davee123

I didn't know the stagnation was happening as early as 1993, but I know 1998 was the first year they ever suffered a loss. At the time, LEGO took something like 2-5 years to create a product. So (generally speaking), the stuff they would come out with in, say, 1998 was actually entering the design phase around 1993-1996. On that note, LEGO started ramping up the crazy train around 1996-1999, which probably makes sense, and that they were designing things starting around 1993 when sales started doing poorly. At the time, LEGO had crazily ambitious goals. They wanted to be the next Disney. Specifically, their goal was something like: "By 2005, we want to be the #1 brand in households with families". Note: They weren't trying to be the most profitable, they just wanted to be the one that customers rated the highest. On that note, they started mixing things up quite a bit: 1996 - Primo 1996 - LEGOLAND Windsor (the plan was to open up a new park every 3 years) 1997 - Juniorization (Town Jr - first big conscious effort by LEGO to juniorize) 1997 - Scala 1998 - LEGO Mindstorms 1998 - First LEGO video games 1998 - ZNAP 1998? - Clothing lines 1999 - Licensing - Star Wars 1999 - LEGOLAND California 2001 - Licensing - Harry Potter 2001 - Bionicle (following explorations into Throwbots and RoboRiders) They also started doing different types of themes like Ninja, Adventurers, Rock Raiders, Aquazone, Wild West, Time Cruisers, etc, which were a departure from their tried-and-true "Town, European Castle, Space Fantasy, and Pirates". Most of their new initiatives weren't very successful. But some were. In 1998, they had their first financial loss, but I think their thinking at the time was that they were losing money thanks to the initial investments they were doing, and that it would magically start paying off "really soon". Plus, everyone in the toy market was suffering, and claiming it was due to video games, so LEGO played that card too. DaveE

Yeah... so is he :) In fact, one of the intentions of the Brick Testament is to show ridiculous parts of the Bible. DaveE

I think it's possible to cram a 6-brick 90-degree orientation into 53 bits by identifying the connection points rather than X,Y,Z coordinates. In fact, you could do it in 51 bits (although... odd), and probably even less if you make some assumptions. Basically, identify the "stud you're going to connect to" with 4 bits, the "stud you're going to connect with" with 4 bits, the orientation (1 bit), and the "brick you're going to connect to", which grows from 0 bits at Brick 2 up to 3 bits by the time you get to Brick 6. That's 7 bytes-- so with 915103765 different configurations, it yields about 5.97 gigabytes of storage? Granted, that storage mechanism means a lot of interpreting, so you'd likely be using up more CPU per comparison than the X,Y,Z coordinate representation. The non-grid representation would be... different. Certainly you could go with: Brick N is an array of 'connections' identifying the stud/hole being connected from (4 bits), the brick being connected to (3 bits), the stud/hole being connected to (4 bits). You don't need an "orientation" anymore, because you would presumably determine that from the list of connected studs. But then you may be storing extraneous connections (if 2 bricks are stacked exactly on top of each other, you KNOW all 8 studs are connected after identifying any 2 connected studs, and then you're uselessly storing 6 connections of 11 bits each). Best case, you're storing 6 bricks in 55 bits. Worst case, you're storing 6 bricks in 440 bits. If it's evenly distributed, that means roughly 247.5 bits per connection (fits into 31 bytes). And with at LEAST 915103765 combinations, you're looking at 26.42 gigabytes. Certainly ... possible ... with a LOT of paging. But then you've got a whole chunk of processing that has to go on for each possibility for collision detection and viable angles. Yeah, I would expect it to take ... months or years to calculate. DaveE

