howitzer

Maybe we're going to get a physical remote with it. And maybe it's also going to rain candy next week and then the hell freezes over.

No it didn't, as I_Igor mentioned, the gear was introduced in the BWE, though I believe the RTC had it for the first time in black.

You don't consider the Bugatti a flagship? RTC had bigger parts count yes, but it was much less expensive and didn't have much of anything new, it was just a really big set.

I guess subtractror gearbox could be made to run with two (coupled) motors for drive and one to steer, as single motor drive would probably be too weak for this scale. Coupled with blade+ripper lift and tilt it would take 7 slots in the two hubs. 8th function could be the three speed gearbox for drive. No cool new parts though with this configuration so I'm not really expecting it. 2018 had the wave selector, which was probably one of the most important parts of the decade.

Ooh, those fan blades! I had a GBC idea involving them but didn't get around to actually trying to build anything as the blades are pretty expensive today... I hope the prices come down with the release of these sets. Nice to see those pneumatics parts (though still no thin long cylinder) and recolours of various pieces too!

Yes, partially. 42043 also contains the long thin pneumatic cylinder (the only set so far where it appears) which is tens of euros by itself. It's also very highly regarded set and released in 2015 (while 42100 is 2018 set) so it's both sought after and generally less available than 42100. You could also go for an used one, they should be less expensive.

I can't say much of the alternative builds (especially the unofficial ones) but for the main model all of these are great in their own way. 42043: expensive, but it a lot of different kinds of parts (some of which are quite pricey), in itself a great model packed with functions 42100: if you want to get into PU, this set is the way to go with it's two hubs and seven motors which will keep you busy for a good while 42082: great parts pack for its price, fun to play with as the functions work great and are nice to use 42055: even greater parts pack for its price and one of the more complex and unique models out there, with lots of interesting functionality Oh, and regarding your old interest in Mindstorms: the new iteration (51515) was released just recently and the hub is fully compatible with PU motors, but better in almost every way than the C+ hub so I might recommend that also, if you want to do more programming stuff.

That FB post doesn't include any source either, so I have to assume it's just guesswork concerning the details like number of motors and such.

I wonder where that guy's info is coming from. Does he have an actual source or is it just speculation and guesswork?

That would be the letdown of the decade, and it has only just began.

There's been tons of sets with linear actuators recently, I'm really hoping for a set with lots of pneumatics for a change.

42043 is indeed a great set, even as a parts pack, because it includes motor, pneumatics, suspension, many wheels and even a linear actuator. The only downside is that it's also pretty expensive today so unless you find it at bargain price somewhere, I'd recommend something else. 42055 is also great though it lacks pneumatics and suspension parts and only has 4 small wheels, but the amount of structural parts is unrivaled and there's a lot of track-links and gearbox parts too. All in all it comes down to what you want to build though - no single set covers everything and if for example you want to build fast cars you have no need for tracks, for those it's better to invest in lots and lots of panels in the colour scheme of your choice.

Haha, good for you for having the patience to wash the parts afterwards. The electronics indeed have pretty good water protection, though I guess the various gaps are not exactly watertight and might let some in which could corrode internals over time...

Wow, I missed this when originally posted but it's great! I love the details and the functions both, though seeing it in the muddy creek made me cringe a bit for the damage the parts would sustain from such an abuse...

A lot depends on how much time and/or money are you willing to invest and what do you think you need. Also, don't forget to consider sorting and storage, Lego can take a lot of space especially if you want it properly sorted. I would say that if you want to gather a wide selection of parts, buying sets is the best way to do that but comes with some serious caveats: The upside is that all the parts you'll be getting are brand new, you'll know exactly what you're getting, and you'll get a full set to build which you can examine and learn stuff from. The downside is that the options are very much limited by whatever is currently available so you'll really have to do your research first on what types of parts you need. A lot of useful parts can be rare in official sets (some were common in the past but not anymore) so you're better off buying those separately. As you expand your collection, buying sets for parts becomes less and less useful, as you're going to already own those parts in large enough quantities so after a while you should only buy sets which offer something new for you, like a colour scheme that you want to add to your collection (for example, I don't have much of yellow parts so buying 42114 might be a good investment for its exterior but its chassis is still built mostly from LBG bricks so I'd have those in excess quantity). Buying parts individually from Bricklink or whatever gets you exactly the parts you need but can get expensive. My gut-feeling is that the price per part is about five to ten times higher compared to official sets but this can vary a lot depending on many factors. If you have a parts list of what you need, you could cross-check it's contents against sets currently in production and see if some of those would fit well into your needs so you'd get most of the necessary parts from the set and then fill the gaps with a Bricklink order. As for buying used Lego, the very cheapest way is to go to an auction site or fleamarket, and buy batches of mixed Lego. You can get a lot of stuff for very cheaply, but you don't get to choose what you get very well and used Lego is often dirty so you should wash it first, which is a lot of work and it's even more work to sort it properly. Buying used sets gets more expensive (again, a lot of variation here), but they are often in better condition than mixed batches and you'll have a good idea of the contents and whether it fits to your needs or not. Something to keep in mind if you're doing digital builds: unless you're experienced and know what you're doing, it's very easy to build digital models that are impossible to build in real life and/or are very weak and fall apart if you attempt to build them from real bricks. You really should get yourself some physical bricks first and attempt to build something with them so you'll have some feeling of how a Technic build is put together. I often build simultaneously both, the digital and physical model, alternating between the two as the building progresses. As for the Mindstorms, you can find more info on its specific forum but personally I can highly recommend the new 51515 set. It includes four motors and a hub of 6 ports (as opposed to 4-port hub of the C+ sets) and a rechargeable battery so it'll get you a nice starting pack to build motorized MOCs.

