I've both played and GMed this scenario. Both times, I walked away dissatisfied. However, a good amount of this was because because of party makeup when I played it. In addition, I think I brought my negative feelings playing it over to my GMing experience.
This was the second or third PFS scenario I ever played in. I was playing a rather terrible elf rogue, and the rest of the party were either Aasimars, Tieflings or had the wild empathy class feature. I failed my save against the effect on the first time and had to spend the entire scenario as a sad lion. Remember the sad kid monkey kid from Jumangi? Yeah, that was me the whole time, but lion-y. I feel like the combination between that scenario mechanic and my party make up just punished me for being the only player stupid enough to bring a character that wasn't immune to the condition.
All in all, this scenario wasn't difficult to prep or otherwise run. I feel as though the new secondary success condition, preventing the fire from spreading, is extremely difficult for players to achieve and un-intuitive. There's nothing to indicate that it's vital to save his farm, and every second counts when the fire spreads so rapidly. However, the end encounter is flavorful, though the unique map is a pain to draw.
I've run Scars of the Third Crusade several times at this point, and it's honestly one of my favorite scenarios to date. In general, I enjoy roleplay-heavy scenarios, and this one certainly fits the bill! Players can really make their characters shine in the way that they approach the investigation, though it can definitely take some prompting! While not every table I've run has participated in more than one combat, they have all left smiling and feeling like they've had fun (at least, from what they told me!).
However, it is a pain to prepare (at least the first time). Luckily, the PFS shared prep site has some nice worksheets to use for all the tracking going on. While obviously the scenario should stand on its own, this is something you should look into if you go to run it.
In addition, it's very difficult for a table to have complete success. There's not much the PCs can do about it, unless they're extremely careful. While this is understandable, it just feels like it's too easy for them to be significantly punished for bad luck and their approach to the solution. In addition, the way that the track advances means that players may want to skip forward to encounters, which makes it a delicate work-around between following the scenario and throwing up a Plot Barrier. Specifically, when my players
wanted to go out of their way to meet Erika before Encounter One was triggered, I had to make it so they had to find her first. In this case, I used the Word on the Street encounter in order to illustrate the process of asking around about her.