13 is unnecessary. I added a few screenshots, where the lights on the right are different. If you can discern any possible way how the puzzle is supposed to work, feel free to let us know.
DeleteThanks! It looks like <> mean are the inputs different and = means are the inputs same, with RED indicating YES and BLUE indicating NO. If you look at the screenshots you will see on everyone there is a pair of the inputs on the left. In all the screens above it is always the 15/16 pair giving the wrong answer, but changing them both to RED will still give the correct inputs for the equation.
DeleteAlternatively changing 13/14 on the left will also satisfy the equation.
Another idea has hit me, what if the point is to get all the lights on the right-hand side RED? In that case, you will have to just change 1 of the input pairs (on the left) for each. So if The top light is blue changing any input from 1-4 would make it red, then if e.g. the 3rd light is also blue changing any input from 9-12 on the left will also change it to red?
DeleteI noticed in your game-over screen one the input you were changing would have resulted in the last LED to change from RED to BLUE which would maybe have tripped the alarm.
I now tried to test the "all right lamps to RED hypothesis", but to no avail - I still got a game over. So, it's at least not the whole solution.
DeleteI didn't speak about this in the blog text, but what makes this puzzle really infuriating is that testing possible solutions is really cumbersome. Firstly, the right-side lights keep changing randomly after every few seconds, which makes timing your move difficult. Secondly, every false move is followed by a game over and DOS prompt, without any chance to load your game in between, forcing you to begin the game anew (watching some animation etc.), if you want to try the same puzzle again.
Would so love to see you finish this! Curious--are the 16 buttons blue buttons at the left the *only* things you can click? Can you click the logic gates (= and <>) too? Or no? What about the other lights?
ReplyDeleteYes, the 16 lights are the only hotspots in that screen, so the other lights or logic gates won't help here.
DeleteIf you are interested of B.A.T.II, someone, who seems to know the game inside out, is doing a Youtube series on it:
ReplyDeletehttps://www.youtube.com/channel/UCShXmBZGyZarJi6Lq1DN6KA
Unfortunately, the series is still in very early stages and the videos appear to be updated quite unregularly, so you might have to wait for a long time, before the series will reach the end sequence.
I am also guessing that the version played in the videos is different, since there are slight differences - for instance, the senator who should have owned deeds has actually deeds to buy in his inventory.
Yes, I ran across that too; pity that person isn't farther along.
DeleteI've spent a fair amount of time staring at the pics you've posted and trying to figure out the pattern. The bottom left definitely seems like the outlier. But I think it's a tad hard to parse from static screenshots. Is there any possibility of you posting a video that shows how the lights change over time, what happens when you press buttons, etc.? Maybe that would help us crowdsource a solution.
I did a short video showing how the lights change and where all the relevant hotspots in the screen are (16 lights and an exit arrow). I do otherwise nothing but move the pointer around the screen and when I finally press one of the lights, I get a game over:
Deletehttps://youtu.be/jgBEb4gT9G4
Thanks--so here's my guess:
DeleteThe final output seems to be the four lights on the far right. The solution can't be to get them all blue, since all blue is one of the combinations it cycles through. But it doesn't seem to cycle through all orange. I wonder if that's the winning combo.
If it were me, I'd try to push a button during one of the cycles where there are three orange buttons on the right. I think I'd start with what appears at 1:30 in the video (from top to bottom, OBOO). It seems like if, at that point, you pushed the 8th button down from the top (or maybe the 7th, 6th, or 5th, too), the colors and logic gates would change so that the one blue button on the right would change to orange.
If that didn't work, perhaps the same kind of strategy could be tried with other combinations where there's one blue button on the right--e.g., at 0:33, trying one of the bottom four buttons, although that seems a tad dicey given the logical weirdness going on at the bottom left corner of the circuit board).
Just thinking out loud here... who knows, this is indeed a beast of a puzzle! So many possible combos it'd take awhile to brute force them all.
I've been now trying to take advantage of the moments, when there's only one blue light on the right, but again, only response has been game over. I would still point out that cycle of light switches is completely random - if I load a save game, the cycle of switches is different than on the first time.
