N64 jump of the gun to 3d realms and forget all 2d classics/arcades types Nintendo has launched. One CD-based and 2d/3d ecosystem, in time, is a better choice than a cartridge focus in 3d games only. The N64, besides greats games, suffers a lack of variety and quantity of your library. Plus, Nintendo forget your ethos: arcade base/2d games.

3DS has this but fixed it because Nintendo portables always have the arcade/2d games besides 3d technology. Nintendo lost all the children with 3ds glass free

But 3ds is not failed. N64/GC/Wiiu is all missteps.

.."gradual decline not huge mistakes after massive successes like Pyro said"..

Losing 50% of your customer base is never really going to be considered a "win", if the PS5 loses half the PS4 market base, people will call it a disappointment, some will call it a failure.

- Is it still a failure if it takes MS out of the game?

Smartphones do dominate the casual market by a monstrous amount at that, Switch is successful precisely because it offers experiences deeper/bigger/more console like than the smartphone and got away from the sinking DS/3DS format of being too small scale of an experience. If Nintendo tried to release a Switch with more casual type experiences as the main killer apps and lower specs they would be in a lot of trouble right now.

 - Fortnite and PUBG are available alongside Mario and Pokemon but no-one ever claims it'll result in Sony/MS/PC casuals moving to smartphones.

The top 10 selling Switch games are very different from the Wii or DS, the Wii's top 10 was dominated by stuff like Wii Sports, Wii Play, Wii Fit, that is not the case for the Switch. Ring Fit is a nice hit buoyed by the fact that gyms are closed or a no-go for hundreds of millions of people right now, but it's not going to be one of the 5 best selling Switch games like Wii Fit was for Wii, not even close. 1,2 Switch is not going to be anywhere close to the top 15 Switch sellers the way a Wii Play was, or Nintendogs for DS. All the IP that are selling are long established Nintendo IP like Mario, Mario Kart, Zelda, Animal Crossing, Pokemon, the difference is those are "big brother console" versions of those IP and not the tiny weeny portable versions but because of advances can be played portably too. That puts the experience on a different level from other systems including Nintendo's own past handhelds.

- Who cares about 'top selling'? It just shows how gullible the hardcore are to marketing. Ring Fit and Switch Sports will sell 20M+ like their unbundled predecessors did. Mario Party on Switch has outsold Mario Party and Wii Party on Wii and DS. Since when were Pokemon, Animal Crossing, SMB and Mario Kart not czualz games?

...

Besides the Switch's sales growth in 2019 in the U.S. prior to the Lite (which BTW was mainly in just the Jan.-May period; June was nearly flat and July was down slightly), we have only two other clear-cut cases of systems that experienced long-term (more than one quarter) sales growth absent a price cut, the release of a new model, or an external factor like the pandemic. And those examples do not appear to be global. Those examples are the Wii and the DS, and both examples are limited to the U.S.

The Wii did experience sales growth over time in the U.S. over its first two years. However, it's not clear what caused that growth. The Wii did have well-documented supply issues plaguing it early on in its life, so its sales growth could simply be an artifact caused by supply not being able to fully meet demand. Smash and Mario Kart did give some big boosts the months of their release, but sales trended downward over the rest of 2008 (until a big restock in October). Beginning in March 2009, the Wii's sales declined significantly, with the yearly total only being as close to 2009's due to the price cut to $199 in September, which was followed by what ended up being the single largest holiday performance of any home console ever in the U.S. This sales trajectory did not happen in Japan, where it was much less popular, peaked in its first year, and declined every year afterward.

The DS is a more complicated example. It had four consecutive years of overall net growth for the year, from 2006 to 2009. However, its growth wasn't consistent and its sales trajectory was... weird. Now, bear with me here because this is a lot of detailed sales history to cover.

