The thread [POSSIBLE SPOILER] Dies Irae - high lab totals has the problem, that it cannot discuss the conditions of the lab total required and possible in DI The End of Time, without spoiling that campaign. So I took its discussion with Lebo77 - on his request - offline soon.
Anyway, The End of Time does not handwave the lab task in question.
Of the dangerous options for lab totals, the thread [POSSIBLE SPOILER] Dies Irae - high lab totals has AFAICS only addressed in passing the loop-hole in Covenants p.121 Magic Items in Laboratories, which allows an arbitrary number of different effects from enchanted items to add their points to the same lab Specialization, and as balance uses lab Warping added at SG's discretion.
To clarify my position, I do not think Lab Totals are all that out of hand. Perhaps for some and/or in some situations. But it has not caused a problem for me as of yet. What does throw me are some of the crazy things mundanes can do with rules in A&A or C&G. The high-level Hermetic stuff I can deal with. I am familiar with the concepts and potential possiblities. But when a scholar or a baker can create natural longevity items that grant a +10 bonus with no Warping or sterility, well, that is the sort of stuff that catches me off guard.
As for the Seventh Magnitude, I still hold that as a regarded benchmark. It is a sign that the wizard is growing strong, at least in a specific area if not also in general. It is difficult to start with a Level 35 spell from Gauntlet without a focus. I also reset Warping to highlight that mark. As an HR, Warping begins at level 35 instead of 30.
I should probably clarify too. My concern is not that high lab totals allow powerful effects. High powered and low powered games are both fun. The issue is that high lab totals blow through the top of the scale that the core system is designed for. The result is that instead of simply having high level spells the game has to add complexity with subsystems like the process for original research of the use of Finesse for craft magic. The basic lab process and effect level system are much more understandable and usable.
As to the mundane effects, I agree. A related problem is that everything in stackable, so that two effects that each sort of push the limits combine to go way beyone.
Setting all that aside for the moment, the OP has inspired ideas and has me crunching numbers. My saga has the concept of portal hubs and networks as a major background element. But I have introduced new elements that could radically change that and pressure characters into finding a way to compensate.
Now, the way I like to think of it is the Hermes Portal came first. Based on old Mercurian rituals or whatever, it was the way things worked back in the day. The invested device version Mercere Portal came about after the Schism War. Their primary advantage being cost effective in the long run but also having the advantahe of security (trigger design) and potentially having additional effects (it is a pair of invested devices after all).
Now why is the original a Ritual? Besides having a high level, it provides true instant transit and either disregards all requisites or metaphysically links two geographic places via magical co-location.
Does that make it non-Hermetic? Not really. Any magus with the proper skills can duplicate similar effects and/or can learn the Hermes Portal spell as they choose.
Casting it can be tricky, requiring two magi that know the spell and thirty pawns of vis (15 each). Wizards Communion and Mercurian Magi are both a big help and can bring the cost down to 16 (8 each).
Creating the Mercere Portal is a heavier investment of both time and vis. One must study a lab text to "learn the secret". Perhaps the secret is that you have to know and master the spell and that is what you spend that season doing?
Then two Invested deviced have to be opened and enchanted. 28 pawns of vis and several seasons of time. At full Verditius scale (3xCost), that is a value of 84 pawns. But it is more restricted and a Verditius may not access the "secret text".
Plus a cerimonial casting of Hermes Portal when the items are fixed in place. I would still charge vis for this "ceremony" because I am a d!k and it helps my story idea.
Now, with the righ scores in place and Labs provided by House Mercere, it will take two magi only one year to craft a single portal (1s each to learn spell, study text open item, and enchant item, then cast spell at the end). It would take a lone magus with a casting partner about a year and a half (1s each to learn spell, then lone magus 1s to study text 2s open devices, 2s to enchant).
Either way will require 28 pawns up front. A tad cheaper than HP and permanent. Having the link ceremony cost vis is what balances it out and makes my idea work. That adds up to 30 pawns, for an up front cost of 58. Minimum 44 (Mercurian Magic to cast the link).
So let us presume every peg is in place and the gears are oiled. It will take a team of two magi five years and 220 pawns of vis to build a network of five portals.
So let us say there are projects like this in the works scattered here and there. More or less. A ten year plan to invest in the Hermetic Infrastructure. An idea worthy of Changing Mythic Europe. Money has been bugeted, some of it spent. A grand work in progress.
Then everything hits a snag. A simple banking crisis. The Mercere banking system, while advanced by medieval standards, is quite primative by modern standards. Going off on my own little thing here, so just follow along.
Things hit a snag, and there are other problems to sort out that are foreground material. But in the bakground, there is the Portal situation.
The experienced team can do it faster I just realized. After learning the spell and studying the text, they can make two pairs per year.
Anyway, the network will make money once money is spent. But they are short on money.
Now presume two Mercurian magi casting just the spell. With an up front cost of only a third (just a tad over), they can set up a network all at once and start making money right away. When things stabilize &/or the system makes enough profit, invest in the original system.
I can play with further complexities, but that is the basics of my idea as inspired in part by the OP.
See, players are always going to find ways to make really big numbers. Breakthrough, "non-Hermetic," secret, Ancient Magic and other stuff like that is a way to say "these effects either aren't available to Hermetic magi or are only available in a limited form, and developing and spreading them would likely alter the dynamics of the Order of Hermes and become a focus of your saga real quick." They're plot devices, and the system provides loose framework rules for them, but it's more complicated than stacking a giant lab total and holding it for ten seasons.
So yeah, a lot of stuff (often high-powered) is "non-Hermetic," because if it could be done through standard Hermetic guidelines, then the Order of Hermes would have forgotten the nonstandard method or only kept it around as a curiosity. I imagine that there's plenty of Lab Texts still in Durenmar describing curious old spells that use the pre-Vim Realm Arts, or a quasi-Hermetic half-integrated version of Pilum of Fire that requires an Arcane Connection even if you can see the target. And there's certainly plenty of nonstandard spells with interesting and useful experimental side effects (it's too bad we didn't see more of those in published material; it would be cool if some such spells had become standard, buy-it-for-a-pawn Lab Texts to allow for quasi-Hermetic effects at a particular level).
Because Ars Magica is a game of nerds, these kind of corner cases should exist exactly so far as they're entertaining.