DeleteI just thought of a possible answer, based on the following comment from the walkthrough I found and was citing in my ROT13'd comments last time:
ReplyDelete"Even when the top four lights show the same combination as the last time you tried it, clicking the same blue light can produce different results. In my case, I saved the game with the top four lights showing OBBB (as shown in the picture). If, when I reloaded my game, the next change was to BBOB, clicking the 11th blue light from the top would win the game. However, if I waited until BBOB came up again, the 11th blue light would then not win me the game, but usually kill me instead."
Thinking about this, I'm wondering if all the logic gates are red herrings, and what you're REALLY supposed to do is this:
1. Look at the four changing lights. Treat blue as 0 and orange as 1. Convert from binary to decimal. So, OBBB = 1000 = 8.
2. Wait for those lights to change. Again treat blue as 0 and orange as 1, and convert from binary to decimal. So, BBOB = 0010 = 2.
3. Before the lights change a third time, press the blue light on the far left that, counting up from the bottom (treating the bottom as 1), corresponds to the difference between the above two numbers. Here, the difference between 8 and 2 is 6. So press the 6th blue light up from the bottom (which is also the 11th blue light down from the top - the light that the walkthrough writer used to win the game after the OBBB -> BBOB change).
The walkthrough writer also comments that sometimes pressing a blue button doesn't cause a game over, but rather changes the light setup. Maybe you need to do what I've described above a few times in succession.
That sounds a plausible explanation, but I haven't been able to make this work yet. Admittedly, it is a bit difficult to calculate the result and then push the button I'm supposed to, before lights change again, so one might think it's just due to my poor reflexes. Yet, even with DosBOX cycles lowered so much that I have no problem doing the math and pushing the button I want, I haven't managed to crack the safe.
DeleteIs it plausible to just wait for an OBBB -> BBOB cycle like the walkthrough writer describes, with the mouse ready over the 11th blue light from the top, and click once that cycle occurs?
DeleteThat was the first thing I tried, when you sent the walkthrough, so that wasn't the answer.
DeleteI am beginning to suspect that the version I've been playing is somehow bugged, making the end puzzle more difficult than it should be.
It looks like those four horizontal lights are actually inputs for the last row of logic gates. If you look at the circuitry pattern, you can see a connection between the rightmost light and the top gate; the second from the right and the second from the top; and so forth. That introduces a few additional logic errors, though, and it's still not clear what your goal is. You're hacking an alarm, so maybe you need to turn all the lights off?
ReplyDeleteThe top inputs do definitely correlate to the output on the right, you'll see that they always make logical sense, and the logic gates are the same everywhere. The problem seems to be that there is no indication as to what the outputs should be. So far suggestions have been to get all the outputs red, matching them to a number and my latest theory would be that the output may have to match the input. The problem being that every-time an input is selected it is an immediate game-over and thus it is nearly impossible to work out what the game wants from the player. I can see how this has been driving the poor author mad...
DeleteYou could try using a version of DOSBox with save states to dramatically shorten the trial-failure-reload cycle.
ReplyDeleteYes, I've moved on to one of those, but it's still quite slow. Frankly, I am not sure whether this puzzle really deserves all the time I've spent on cracking it.
DeleteI'm reminded of BAT 1's blind maze, where one wrong button press made the screen stop working. I wasted so much of my life on that thing..
ReplyDeleteThat was more of a bug, though? This, I don't really know, if it is solvable and I just haven't been able to find the solution or if there's some bug making it harder than it should be.
DeleteSo, I solved the puzzle, at least semi-reliably. Here's the description of what I did:
ReplyDeleteSvefg, vs lbh'er jbexvat sebz Vyznev'f fnir, lbh zvtug gel rkvgvat gur fperra. V jrag onpx bhg gb gur bhgfvqr (gur ragenapr gb gur Xbfuna gbjre) naq jrag onpx va. Vg frrzrq gb jbex orggre gung jnl.