There was that initial spike in June 2006 from the Lite's release, then it dropped off slightly in the July-Oct. period, though sales were still higher than prior to the Lite's release (an average of 83.4k/week for the 4-month period, vs. the 35k to sub-40k range in Jan.-May). Then after the 2006 holidays there was a big drop in Jan. 2007, with a rebound in February. From February to October 2007, the DS managed a fairly consistent (and every so slightly downward trending) baseline that averaged out to ~107.5k/week. That's better than the July-Oct. 2006 period, but there were no huge blockbusters in late 2006 or early 2007 that could account for this. So, what caused the Feb.-Oct. 2007 baseline to be 28.9% higher than the July-Oct. 2006 baseline? We don't know. There is no obvious cause.

Then the DS experienced much larger YoY gains in the 2007 holidays than its more modest YoY improvements from the previous several months. Again, no huge blockbusters that holiday that could be the culprit (the biggest DS game of 2007, and the only one to crack the yearly Top 10 in the U.S., was Pokemon Diamond & Pearl, which was an April release that, for some inexplicable reason, caused no appreciable HW sales growth that month). However, Nintendo did issue some big holiday bundles, which may account for it.

2008 was an odd duck of a year compared to the more stable 2007, with weekly average sales jumping all over the place. There was another unimpressive January (relatively speaking; it did well, but was a weak month by DS standards at the time, placing #5 for the month) followed by a very strong February & March, which were way up YoY—and were the two best non-holidays months to date at the time for the DS—for no clear reason. No big games released in either month. Then April was down YoY by over 11% while May was up only about 7%, nearly cancelling each other out. June then set another non-holiday record for the system. That one does have a plausible explanation at least, since there was a special edition Guitar Hero-themed system & bundle for the release of the "On Tour" game that month (another example of a limited edition hardware release spurring short-term sales growth). July sales were still strong, possibly a residual effect of the Guitar Hero bundle. Then sales started to drop back to about where they were in April & May. November was barely up YoY (+2.6%), but December was up 23%. There were some more limited edition bundles released on Black Friday, which may account for December, but November being flat despite the bundles indicates that month may have been posting worse than expected sales in the weeks prior to BF. So, while 2008 was a bigger year than 2007, it wasn't due to any sort of long-term growth continuing from 2007, but because of intermittent spikes, one of which doesn't have any clear cause.

Now, 2009 being bigger than 2008 was a bit more understandable. Q1 as a whole was up a bit, but only because January didn't see another unusually large drop-off; February and March were down YoY, though. Then the DSi was released in April, which put the DS over one million units, which in terms of weekly averages is still the best non-holiday month ever for any system. The DSi has a significant residual effect over the following four months, with May & June averaging over 150k/week and July & August averaging over 100k/week. That put May & June ahead of February (which is almost always bigger than either month) and July & August ahead of March (which is also usually one of the bigger non-holiday months of the year). The DSi bump had fizzled out by September, but it was still a 5-month bump, and it helped drive DS sales in the first three quarters of the year to over 5.7M units, a record that stands to this day. When compared to 2008, though, nearly all of the DS's gains (in numerical terms) in the non-holidays months of 2009, which amounted to 835k units, were in just four months. Those months were April (+265k), January (+260k), May (+181k), and August (+35k). The DS was down YoY in March, June, July, September, and October, which somewhat offset those gains (February was almost perfectly flat, being up a mere 1000 units). This suggests that without the DSi the DS would have been down for most of 2009. The 2009 holiday season saw further gains over Holiday 2008, which again appears attributable to Black Friday bundles that included limited edition models.

So, in summary, we see a stable 2007 baseline that's better than what the DS was at in the July-Oct. 2006 period, but with no clear cause. It could have been stock. It could have been increased interest from casuals. We don't know. Then the less stable but overall improved 2008 numbers, which were almost entirely due to just six random months (Feb. & March, June-Aug., and December), were not fully explicable. Limited edition bundles can account for June & July and the holidays, but not for February & March, which accounted for 18.5% of the total growth that year. 2009's gains were entirely from January, April, May, November, & December, with January being an unknown and the rest due to the DSi and holiday bundles.

