It's not an assumption. It's a presupposition of some people's preferred approach to play.
My "preferred approach", then, is not to have the full details of each ane every encounter laid out before me in advance so I may decide whether it has sufficient relevance to satisfy me that it is worthy of being played out. I don't need to know whether the fellow in front of me with a crossbow is likely to slay my character should he refuse the orders being given to him (high stakes indeed) or will inflict trivial damage, then fall to my blade (low stakes) to decide how my character will proceed. Providing that information removes all uncertainty and reduces the enjoyment of the game for me. This moves my "cards on the board" analogy away from even having information on the underside of the card - I know up front exactly what my odds are and the results of success or failure. A "choose your own adventure" book becomes more exciting.
Because once in the castle the key stakes - assassinating the duke - are in play. As @Hussar said, there is no mechanical rule here - it's a matter of taste and judgement, a bit like editing.
I agree with the latter, but the former is ludicrous. If there is, for example, a haughty guard who will seek to deny me access to the Duke, he is relevant to the assassination if he is inside the castle, but not if he strides out to challenge me before I can gain entry to the castle?
An encounter on the road to the castle with its architect, who knows of a secret way into the Duke's chamber is somehoe less relevant than encountering the same architect in the city, or in the castle itself, or even find his body where it was left in the desert (after the Duke had him killed to protect his castle's secrets), with a page or two of notes that would provide a clue as to that secret passage? Physical proximity to the Duke is the only thing that can make the situation relevant?
Under the proposed model, it can be no surprise that the prisoner of those desert nomads, or the fellow we're interacting with in the tavern, has this relevant knowledge, as we must present the stakes to the players prior to the encounter so they may judge its worth. "Guys, I think it is important that you not just skip over this section, because there'sa character in here who is very important going forward - here's all his backstory and future potential involvement. So can we please NOT skip the travel through the Gobin King's territory, as I have now shown you it has relevance" What do we end up with? Relevant characters walk up to the campfire with their resume in hand so the players can judge whether interaction will be relevant? Oh wait - now we're back to 90 minutes of interviewing NPC's aren't we?
Yes. That would be equally frustrating.
So how is it impossible to have situations outside the castle which would be equally exciting, and relevant to the assassiniation,to situations within the castle?
You are assuming here that whether or not something takes time at the table should be determined by whether or not it is hard in the gameworld. That is one way to RPG, but not the only one. Hussar and I are articulating a different approach - as Hussar said upthread, it is non-simulationist.
So who decides which of:
- travel to the Duke's city
- determining the lay of the land with its people
- finding a way to gain access to the castle;
- once inside, finding a means of confronting the Duke at a time and place where we can pull off the assassination;
- the actual assassination
- the escape from the location where he was assassinated
- the escape from the castle itself
- the escape from the city
- subsequent efforts to capture the PC's and/or avenge the Duke?
It seems like your/Hussar's vision is one of either unanimous group consensus, or one player's preference (depending on whether the single player or the group as a whole decides how we will proceed). So, once we have agreed on which of these aspects should or should not be played out, I'd think the GM now has to break so he can write the adventure, as it seems likely at least one item the group included would lack any relevance, and/or at least one item the group removes would have had something relevant, if the GM had simply prepared the adventure himself after the players ended last week's session with their announcement that "we must assassinate the Duke". Hopefully, no one gets bored with the concept in the intervening period, or we'll want to skip past all this borning Duke stuff and move on, and all that preparation will be for naught.
If there are no salient stakes, then the player does know its trivial (by the sort of measure that Hussar and I are putting forward).
When the players arrive at City B, they find an inn at which to eat and rest, and there are half a dozen patrons in the tavern. Trivial day to day maintenance, or could one of those people in the tavern be:
- a Duke loyalist who may pass information on these new arrivals up the chain?
- a builder who knows about a secret passage into the Duke's chambers?
- ordinary townsfolk who can tip the PC's off as to the level of resistance or support the locals may have for the Duke and, by implication, his assassination?