1) LOTR license - Correct to the time (that IS how they felt), but wrong now, since LEGO *DID* pick up the license 2) San Francisco store - Now there are 3 brand stores in the SF area (at the time there weren't any LEGO Brand Retail stores in North America, just 2 "outlet" stores in Minnesota and Virginia). 3) Color mosaics - Ok, he's still sorta right, unless you count 6162 and 6163, which... aren't really what was being asked about. 4) Molds for LEGO Direct - Technically, they DID for the Emerald Night, even though it's not necessarily called the "Direct" group any more. 5) New Track Geometry - New flexible track came out in 2009, although granted, he probably just couldn't talk about what they were planning. 6) Death Star - We got 2 of them! And again, he may have known, just couldn't reveal it. 7) Skin tones for real people - He's right, LEGO started using flesh tones in 2003 for real people. 8) Why not Spielberg and Zidane in yellow - He SHOULD have known this. Basically, they were designed before the policy went into effect. 9) Skin tones for characters vs. 'real' people - Wrong! LEGO *did* do skin tones for characters. Again, he should have known this-- LEGO was in the middle of transitioning at the time. 10) Lando/Mace Windu will be yellow? - Correct, they weren't necessarily yellow, proving that he should have answered #8 and #9 differently. 11) Bulk brick variety - No answer, but we DID eventually get things like Pick-A-Brick in stores, and LUGBULK. I think the funny part was mostly that the comic's depiction of Brad's attitude is nearly spot-on. He'd often say VERY little-- often just a "yes", "no", or short phrase like "I won't comment on that". Comparing to other LEGO employees that we get nowadays, it's like night and day. These days, we're told the boundary lines of what they can and can't talk about, and they're MUCH more vocal about things they CAN talk about. DaveE

I think the way to do it would be to take "standard" sets (MISB or whatever), and dump them in a container one at a time until it "looked full". That way your ratio of pieces will be pretty on-par, as well as the "looking full" factor to determine when you should stop. Yeah, you have the ability in a large box/bucket to use space more efficiently than in a PAB cup. Generally speaking, people that buy PAB cups will optimize them in some way-- either by carefully stacking elements, or by sprinkling in small pieces in order to minimize the empty space. But when your typical parent decides to sell the old LEGO at a yard sale, they just take all the LEGO and dump it quickly into a box, such that everything fits. There's a large variety of elements, and you get a lot of empty space near the top, where the bigger elements tend to hang out. And as long as the lid closes, they're happy. DaveE

I believe the LEGO company originally asked someone to come up with the statistic in order to promote the brand. IE, demonstrate mathematically just how gigantic the possibilities are so that they could justify saying something like "effectively infinite", or what-have-you. However, the original calculation was also very wrong. It was done in the 1970's sometime, when computing power wasn't where it is today. The person assigned to the task decided to see how many combinations there were with simply stacking bricks-- so no 2 bricks were on the same vertical level. Plus, nothing other than 90 degree rotations. Then, in the ... early 2000's somewhere, someone realized that the calculation wasn't complete, and decided to do a research paper on it. Probably for a class project or thesis or something. I think that's the one that's linked to above. They decided to allow bricks at ANY level (not just stacked vertically), but still only limited it to 90 degree rotations. Not outrageously so. Somewhere in the gigabyte range, depending on your data structure. Offhand, I can imagine 11 bits per brick, which is 66 bits per combination, which fits into 9 bytes-- about 7.6G. If you database the information, rather than keeping it in memory, that's peanuts. And there's probably a lot of optimizations you can do on that. The newer algorithm by comparison is ... daunting. Speaking as a programmer, I'm not really sure how I'd approach the problem. Anything I can think of would probably take many years to complete. Kudos if you can figure out how to do it! DaveE