This means that we have two systems that experienced some form of growth over time that appear to be entirely due to reasons other than major software releases. The Wii's growth can be explained by stock improvements to a system that was perpetually sold-out for at least two years, and the DS's growth was irregular. Of its three periods where sales improved for a term exceeding 2-3 consecutive months, one (Feb.-Oct. 2007) had no clear cause and the other two (June-Dec. '06 and April-Aug. '09) were due to a new model. The rest of its gains were short-term bumps.

From 1995 to 2018, we never saw a single instance of a software release being a long-term (>3 months) driver of sales by itself or in conjunction with another major title released close to it, something that probably would have been completely uncontroversial to point out prior to a couple of years ago. Except for the examples of the Wii and DS, all long-term sales growth in that time span was fully explicable as a result of price cuts or hardware revisions, and in the DS's case most of its gains are explicable in terms of hardware releases. Plenty of other systems have seen strong lineups of games do absolutely nothing to improve sales, even when those games were released in rapid succession (and better AAA third-party support for PS & Xbox means more big games each year). But price cuts and new models do. The May 2002 price cuts for the PS2, GameCube, & OXbox. The PS2 Slim (once stock issues were resolved at the start of 2005). The OXbox's third and final price cut. The PS3 Slim (which was concurrent with the PS3 getting cut to $300). The 360 S. The DS Lite and DSi. The GBA SP. The One X. The 3DS's first price cut. The Wii U's first price cut. And so on.

It is for this reason that I don't believe the sales growth we saw from the Switch in the U.S. in the first five months of 2019 was a result of late 2018 software releases. Yes, the Jan.-May period was up 29.1% after adjusting for Jan. 2018 being a 5-week period. While it's not massive (we've seen far better percentage growth out of other systems), it's not insignificant, either. But given everything we seen from over a quarter of a century and several generations of systems worth of sales data, there isn't any reason to believe that the Switch's growth in the first half of 2019 was the result of software. Not when growth was never the result of software before for other systems, especially systems that were selling well. You don't just assume some new phenomenon (i.e., long-term growth coming from software) suddenly manifests itself just because, nor do you assume it exists because it sounds intuitive (individual games can and sometimes do cause growth, but that doesn't mean they can do so over a long term).

I also find it unusual that the YoY gains were relatively stable with no clear trajectory, then suddenly sales go to being flat over the summer. Why would those games suddenly stop moving hardware after managing to cause YoY growth ranging from 24.4% (in April) to 32.7% (in May) with no discernible pattern? Typically, when we see a period of growth we see a downward trend over time in the YoY gains.

Is it impossible that those late 2018 games boosted sales in the first half of 2019? Of course not. It can't be conclusively disproved, either. But in the absence of any other instances of software releases doing anything like that before, I'm skeptical that they did it with the Switch. It would be the first time on record of such an occurrence.

Just because Switch sales in the U.S. improved after Smash & Pokemon Let's Go doesn't mean they were the cause. Post hoc, ergo propter hoc. There has to be some sort of evidence in the sales data. Something that can't be explained by anything other than software. And you have to prove that individual games can have an impact lasting many months after their release, something that we've never seen before. The Switch's gains in the U.S. in the first half of 2019 could have been just as easily due to other factors (I seem to recall various deals and bundles at the time). And the DS showed that it's possible that sales can improve absent of any obvious stimulus, assuming said improvements weren't just statistical flukes.

Shadow1980, thank you for this thoughtful post.  I want you to know, for the record, that I don't think you are 100% wrong.  Especially, if we focus the discussion back toward Switch sales, I think you have stumbled onto something that most others have missed.  For the record, I think both Animal Crossing and COVID had boosted Switch sales in 2020, but I don't think either one is the actually the biggest cause.  There is actually another sales factor going on that is not being accounted for.  The console market is actually going through a transition/transformation period that the PC market already went through.