The simple fact that the GM plays out the tavern encounter is, frankly, a tip off to the players that "something relevant this way comes", unless your group plays out every stay at a tavern (which would become trivial and boring very quickly). To me, the game is not enhanced by the GM spelling out exactly why this scene is relevant.
No. City B, the Duke's castle, the grell, are situations - ie NPCs and PCs in some sort of confrontation over something at stake with which the players are engaged.
Semantics. By the same phrasing, the desert, the road to the Duke and the spearcarrier recruiting are also situations. The players are engaged in killing the grell, and they have decided to hire assistance to do so. They returned to the city to recruit those allies. It is the players who set the scene of choosing their recruits. I have no problem with them choosing to rush the process, and risking poor choices of allies, or of detailed interviews to get to know the people they will be trusting with their lives in the near future, but they have set the scene where they need to make that choice.
By choosing to assassinate the Duke, they have chosen a path that requires they reach the Duke's city. They have some nebulous goal requiring they get to City B, which requires they cross the desert. They have then set the "cross the desert" scene. Presuming their goal is not just "get to City B and purchase a packet of smokes", I assume there are also scenes which will occur in the city - things between them and whatever that nebulous goal in City B happens to be.
Now, I suppose we could play Improv D&D - I'll describe out the one scene I want to play, and the GM now makes that scene, and we play it out (unless one of the other players says "naw, that's boring"). Then we can brainstorm to come up with another scene, having little or nothing to do with the first. That's not the game I'd want to play. It seems very much a tale told by an idiot (in this case an idiot committee), full of sound a fury, signifying nothing.
I said simply overcoming challenges. The relevant difference is, what are the stakes of the challenge? Tomb of Horrors and White Plume Mountain involve challenges, but I have zero interest in playing or GMing either, because there are no interesting stakes.
The older modules come without stakes, or links to the campaign. One can certainly add stakes. Make one of the weapons in WPM important to the game. Change one of them to the Sword of Whoever that was in Player #3's background. I agree that the GM has a role to play in taking disparate modular adventures and making them relevant. That doesn't mean the players (much less one single player - and your posts still indicate a hive mind player team) decide "oh, we have to get past a Grell to get to the Sword? Not really feeling it. Can't we just cut scene past the Grell - tactical combat bores me!"
Disregarding the "useless", that's what colour is - description that adds flavour and vibe to the game. (I don't know why you'd think it's useless - description and flavour are part of being immersed in play.)
I don't find "so scratch off eight potions of heat resistance" provides for that immersive experience you are describing. By reducing the impact of the desert to such a trivial level, you have removed the flavour it could have added to the game. And how much crap are we carrying around to turn every issue into a "scratch something off your character sheet" moment? Sounds more like a board game than an RPG, and not even a good board game at that.
First, why will the game play out identically? To cross the desert, for instance, the PCs might cross off their "heat resistance" potions, whereas going through a forest wouldn't have the same impact on resources.
The game will play out identically because you have decided the travel is trivial and should be ignored. "Cross off X" is not gameplay to me. Do you find excitement in the GM asking you each day you have travelled whether you wish to cross off some food and water? I don't think Hussar does - in fact, tracking that minutia seems to be a key aspect he wants to avoid - and in that, I agree with him. If obtaining food and water is trivial, then we don't need to play it out. However, if the GM stops and asks "how much food and water are you carrying", my guess is that how much food and water we have is about to become a matter of concern, and no longer trivial.
Second, why is description and immesion not an end in itself? Apart from anything else, it plays a role in setting the fictional stakes.
Oh look - the setting just became relevant.
In 4e, for instance, a thief in leather armour and a paladin in full plate have much the same AC, but the difference in colour is hardly irrelevant - it gives the two characters very different fictional positions.
There are also some pretty material mechanical differences. Let's deny both their DEX modifiers. Oh, here comes an enemy that attacks touch AC. That one has electrical powers and is unaffected by metal armor. Which one is easier to drag out if incapacitated (HINT: all that plate adds considerable weight!)? Which one can swim better? Which one has issues trying to creep past a sleeping guard? Those differences seeem like a lot more than colour and description to me, and took no time at all to come up with.