I did some checking on average weight some years back, and came out with about 350/pound: http://news.lugnet.com/general/?n=52892 But that'll depend (obviously) on what types of elements you've got. If you have basic building tubs, your average will probably be lower, because they're full of large pieces. And if you've got a lot of technic, it'll probably be higher. As for using PAB to determine volume, it's probably not a good way to estimate, given that people often pack their cups pretty densely, and usually people put in a lot of small parts. For the record, I have data for 64 PAB cups that I've gotten over the years, which have contained 22,191 pieces. Assuming that my approximations for PAB volumes are correct, that's 50915 mL, or about 13.45 US gallons, which works out to about 1650 pieces per US gallon. From there, assuming the above average weight per piece, you could guess 2138.36g of LEGO in a gallon. But that's way off from the above guess by @legomr. As stated, I think it's probably too high thanks to the amount of packing that typically gets done in a PAB cup. So, although I can provide a chunk of nice statistics on PAB cups, it's probably not very applicable to generic LEGO by volume (or, such would be my guess). DaveE

That's probably an understatement :) With axis-aligned connections, you can grossly simplify the rules to the point where a computer representation of the bricks is easy to define. But this really pushes the boundary since you've got to do things like calculate whether or not the corners of the bricks collide with the studs of the bricks upon which they're attached-- not to mention the more obvious collision detection, and of course the plethora of possible angles. Essentially, the algorithm is SIMPLE enough-- it's just recursive to a degree that I can't easily come up with a way to make it efficient. Essentially, for N bricks, you take all the known "configurations" of N-1 bricks, and attempt to add another brick at every possible connection point. That part sounds pretty easy. But for each possible connection point, you MIGHT have to consider MANY "models" that are possible with that one "configuration". Plus, you need to essentially explore "having 1 end connected" for the brick you're adding, and "having more than 1 end connected" for the brick you're adding. I can imagine an algorithm to do this, but I just imagine it being VERY slow. The trick will be optimizing the algorithm, and PROVING that it really DOES explore all possibilities. Relatively speaking, perhaps-- but I wouldn't think that de-values the final result if someone's willing to do the work! To me, it sounds somewhere around a master's degree thesis in programming or something. I'm not sure if it's really worth the effort for the number of curious minds out there, but it's more likely to simply be satisfying to accomplish! If you actively want to find the answer, and have a good solution for efficiency, I'd say go for it! DaveE

I'm pretty sure that the UCS ISD was available at Toys 'R Us in the USA. And I *THINK* some select TRU's also had the UCS Millennium Falcon. I seem to remember seeing a box of one at TRU with a security device wrapped around it, unless I'm misremembering. I also believe that a bunch of other UCS sets were available retail (but not all of them): 7181 - UCS TIE Interceptor - Definitely normal retail 7191 - UCS X-Wing - Definitely normal retail 10019 - UCS Rebel Blockade Runner - Definitely normal retail 10026 - UCS Naboo Fighter - Definitely normal retail 10030 - UCS ISD - Pretty sure normal retail 10179 - UCS Millenium Falcon - I think normal retail? 7194 - UCS Yoda - Not sure... 10018 - UCS Darth Maul - S@H only? 10129 - UCS Snowspeeder - S@H only? (and LBR) 10134 - UCS Y-Wing - S@H only? (and LBR) 10143 - UCS Death Star - S@H only? (and LBR) 10174 - UCS AT-ST - S@H only? (and LBR) 10175 - UCS Darth's TIE - S@H only? (and LBR) 10186 - UCS General Greivous - S@H only? (and LBR) 10212 - UCS Imperial Shuttle - S@H only? (and LBR) 10215 - UCS Jedi Starfighter - S@H only? (and LBR) 10221 - UCS SSD - S@H only? (and LBR) 10225 - UCS R2-D2 - S@H only? (and LBR) 10227 - UCS B-Wing - S@H only? (and LBR) I... *think*... that's right. But that's from memory, admittedly. DaveE