Yeah, those are valid points too, though the cables of PU motors are a bit longer than those of PF motors so you can create pretty large models even without extension cables. Hopefully we'll see them released for PU at some point.

There's also PU AAA battery box, though it only has two ports. Still perfectly usable for small applications. The PF rechargeable battery box (84599) was discontinued in 2013 and costs 100€ or more in Bricklink, even used. If you just want a motor which runs and don't care about position encoding or speed control, PF is indeed better and the same goes if you have to run huge number of motors simultaneously with low torque. Other than that, PF doesn't really offer you any advantages. I'd say the main downside of PU right now is the need for programming for MOC operation, as that sets the bar higher than PF even with the simplistic block language. Of course it also means that every PU MOC is RC while PF would require additional components in order to turn them into RC. The number of motors is also somewhat limiting, but that's probably a problem mostly for hardcore AFOLs, who like to make huge and complex machines with absurd number of motors. Vast majority of the people will never need more than the 4 ports offered by the C+ hub - and you can increase this to 6 with the RI hub with the added bonus of a rechargeable battery (though that's going to be a bit more costly.) There's also a wider selection of different kinds of motors, which allow for more options in design than PF. I have to wonder why people keep repeating that PU components are somehow very expensive, when they are available cheaply in Bricklink? A C+ hub costs less than 20€, so that's hardly a huge investment and motors cost around 12-13€ each which is maybe 1€ more than PF equivalents. Yeah, they are hugely overpriced in the TLG's online store but only a fool would buy them there rather than Bricklink.

Divide km/s by 3,6 and you'll have m/s.

The pneumatic switches in Arocs are indeed easy enough to use, but the same is not true for other controls. Remember, there's outriggers, crane slewing, dumper bed and pump to control too, and it's all too easy to get confused with them as the direction switch is at battery box which is bad design in the box itself as it crosses midposition too easily and starts to reverse when you'd want it to stop, and it uses the same reversing switch for multiple different functions. So unless you're using the pneumatic switches, you need to have one hand all the time at the battery box (which requires very careful operation) and the other must be turned around the model to use the other switches. This isn't just me, but it's a point that has been brought up in several Arocs reviews, and even Jim commented on this on in his review of 42082. This is probably true. I haven't built the Sian myself so couldn't comment from personal experience, but TLG really wants to make the gearbox build as easy as possible, because correcting mistakes later is extremely painful and they might not be apparent before very late in the build.

Designers of the real things aim (or at least they should) to make the controls as easy and intuitive as possible though, so unintuitive and difficult controls are a sign of either bad or heavily constrained design. I'd say it's the latter in the case of 42043, as I don't really see how they could be improved without redesigning the whole thing and compromising it elsewhere. So yeah, the controls don't make 42043 a bad set but 42082 still had it much better as it didn't have similar constraints in design. Having more gears for the sake of having more gears is bad design too. For some reason the official sets sometimes seem to have many more gears than necessary but there might be other factors in play (marketing or whatever) beside optimizing the gearbox from engineering perspective.

Maybe this is a similar case to the bushing-type axle holders breaking? Those which for some people seem to break constantly and for others almost never.

@Kumbblhas a point there in ease of operation. 42043 is quite hard to initially learn to operate smoothly, while 42082 is really intuitive and easy to operate. I don't like using the battery box as a direction switch, it somehow feels a cheap solution compared to using a gearbox for that. Whichever it is though, the controls should be located in places such that they are logical to the function they operate. Again, 42082 shines here, while 42043 not so much. Don't get me wrong, I absolutely adore 42043, it's one of the best sets ever, but it's not without flaws either.

Yeah, you couldn't push pin through them, but you could still insert bar's end in the pin to help with extracting the pin from another part. This was especially useful with the first iteration of pins (from 1988 or so) as they were horribly tight and sometimes impossible to separate with fingers only.

I'm thoroughly annoyed by these design changes concerning bars inside pins, as the older design allowed bar to be used with liftarm to remove the pins. Sometimes in tight spaces this would be a very useful feature but only the black 2L pins and blue 2L axle-pins allow this currently.

All of those are effectively two-dimensional, I was thinking more of a part that would be large in all three dimensions. And yeah, I don't think such a part is necessary for the current scale either, but if TLG starts to push for the ten thousand parts scale, they're going to need something more robust.