Stuck on SotBE:Barag Gor (beta)

[Maiklas3000]

Son of the Black Eye scenarios have been modified in the development version. I'm on 1.11.12, and I'm stuck on the new Barag Gor on nightmare difficulty.

The basic situation is that you need to rescue 3 Orcish Shamen from a besieged castle and defeat all 4 Elvish leaders, who will unleash a whirlwind of hurt with hordes of Elvish Archers and Fighters. You start with your leader (mine is L2), Gruu (L2 Troll), and a scout (mine is a L3 Direwolf Rider.) There are next to no villages and you have no possibility of getting healers. Your roster is of no help, except maybe if you have leveled Wolf Riders, because you have no ability to heal anyone. This all sounds pretty dire, however you have just acquired the ability to recruit Troll Whelps. Help me Troll Whelps, you're my only hope.

Healing is the key to victory in most scenarios, and in this scenario that means trolls. My basic strategy has been to recruit almost exclusively Troll Whelps with my leader and form a line in the hills in the northeast corner. Meanwhile, the Shamen can easily escape the castle and make their way to that corner. In the castle I also have another leader (a Slayer), and I have him recruit Grunts and Wolf Riders, blowing his gold all on the first round. The castle force is doomed, and must either fight a delaying action or flee or both. The northeast enemy leader falls relatively easily. Sometimes I'm able to sneak a small raiding party down to the southeast leader, but so far not enough to take him out. Meanwhile, my main force retreats to the northeast hills to await the main assault wave.

Eventually, I have a fine crop of L2 Trolls, but by then the turns are running out, and there are still an annoying number of incoming Elves keeping my L2 Trolls at bay. So, holing up in the northeast corner seems not to be a winning strategy on nightmare. The screenshot below is on turn 21 of 24, which obviously means I wouldn't be able to kill the three remaining leaders in time. Don't be fooled by the blue dots... those are my main force and are under my control, alternately with the green dot units.

Any ideas?

[mattsc]

Hi Maiklas3000,

Thanks for the feedback. I am going to do it this way in case somebody does not want to see this:

Spoiler:

You pretty much answered your question yourself:

So, holing up in the northeast corner seems not to be a winning strategy on nightmare.

While playing defensively is often a good strategy, this scenario is (intentionally) set up so that that won't work (and not just on nightmare level). It is meant as a scenario in which you have to play aggressively and, as a result, likely won't be able to avoid a certain amount of losses, even of leveled units.

The strategy that worked for me (on medium, I did not do the playtesting on nightmare) is to roll over the NE elves as quickly as possible and join your ally in the fort. Then hole up there in the center of the map for a few turns, using the ally's troops as cannon fodder (as the objectives say, none of them will join you later). At some point you'll have to go into the offensive again, probably (hopefully ) slightly before you'd generally be comfortable with it in order to get to the other leaders in time. As I said, it's meant as a scenario in which you will suffer some losses (the next scenario will allow you to make up for that), and the pressure of having to get it done within the time limit is very much intentional.

Hope this helps. Thanks for testing and the feedback!

[Velensk]

I downloaded and looked at your save and I think that you actually have pretty good odds of just smashing through if you play more aggressive (when I have time I'll see if I can do this and make a replay). Remember, you're orcs. Worrying too much about healing is for elves and humans. If you can get your main force to the castle quickly then with the terrain advantage and a good night-time counterattack you should be able to clean the field and win a good early finish bonus. The only thing that I found a bit worrying was the fairly low starting gold. You have a lot of about to level archers which can expand the value of your gold out quite a bit but you still don't have a lot for a smashing offense.

[Anonymissimus]

Interestingly, I recently played the campaign on the hardest difficulty from the start as well, and decided to go back one scenario for more gold because of Barag Gor too. But I played this particular scenario one difficulty step lower (on medium) then.
Main strategy variations I tried are teaming up against the northeastern leader (and giving up the middle castle) with both sides versus holding the middle castle for as long as possible. Well, it seems that the ladder is better, as it keeps enemies from getting villages, and makes use of 60% def for orcs vs 20% for elves in water.
I had 210 starting gold - that's about one unit more than you... (The second time, when I aimed for maximizing carryover.) And I also spent almost all gold on recruiting whelps.

Worrying too much about healing is for elves and humans.