Yeah, that's why I mentioned that it may not be quite in line. It's simply the biggest LEGO box for a LEGO product that I'm aware of offhand. As stated, I expect there are even bigger boxes that contained things like playtables and such. There are quite a few caveats you could consider, like: 1) Is it an "education" brand? Stuff from LEGO Education/Dacta/etc are often meant for classrooms, NOT for individual people. Hence, you could argue that if it's from one of these, it might not really qualify. You could argue that for #7100, where it's meant for a classroom of kids. 2) Was it sold retail? Some stuff was never sold retail, or at least *general* retail. Some things are "custom", or otherwise not for general retail, so they might have HUGE boxes. That could be reason to discount anything at that level. 3) Is it a "bundle" or "combo" set? In some cases, LEGO associates a product number with MANY sets, or many COPIES of a set. For instance, LEGO Education might sell 16 copies of a kit, boxed together with a single set number. I've never ordered one of these, but it might ALSO raise a question of whether or not the packaging actually contains the "set number" in question (and isn't just a generic shipping box). 4) Does it contain actual LEGO building elements? In some cases, like playtables, sorting cabinets, etc., there aren't any actual building bricks, so it might not be worth considering. 5) Does it contain LEGO "System" elements? You might want to discount things that include other scales of LEGO, like DUPLO, LEGO Soft, Primo, Quatro, ZNAP, Scala, Clikits, "Jumbo", Modulex, etc. They might contain "building elements", but they're not quite the same thing (as is the case with #9020). Anyway, I would guess the ISD could be the biggest in the strictest sense of the above (disallow 1,2,3,4,5), and 7100 is likely the winner if you disallow 3,4,5. And #9020 might be the biggest if you disallow 3,4. I suppose it COULD be the biggest if you allowed ANYTHING, but I doubt that pretty highly. ... I just don't have an example handy of a box that's bigger. DaveE

The LEGO Soft starter kit box (#9020) is roughly 39cm x 39cm x 48cm, which is about 73,000 cm^3. But that's a LEGO Dacta/Education box, so I'm not sure if that counts. It's also not "system-scale", either. There may have been bigger boxes for other things like playtables and such that WAY outclass everything else. I don't have the LEGO Soft box in front of me to measure, but I have one at home-- and I know it fit 8x8x8 "LEGO Soft" bricks dimension-wise, and each LEGO Soft brick is 6x the normal size. So the INTERIOR of the box is at least 384mm x 384mm x 460.8mm (plus some extra on top for the studs). It's also in a corrugated cardboard box, and I'm sure there's a little bit extra around the edges. DaveE

I think it's just that in the past, they typically have correlated, which is why the numbers are so surprising to me. Again, it's not that I necessarily disbelieve the numbers outright, it's just that the numbers as stated go against what we've seen in the past, with strong LEGO purchasing markets lining up with AFOL population, in addition to other factors. I think the reason they've correlated in the past is due in part to nostalgia factor-- IE, areas where LEGO sales were strong 10-20 years prior will produce adults that have nostalgia for LEGO, and that will be drawn to the hobby. Oh, I'm mostly surprised that the number of clubs that DO participate in the Ambassador program appears to be LESS than the number of clubs that contributed to the presented data. I would think that most LUGs that participate in the Ambassador program would similarly have statistics for the data collection-- especially since (in theory) clubs are supposed to re-submit their qualifications for Ambassadors (I'm not sure if this is really "enforced" or not) That's more along the lines of what I'm trying to determine. I guess my discomfort with the statistic is that to me, "AFOL" or "LUG member" has a certain definition, so to say that there are 130,000 of us worldwide somehow implies something that's misleading. So I'd like to understand the composition of that number better (assuming it's not a typo or something). If an "AFOL" or "LUG member" in South Korea means "I enjoy looking at LEGO models, and I've even purchased a few sets for myself", then that's an important caveat, IMO. Well, in this case, I think it's a bit simpler than the AFOL market share. The AFOL market share numbers are calculated in many countries by analyzing sales in a variety of venues, combined with customer feedback, surveys, and other market data through 3rd parties. It's a phenomenally complicated value (or so we're led to believe) that I'm comfortable taking on faith. However, I'm familiar with the "Tell Us About Your LUG" surveys that they do, which is what it sounds like these numbers are based on. Those aren't cross-checked by LEGO with sales or population or ... anything, really. They just ask "how many members do you have?", and we say "we've got 100 members", and they accept that. So, in this case, it could very well be a single contact that's reporting these numbers, and LEGO is taking them on good faith. I'm not aware of any market data analysis on those figures by LEGO, which, again, is why I was asking who compiled the data, so I might be able to verify with them. Anyway, I don't mean to make a mountain out of a molehill-- I just want to understand the numbers. DaveE