Gur xrl vf gur yvtugf ng gur gbc, juvpu fubhyq or vagrecergrq nf n urknqrpvzny ahzore. Fb sbe gur yvtug ba gur sne yrsg, na benatr = 8; gur arkg bar gb gur evtug, benatr = 4; gura sbe gur arkg, benatr = 2; gura, svanyyl, benatr = 1. Fb, n srj rknzcyrf:
BOOO = 8
BBOO = 12
OBOB = 5
OBBB = 7
Jung guvf gryyf lbh gb qb vf juvpu ohggba gb chfu ba gur yrsg, pbhagvat sebz gur gbc (ohg n pnirng ba gung va n zbzrag). Fb, vs vg qvfcynlf BBOB, lbh jbhyq chfu gur frpbaq ohggba sebz gur gbc. Be, BBBO, gur sbhegrragu ohggba sebz gur gbc. Tvira gung gur yvtugf punatr snveyl encvqyl, V znqr hfr bs gur cnhfr shapgvba va QbfObk gb tvir zlfrys gvzr gb qb gur pbairefvba, pbhag qbja gb vqragvsl gur ohggba, naq gura V hacnhfrq gb cerff vg.
Vs lbh qb guvf pbeerpgyl, gur ohggba lbh chfu jvyy punatr pbybe. Gura nabgure urknqrpvzny frdhrapr jvyy nccrne. Qb vg ntnva. Naq ntnva. V'z abg fher vs gurer'f n frg ahzore bs gvzrf lbh arrq gb qb guvf, ohg riraghnyyl, gur fnsr jvyy bcra, lbh chg gur pbagenpg va vg, naq gur tnzr raqf.
Va bgure jbeqf, gur jvaavat cnggrea frrzf gb or qvssrerag guna gung qrfpevorq va gur jnyxguebhtu jr nyy jrer pbafhygvat ng FanexQernzf. Vg'f frireny ohggba cerffrf gung ner arrqrq va beqre, abg n "zntvp" bar pyvpx.
Naq fb gur pnirng: Gur svefg gvzr V qvq guvf, V npghnyyl pbhagrq sebz gur obggbz bs gur fperra, engure guna gur gbc. V.r., sbe OOBB, V uvg gur guveq ohggba sebz gur obggbz. Naq gung jbexrq. Ohg gura gur arkg gjb gvzrf V qvq vg, V pbhagrq sebz gur gbc, naq gung jbexrq. V'z abg fher jung qrgrezvarf jung qverpgvba lbh pbhag sebz. Vs nalbar unf nal vafvtug ba gung, yrg zr xabj.
Ubcr guvf znxrf frafr. V zvtug gel gb znxr n ivqrb bs vg naq cbfg vg yngre.
Here's the link to the video showing the solution: https://www.youtube.com/watch?v=7KBonyguS7E&feature=youtu.be
DeleteYou also see the beginning of the ending sequence. There's a bit more after that, but not much (a rather unsatisfying ending IMO).
A note regarding the video: Va guvf pnfr, vg bayl gbbx zr guerr pbeerpg pyvpxf gb haybpx gur fnsr; va zl cerivbhf gjb ehaf, vg gbbx frireny zber. (Creuncf guvf vf jul gur Fanex Qernzf jnyxguebhtu unq bar zntvp pyvpx? Znlor gur frdhrapr pna or nf fubeg nf bar?)
One more comment: Gur = naq <> ybtvp tngrf frrz engure fhcresyhbhf gb fbyivat gur chmmyr--fb creuncf Ibygtybff vf evtug gung gurl ner n erq ureevat.
DeleteGreat job, Tornado! So you had to exit the screen and come back to make the solution more reliable? No wonder I got just blanks with my experimentation. I wonder if it was an intended feature that saving on the puzzle would make it more difficult.
DeleteAnd I truly wonder how anyone solved this puzzle back in the day without the DosBOX Pause command! Counting the right button in the limited time available is a challenge.
Thanks Ilmari! I agree, without the pause command, it seems like the player has to do pretty quick math. But I also think there isn't really a time limit on how long you have, so the player might be able to just wait for a number he or she recognizes to come along.
DeleteI didn't think about the possibility that the game nerfs saves during the puzzle. That seems plausible, if mean. Then again, it's just plain mean to have a difficult and somewhat obtuse math puzzle at the end of a game that, as you noted, didn't have much puzzles of any kind up to this point. Certainly an example of bad game design--not building your player up to the challenge. If this puzzle had appeared at the end of a game with content and structure like Castle of Dr. Brain, it might have been more sensible. Maybe.