I'm not sure how that would be. How would it come about that a player-created goal would be simply a plot device for propelling the action along? Can you give an example?
Let's take that Sword of Whoever. Pop it in to replace one of the White Plume Mountain weapons. Suddenly, the player is interested in playing out that scenario, and will try to persuade the rest of the players. Or maybe he'll just moan that he finds dungeon crawling dull and boring - can't we just cut scene to his character gloriously holding the sword aloft? No sense playing out his trials and sacrifices to find the sword (which have finally culminated in his determining its last known location), just relegate it all to backstory and the sword is just as worthwhile. "We cut scene through searching the libraries, questioning the sages, travelling the world to find small clues and battling past hordes of bizarre enemies - and lo, we have found the Holy Grail
Lost Sword of Whatsisname!!" Hey, that only took 15 minutes (14 of dismissing scenes we couldn't be bothered to play out, 30 seconds of triumphant high fiving and that wasted 30 seconds of the GM describing the room the sword was in - BAD GM! So, should we play another campaign? We've got an hours, so we should have time for at least three more!"
Then I think you may be mishearing.
I don't. Can't we just cut scene to the conclusion that I am right?

To go back to the grell example, for instance: there's a pretty big difference between turning up to a rematch with the grell and finding it's recruited a gauth ally (a complication within the context of the grell situation), and spending 90 minutes of play recruiting hirelings to fight the grell (which is not a complication with the context of the grell situation, but a completely different, and also in context rather low stakes, situation).
I see...so the Grell recruiting an ally to fight the PC's is an exciting and relevant complication, but the PC's recruiting an ally to fight the Grell is a low stakes, boring, irrellevant waste of time. So why did the players choose to recruit allies? Did they want to make the game dull and boring? Hussar tells me these spearmen were very important to their rematch with the Grell - it seems that would provide their recruitment with some stakes. So, if you and I are both at the same table, how do we resolve the fact that I see the recruitment as high stakes, exciting and relevant, and you dismiss it as low stakes, trival and boring? Note that, to me, a rematch with the Grell is low stakes (we have to get past to move on - will the GM just end the campaign here?) and boring (played that combat once - rematches bore me out of my skull - it's not a videogame where we just beat our head against the exact same challenge over and over until we finally slip through), so I want to play out the recruiting process, then just cut scene to us reveling in our victory over the Grell. Describe the perfect game session we build out of these two perceptions and priorities for me.
No one has said that. There are any number of ways to introduce consequences of failure or less than full success without framing things into tedious and low-stakes scenes.
Then why is it assumed that the desert travel will just be a bunch of tedious and low-stakes scenes, but once we get to City B, the game will turn from grainy B&W to high def 3d glorious technicolour?
I've already indicated with the grell example, for instance: a hireling shows his/her cowardice, or tendency towards aberration-worship, when the group confronts the grell.
You just made the recruitment high stakes and relevant, contrary to how you classified it above. [ASIDE: As an internet posting ettiquette note, it is customary to place self contradictions at least two posts apart, and three or more is generally preferred - at a minimum, they should not follow in immediately sequential paragraphs; although insertion of a second quote between mutually contradictory statements is felt by some to reduce the faux pas, this is not the case by the standard rules of ettiquette.]
That's a caricatured framing of things
Well, we can't have topped 125 of those in only 366 posts, can we?
but take out the caricature and yes, you are a bad GM for me or Hussar.
I'm so tempted to ask whether you're referring to Hussar or youself, or whether that should have been plural. Sadly, I can resist pretty much anything but temptation. I hope, and like to believe, that the caricature arises from the manner in which the viewpoint has been presented (internet posting is not a full suite of communication tools full of subtle nuances), but the fact is I don't think I'm the one who draws the same conclusion from the information we have been given. But I doubt a detailed analysis of my own posts would reveal much more than another caricature.