Yeah, Google probably wouldn't be good for determining those metrics easily. I would expect corroborating evidence to be something like LEGO reporting that AFOL-targeted sets sold 10x or more higher volume in South Korea, or hobbyists in the area that are "in-the-know" confirming the source of their data in a much more in-depth fashion. Basically, with a number like 30,000 LUG members, it's just so out of whack that I'm imagining one or more of the following: - They allow kids to be members - They allow parents of kids to be members - They have a website with that many registered users, and the users may-or-may not be actual AFOLs, and/or from South Korea - They had some event with that level of attendance, and considered attendees "members" - There was a typo or miscommunication in the number reporting Huh-- I thought Japan was the largest Asian market... I know they already have a version of the LEGO site for Japan (I guess I assumed they had S@H direct, although I wasn't quite clear). They've also got Click-Brick stores (I actually got to go to one back in 2011!), and they've long since had Kabaya promotions, which is Japanese. Also the line of collector books in Japanese, which have been around since... 2003 or so? Hm. Now I'm curious to go back and re-read some of the annual reports to see if South Korea gets any special mentions in the list of countries with large sales volumes. Yeah, I've been linked to by BrickInside a few times since... Hm. I guess I'm not sure when. I've seen them before, but I don't really know how big of a following they've got, or what types of people are on their site. I don't know if it's AFOLs, parents, kids, or what-have-you. Hmm... According to the list of Ambassador-contacts, I see listed: HKLUG (Hong Kong) Thai Brick Club (Thailand) BrickInside (South Korea) T-LUG (Thailand) Pockyland LUG (Taiwan) AFOL of Japan (Japan) TWLUG (Taiwan) That seems... Odd. At least 5/7 of them look like LUGs, possibly all of them. But the stated percentage was 2% of 160, whereas 5-7 would have been 3% or 4%. So it would seem that at least some of these weren't included in the earlier statistics, and PhilBricksters probably isn't on that list? Was Kevin the one that compiled this data? Maybe I should just go ask him what the dealy-o is with this data. It seems like there are some interesting points that may need some explanation! DaveE

I'm not basing my guess on the fact that I haven't seen their online activity or anything. Basically, looking up population for the country, density, and taking into account their lifestyle should yield a relatively round number estimate for how many I'd think would be there. I might compare to Hong Kong or Japan as potentially similar in nature. My offhand guess for South Korea would've been something like 100-1000 LUG members. I'd also say that I've never seen any information from LEGO that supports such a large market. LEGO always suggests that the most lucrative country for LEGO is Germany. And I know German fans are pretty active, even though they aren't always on English-speaking sites. But this would suggest that there are roughly 15x as many LEGO hobbyists in South Korea, even though I've never seen anything from the LEGO company catering to the market. I've seen LEGO cater to the USA, Europe, Germany (language-wise), and Japan, as well as a few minor spin-offs here and there. And that suggests to me that those are probably the larger AFOL areas of the world. I'm honestly curious-- how are you gauging your statement, apparently defending the numbers as reasonable? Do you have corroborating evidence that suggests that the reported numbers are at least close to accurate? I don't mind being wrong, I just haven't seen any evidence to support there being so many "AFOLs" in Asia. My suspicion is that their definition of "AFOL" or "LUG member" is different than ours, so they're reporting something that's not comparable. Either that, or it's a typo or miscommunication. But if there's evidence to support something of a vastly different scale, I'm certainly eager to hear about it. DaveE