I went ahead and uploaded two videos that show the rest of the ending, in case anyone is interested (apologies if there's an easy way to combine these into a single video; DosBox spit them out as separate videos and I'm a novice YouTube user):
https://www.youtube.com/watch?v=ip1N9WyPlFo&feature=youtu.be
https://www.youtube.com/watch?v=ykVRXo8MvSg&feature=youtu.be
... and I caught an error in my description of the puzzle, for the combination I said should be interpreted as 2. Here's the correction: Jung guvf gryyf lbh gb qb vf juvpu ohggba gb chfu ba gur yrsg, pbhagvat sebz gur gbc (ohg n pnirng ba gung va n zbzrag). Fb, vs vg qvfcynlf OOBO, lbh jbhyq chfu gur frpbaq ohggba sebz gur gbc. Be, BBBO, gur sbhegrragu ohggba sebz gur gbc.
DeleteAnd one more correction since it's bothering me: V qrfpevorq gur yvtugf ng gur gbc nf vaqvpngvat n urknqrpvzny ahzore; ernyyl, vg'f zber npphengr gb fnl gur yvtugf ner va ovanel (benatr = 1 naq oyhr = 0), jvgu gur sbhe ovanel qvtvgf ercerfragvat bar bs 16 ahzoref, juvpu pbhyq or vagrecergrq nf bar urknqrpvzny punenpgre.
DeleteSorry for the minor tweaks here; I was trying to get this up quickly in my initial post, partly due to limited time, partly due to excitement. :)
So, when you are saying that this solution is just mostly reliable, could it be that this puzzle is like binary version of Simon, so that you would have to "repeat" a correct sequence of lights? That would explain why the solution doesn't work very well, if you try to do it in my save, without going away for awhile - if I have saved the game at the middle of some sequence, then just finding the correct light corresponding to the next pattern of lights wouldn't work.
DeleteI sat down with the puzzle again tonight, and now I understand it much better. It is actually less complex than I made it out in my comment above. To summarize:
Delete1. Gur chmmyr vf ba n gvzre gung fgnegf jura lbh ragre gur Xbfuna gbjre. Va lbhe fnir fgngr, gur gvzre unf eha gb gur cbvag jurer vs lbh pyvpx ba n ohggba, lbh'er pnhtug. Fgrccvat bhgfvqr gur Xbfuna gbjre erfrgf gur gvzre.
2. Gur chmmyr npghnyyl qbrfa'g vaibyir pbairegvat ovanel. Vg'f whfg n znggre bs trggvat nyy gur ohggbaf ba gur evtug gb or benatr. Gur gvzre vf cebonoyl va cynpr gb tvir lbh n yvzvgrq nzbhag bs gvzr gb qb vg (naq znlor sbe n frafr bs ernyvfz).
Ilmari, would there be any possibility of me writing a guest blog post explaining the puzzle and the end of the game? I do have a Blogger account. Not sure if that's breaking protocol around here. I'd just like to see a post that explains this puzzle, given that the few Internet sites that describe it, describe it as a mystery, when it isn't--it does have a rather simple set of rules that govern it. If future adventurers get stuck on this, or if they're just mystified by it and want more detail, it'd be helpful to have a place that sets the record straight. Of course, if you'd rather write such a post, I'm happy to see you do it too--after all, you're the one who's suffered through 99.9% of the game. I'd just like the solution to the puzzle to be in the public Internet and not hidden behind ROT13.
I think that since you were the person finding the correct solution, it is just fair that you shall write the post for it. Assuming you have a Google account, it's easiest, if you upload the text file of the post and the screenshot files to be used to Google Drive folder and share that folder with adventuregamer@googlegroups.com - also, indicate in the text the places where the screenshots are to be placed. We'll then take care of scheduling your post on the blog!
DeleteWonderful! Thanks for the opportunity. I've started drafting it up and hope to have to you soon, hopefully by the end of the day.
DeleteJust shared it. One note, you might see one google doc with the screenshots inline, and one with the screenshots called out (as you suggested). I also made a few text edits in the called out one, so use that version (and feel free to make minor edits if you wish). Let me know if you need anything else from me. Thanks!
DeleteThanks for the post! It should come out in our blog in about 8-9 hours. If you still catch something you want to edit in the published post, just send a word and we'll fix it.
DeleteThanks Ilmari! I caught a couple of typos (see my comments on the published post). Really appreciate you getting it up so quickly!
Delete