If you go back 20-30 years, there were consumers buying laptop computers, but desktop computers were much more common.  Today laptops are more common among typical consumers and smartphones are even more popular than laptops.  (Although businesses tend to prefer desktop computers still.)  If a person wants a really powerful computer for the best price, then desktop is still the way to go, so why do so many more people use laptops then?  People tend to prefer the convenience of laptops, and you can often get one really cheaply if you are willing to get something really low powered.  Of course, smartphones are even more convenient and can get even cheaper than that, which is why they are even more popular.    

This sort of phenomenon is described in books like The Innovator's Solution.  The market as a whole often chooses a product that, on the surface, appears worse.  In reality they are sacrificing raw functionality for improvements to reliability, convenience or price.  This especially happens when the product with lower functionality gets "good enough".  Laptops are still not as powerful as desktops, even today, but at a certain point most consumers decided that laptops were "powerful enough" and laptops encroached upon the desktop market.  Many customers switched over to laptops and now laptops are more common outside of a business setting.  The Innovator's Solution even describes the math behind this.  The rate that the new market encroaches upon the old one can be described by an "S-curve", i.e. a logistic function.  (The author uses the term S-curve, but the math term is actually logistic function.)  It looks like an exponential function at first, but once the curve gets past the half way point, growth slows and eventually levels off into an "S" shape.

So, if we bring all of this back to video games, then handheld and home consoles markets would be analogous to the laptop and desktop markets, respectively.  At a certain point handheld systems should get "good enough" and encroach upon the home console market.  If you look at the handheld market closely enough, then it should also be obvious when this happens: with the Switch.  For the past two handheld generations, Nintendo has gone a different route with a 2-screen handheld system.  They have kept the power levels down and tried to focus on games (like touchscreen games) that couldn't be played on a home system.  All of this delayed the inevitable encroachment.  But with the Switch, the power level of their handheld system shot way up compared to the 3DS, and they even included a dock so that it was easy to play the system on a TV.  Nintendo is intentionally trying to encroach upon the home market. They don't just want former Wii U customers either.  They are going for Playstation and XBox customers.

If this is true, then what should the data look like?  It should show that Nintendo is getting a bunch of extra customers that they didn't have on the 3DS or Wii U.  At the end of the Switch's life we should see "extra" customers added to the Switch's install base like an S-curve.  Since we are currently only about half way into the Switch's life, then this surplus is going to look like exponential growth at this point.  And in fact, this is exactly we we do see.  Look at the gap chart (the first chart) whenever trunkswd compares Switch vs. 3DS + Wii U.  For both worldwide and the US, these curves strongly resemble exponential growth.
https://www.vgchartz.com/article/447667/switch-vs-3ds-and-wii-u-in-the-us-sales-comparison-january-2021/
https://www.vgchartz.com/article/447572/switch-vs-3ds-and-wii-u-sales-comparison-january-2021/

The Switch is really a handheld system that is also encroaching upon the home console market.  That is the biggest factor causing the spike in Switch sales.  They are taking future sales away from PS5 and X|S.  It's not obvious now, because Switch launched significantly ahead of the other systems.  Give it a few years though.  It took a few years for Switch sales to really take off, and it will take a few years before it's obvious that the other systems are going to underperform.  Switch is getting its sales growth by taking sales away from PS5 and X|S.  The handheld market is encroaching upon the home console market, just like the laptop PC market already encroached upon the desktop PC market.

Because the Line up is pretty solide if u never owned a Wii U, unlike other Nintendoconsoles which got wat more draughts. Nintendo banked really hard on that in 2020 and probably will do do so in 2021, the first two remasters (3D world and Skyward Sword) are already announced/released. Both Will sell more than 10 million.

Hey Liquid, that's very thoughtful, thank you. Never looked at it from the PC angle but I think you are right. But as you said it will take years, probably, the PS6 (maybe even a new Xbox) will still be a traditional home console at heart (with streaming functions to any device I guess). Probably, the PS6/new Xbox will sell significantly less (but still could make more profit) than both, the last generation PS5/Xbox One X/S and the forelast generation PS4(Pro)/Xbox One (X). The question will then be if there will be another physical generation or not and if there will be another physical generation how will it look like? Just streaming machines, super powerful hybrids or something completely else?