I'm afraid that could be a multiplayer thinking style which doesn't apply for campaigns. The problem with playing aggressively, taking out the northeastern leader quickly, is that you lose too many units in the process, and the low village count doesn't allow for reinforcements. Thus you cannot take out the other leaders in time.

I wonder why the scenario setup has been changed (and not for the better me thinks). In the old setup I only recruited a single castle and ran as quickly as possible to the middle with the leader. This no longer works, of course.

EDIT
Here are start-of-scenario-save (the situation is slightly better I think, although I got only a level 2 rider recalled, but slow attack is more useful) and the replay (on medium, starting with the same save).

[mattsc]

I wonder why the scenario setup has been changed (and not for the better me thinks).

In general, all changes to SotBE were made to make the campaign harder, as it is labeled an expert level campaign but was easier than at least some of the intermediate level campaigns (as explained in the release notes of several 1.11 releases). That's the main reason for the change in this scenario, the other one is that those of us who discussed this thought it would make the scenario "more interesting". The latter is a matter of opinion, of course, as is the second part of your statement.

In the old setup I only recruited a single castle and ran as quickly as possible to the middle with the leader. This no longer works, of course.

Exactly. That's boring. Now it's much more interesting (and harder). I'm not trying to convince you, just saying that my personal opinion is different. If we get overwhelming feedback that the new setup is bad, we can change it back (eventually), but so far nobody else has said so.

[Velensk]

After many attempts I haven't been able to do it on that save.

I have been able to smash through black and the initial attack waves on the castle and get my entire force to the castle several times however at some point after this either my defenses wear thin enough that they can get through and snipe one of the shamans or I encounter a twist of bad string of bad luck at the wrong time (I only restart from scenario beginning).

I think that if I had enough gold to recruit a few more grunts at the start (for either green or red) I'd be able to hold through as the margin of my defeat has been very narrow a couple times. It can really help even things up if you identify your intelligent green archers and feed them all the kills. Still, I haven't been able to quite pull it off.

EDIT: In response to the comment about not worrying about healing not being a campaign thing. I'd disagree (not that you shouldn't do your best to obtain as much healing as possible but as orcs but it's not a primary focus in most cases. It's quite possible to attack black losing only a whelp and a grunt or so with his save.

[Anonymissimus]

I tried again to see whether I could have won on HARD as well (using my own save); while I'm clearly on the winning path (southeast ist dead, positive income, limited losses for red), I'm almost sure that I can't finish in time.

[mattsc]

So, from what I am hearing it sounds like things are just a little too tight on nightmare. What should we do to bring it more into the realm of the doable? A bit more gold for the player? Less gold/income for the AIs? If so, how much? The turn limit is the same on all difficulty levels and I'd prefer not to touch that.

It is supposed to be a seriously difficult scenario, but not so much that it cannot be won without having to rely on luck, of course.

[Anonymissimus]

Well, somewhat between medium and hard, based on the current values ? I would reduce enemy base income, it has the highest impact on difficulty in my experience.

So, do we consider ourselves campaign experts now ? Last time I was involved in such discussion, the author (beetlenaut, not yet a dev back then) assumed there'd be much better campaign players than himself and balanced the easiest level to be hardly beatable (for me too).
For what it's worth, I never considered the strategy described above (quick move to center castle) boring. I can't name any other mainline scenario out of the top of my head which features the same. Back then when I played it for the first time, I recruited much more in the starting castle.

[mattsc]

Anonymissimus: attached is a replay finishing the scenario in 21 turns starting from the save you posted. Note that this is not without reloading, however it is without reloading for luck. A couple times I got careless and tried to rush through during the day, so I went back 2 or 3 turns and restarted from there. With that experience, I believe that I could play all the way through without reloading now, but I simply didn't/don't have the time to do that right now.

Either way (as in, whether I could do it or not) I am quite confident now that an expert player can play this scenario on nightmare without changes and reloading. There are certainly things that I could have done better.

For what it's worth, I never considered the strategy described above (quick move to center castle) boring. I can't name any other mainline scenario out of the top of my head which features the same. Back then when I played it for the first time, I recruited much more in the starting castle.

Agreed - I was being facetious about it being boring. That's what the was about. The main reason for the change was something much more game-play/balancing related, but I forgot the exact reason. So no, it wasn't boring before (but personally I do think that it is even more interesting now).

Oh, and I also agree that the enemy income is the hardest part in the endgame (and again, that is very much intentional).

