There is already a lot of difficulty porting to Switch from PS4/Xbox One, so much so that most AAA games skip the platform entirely, those that don't often come late and require entire seperate teams. With the event of PS5/SX, games will also become less optimised for PS4/Xbox One 

Something akin to the New 3DS or DSi is not what I'm referring to - that's entirely possible when they announce their redesign of the current Switch.  That is not Switch 2, which will be an entirely new system with the start of a new library for the 10th gen.  This splits the user base and the developers' priorities as well.  New 3DS had only a handful of exclusives, with Xenoblade and Fire Emblem Warriors being most notable, and the DSi had even less.  Not the same thing at all.  The idea that Nintendo would release a more powerful system so soon just to cater to 3rd parties is a recipe for disaster on so many levels.

Gameboy Advance was facing impending competition from Sony with PSP.  Nintendo was set with a touch screen gaming device years before iPhone or iPod Touch.  Launching the DS when they did was indisputably the right thing to do; I remember at E3 before launch, Iwata himself said that he was most proud of the DS out of all of Nintendo's accomplishments.  That was the power of the DS and Nintendo knew this.  The Switch faces zero direct competition on the horizon and its sales in its fourth year will far exceed what the GBA was doing aligned.

But you have made yourself clear to me now:  The majority of your argument hinges on the fallacy that Switch's sales will be in a dangerous decline come 2022.  If a redesign is indeed released in 2021, Switch's sales will be over 20m again, and 2022 will see shipments sales well above 15m; far more than the ~7m that the 3DS garnered in its 6th year.

More than conceptually difficult, I'd rather say that for AAA titles it may become time consuming. Scaling down textures and, if necessary, redesign the least important game item models with less polygons doesn't affect the gameplay and game engine, but AAA titles with a reputation on graphics to defend can't just scale down with mostly automatic tools, they must check and tweak the results carefully to keep them as good looking as possible even at lower res, this surely makes managers question the opportunity and profitability of some ports, but as far as we know some ports could be simply delayed to arrive when the NS user base will be larger than now. Another factor is that a large part of NS users are mostly great fans of titles developed for portables and/or family and party titles, this can make devs and publishers predict that some PC and home console titles would attract a too small share of NS users. Obviously this evaluation may be totally wrong, and anyhow those that must decide know that even a very pessimistic prediction about the percent of NS gamers interested in a game can eventually take to an absolute number worth the investment once the total NS user base becomes large enough.
Anyhow, I see that we both agree that the more games will become optimised for PS5 and XBSX, the more difficult NS ports will become, up to eventually become impossible.
I'd add another thing: NS shows Ninty that, at least for Ninty itself, hybrid is the way, so, even with a dropping 3rd party support, unless it drops really too fast,  a new NS can't come too soon, as it would make it too difficult to make it powerful enough when undocked. Luckily for Ninty, thanks to the faster growing mobile market,  ARM CPUs development can react faster than Intel and AMD to new requirements.

I think the switch will last until 2024 spring at the very least. Nintendo still has a huge buttload of games it can bring to the console and in terms of sale, it's not slowing down any time soon. Also Ninty might want to enjoy the harvest period of the console for a bit longer so they could stretch it out as far as spring 2025. Right now looking at sales, it seems like Ninty can do no wrong. I won't be surprised to suddenly learn clubhouse has reached 1mil or something. You never know which Ninty game can be the next big thing like how Ring fit was. And every game franchise seems to be selling the best or close to the best on switch. The biggest example is Zelda.

Post bookmarked for future reference. :p

im suprised you didn't agree with the content of that post

I go back and forth depending on my level of anxiety and depression. For today, both are relatively in check.

Six months is not a long period to wait, we can check that.

I think you are not understanding truly why Wii U failed where the switch succeeded and severely underestimating the switch. The switch will continue to sell with or without big third party support. And support will still come because of the sales it has and the fact that it will have unreal engine 5. The really advanced 3rd party games aren't necessary for the switch's success or any Ninty console's success. Plus the tone of the console is set in the first 2 years and it;s really hard to shake that. If it's poor, turning it around requires a lot of effort and if it's really good, you have to screw up royally to make things go bad. Spring 2023 is the earliest, not the latest. Spring 2024 is the most reasonable and there's a possibility of spring 2025. People forget that the longer Ninty waits, the more advanced the switch 2 can be based on a reasonable price. 

So your assumption is that in 2023, it will be a big weaker than Lockhart. What if I'm 2024, it's a bit stronger and in 2025, it's much closer to XSeX than Lockhart. The switch will continue to sell so why wouldn't they just delay it a bit and make ,ore profit from it. They still haven't released a pro model that will extend the lifetime of the console even more. Or even made one price cut to the OG console.

I actually don't think I'd want to have to wait until 2025 for Switch's successor.

Systems start get boring and feel stale after a while, and 8 years is just too long. Pretty sure by then I'll be sick of Switch.