Assuming South Korea had 25k AFOLs, that would mean: About 1 in 101,677 people in North America is an AFOL About 1 in 31,588 people in Europe is an AFOL About 1 in 1,955 people in South Korea is an AFOL I dunno, it just sounds phenomenally suspicious to me. DaveE

I always have a difficult time assessing whether LUGs are "active" or not-- I think LEGO's "requirements" for a LUG include a certain level of activity and so forth-- do we know which requirements apply to the 76 in question? Hmm... I wonder how much crossover there is in that 130k figure? I know back in Ye Olde Times many people were frequently "members" of many different LUGs. I know in NELUG, we had more than 100 people from as far away as South Africa were were clearly not "really" members of our club. Otherwise, that's a pretty impressive number! I recall guessing there were in the ballpark of 100,000 of us worldwide, but it seems like the number is probably even bigger! Hmm... That... seems... odd to me. This is only based on LUG data? That would imply that about 101,000 AFOLs that are LUG members in Asia are all grouped into ... about 3 or 4 LUGs? That's an average of between 25,350 and 33,800 members per LUG? How is LUG membership defined in these clubs such that more than 25,000 people are LUG members? DaveE

Huh-- so the motor set that's listed on LUGNET/BrickLink was the first one? http://guide.lugnet.com/set/002 http://www.bricklink.com/catalogItem.asp?S=002-1 Interesting that the motor that came out in 1966 by LEGO proper (not Samsonite) was pretty different than this one. DaveE

Can you clarify for #1 whether that's LEGO System motors, or motors for other toys that LEGO may have had? (I honestly don't know if other toys had motors-- and would a wind-up motor count?) DaveE

I think I know answers for 2, 3, 4, 5, 6, 9, and 10. 1, 7, and 8 are befuddling me though. 1 and 8 MIGHT be referenced in one of the LEGO books, but I don't recall either offhand. I also found one reference to Duplo in Australia on BrickFetish, but it... seemed to just be called "Duplo". So maybe there's more to that story... Will we get any answers? DaveE

Yeah, I don't think I've *EVER* had any of my chrome elements lose any aspect of their shine, color, or chrome. I *have* purchased elements whose chrome has worn off in places, however, which I assume have been played with and handled excessively by children. The elements in question, however, I don't own yet, and MIGHT be chromed through a different process. So I suppose there could be a problem with "modern chrome", but I haven't seen any evidence of that to date. DaveE

I forget who came up with the idea, but at a meeting a while back, we decided it'd be a great AFOL contest to redesign classic or otherwise fantastic LEGO sets with totally inappropriate play features-- like the Statue of Liberty with flick-missile eyes, Cafe Corner with trapdoors, or Town Square with "falling axe traps". Who can come up with the least appropriate and most overzealously numerous amount of play features for an otherwise amazing set? DaveE

I have to agree. I haven't seen too many unprofessional actions by Eurobricks staff, but this is clearly an example of one. Can someone in EB administration explain further why this might have been justified? DaveE

I've done static doors (not openable) like this: http://www.suave.net/~dave/tmp/bm/shots/design4de_facade1.jpg And I've done openable (but not paneled) doors like this: http://www.suave.net/~dave/images/parv/building/parv_door.gif But I feel your pain! It's been a long-standing gripe of mine that there aren't many good interior doors without windows in minifig scale! The vertical panel doors are ~OK~... And at least there's the solid 1x4x6 door now (although it only came in one set, and only in white, so it's pretty pricey!) DaveE

It's true-- just my experience generally has been that people that can actually afford to get to the point of preferring something like vertical versus horizontal injection molding are more coherent. The phrasing mostly struck me as a homework assignment that someone decided to post to a LEGO forum, figuring that LEGO hobbyists might be able to give a decent response. But it's such a short description, I can't really tell. DaveE

Sounds to me like a homework assignment that we're supposed to answer for him. DaveE