[Anonymissimus]

Hm. I think the main useful point about the strategy in your replay is that you manage to keep alive and level a lot of green units. You do this by hesitating a lot to move onto < 60% def spots, and you use rest healing rather than killing a few more units. Thereby you have more green units left, forcing enemies to often attack 60% def spots from 40% (or worse) ones. On my last trial, the castle was basically lost, red wasn't there in time, and I finished the remainders of the main wave north-east of the castle. In the late game, you still have several green units alive, which are basically upkeep-free, and expendable. You got careless in the late game, there are several occasions where you let them attack you from forest.

However, one would need to make sure the next scenario is beatable too, with Kapuoe on level 2 and the gold you get out of that.

I wonder what's best strategy for green leader, while he's a very useful unit with poison and marksman, his death enables green units to heal in red's villages without capturing them - gold impact on next scenario! In your case, you chose to steal red villages pretty often.

[mattsc]

Thanks for the analysis. You're right about all that, in particular also that I got careless toward the end. I was getting tired and was just trying to get it done. I also really didn't think about the next scenario, just wanted to make sure that this scenario was playable by itself, so didn't really care about maximizing gold etc. I would do that differently, of course, if I didn't play the scenario in isolation. [I do think that the next scenario should be playable even the way I came out of this one, but I am not going to try to prove this since, as you say, I could come out of this one with somewhat more gold.]

I do like the idea of intentionally letting the green leader die at some point to maximize gold...

Btw, the one thing that I think would have made this scenario a little easier coming in would have been a leveled or close-to-leveling grunt on the recall list (and have one leveled archer or wolf less). I usually try to get at least one orcish warrior early on, but it always seems hard to keep the grunts alive because they have no ranged attack.

[Anonymissimus]

Btw, the one thing that I think would have made this scenario a little easier coming in would have been a leveled or close-to-leveling grunt on the recall list (and have one leveled archer or wolf less). I usually try to get at least one orcish warrior early on, but it always seems hard to keep the grunts alive because they have no ranged attack.

Firstly, recalling grunts is strongly discouraged, as that means 8g difference over a fresh recruit. Secondly, grunts are too difficult to keep healthy until you get healers. And thirdly, they are almost useless in the undead scenario and the desert one (where you need healing desperately). So I get them just for cannon fodder, and prefer whelps.

[Maiklas3000]

I went back a scenario and played to maximize gold and got 186 starting gold, which was less than my previous 195 gold. I tried again but got 186 gold again. Maybe I need to go to remedial school for gold mining.

I bumped my starting gold up to 265 (i.e., cheated) and then I was able to win. I don't recommend that as a modification of the scenario, though, as it makes for too much unit-pushing and too huge a roster gain if things go well.

[Edit: I meant "mattsc" when I wrote "Velensk" below.]

With my original 195 gold, I was able to mimic Velensk's strategy, with partial success. Seeing Velensk's approach caused me to stop recruiting so many Whelps. I also learned to stick to the hills initially rather than extending my line into open ground. My castle crew was in worse shape by the time my main force arrived, so time was lost retaking parts of the castle. Time ran out as my forces approached the northwest and southeast leaders.

Velensk levels a surprising number of the castle's units. I try to steer kills to intelligent archers, and I rest-heal like crazy, but I don't get those kind of results. What am I missing? One difference is that Velensk seemed to tolerate the continued presence of the units initially around the castle, whereas I try to exterminate all of them before the assault waves arrive. I guess if you're surrounded by overwhelming forces, then unit kills basically make no difference, except, of course, for leveling.

I suggest reducing the gold of the enemy leaders somewhat, because even Velensk was not able to win with my starting situation, and it wasn't a bad starting situation. Currently, there's also the cruel twist that if you take the easy way out in the previous scenario (i.e., running Kapou'e rather than defeating the leaders for the early finish bonus), then Barag Gor will be quite impossible to beat. I prefer when taking the easy way out means the next scenario will have high casualties and be very difficult to win, but still possible.

[Anonymissimus]

mattsc you mean, I suppose ?
Here's my replay from the previous scenario. That was on nightmare - how I got the starting gold. 1 unit doesn't make much difference though. The pillagers I leveled are probably more important (slowing enemy leaders before attacking). So taking the other advancement path is a mistake IMHO. Pillagers are also pretty useful in the undead scenario.