Template:SalesChart This might be ready for prime time now. I started working on a means of documenting sales charts a couple of years ago, but then... stopped. https://segaretro.org/Metropolis_Street_Racer/Reception#Software_charts Although the US and Japanese charts tend to have more data, and I'm not totally sure how best to handle that yet. Arguably we don't need to keep another record of release dates and prices... but maybe we do, because if the charts are reporting it wrong, that might be an indication of... something. Eeeee it's complicated.
Some thoughts; We should probably split those chart pages out into specific charts. I didn't allow for that when I created those pages, and have shoehorned in several different charts into some pages. There are also many more charts yet to be added which will make it even more unwieldy. Maybe something like this for naming the chart pages, open to suggestions; [Chart Compiler] [Platform(s)] [Country] [Time Period] [Metric] Example; [Gallup] [CD Consoles] [UK] [Weekly] [Units] Charts Or for the Spanish charts on the MSR page which are currently 2 (3) different charts on the same page; GfK DC ES Monthly Charts GfK DC ES Monthly Units Charts GfK DC ES Monthly Value Charts To confuse matters with this example, Computer Hoy Juegos didn't specify whether the first chart was ordered by units or value. After this the charts were by value, but also with the position by units noted. Later still the charts are only listed by units. It's also not clear on some other country charts whether they are by units or value. We can just put them on their own page for now until we have more information. We can probably stick to regular charts such as weekly, monthly, and annual. The Japanese ones are a bit of a mess at the moment as I've added irregular charts that magazines published such as occasional lifetime sales or financial years. The sales numbers for those could just be included in the corresponding weekly chart, albeit without chart position. Famitsu DC published weekly top 5 for each of the weeks since the last issue, and a top ten for the whole period which could range from 2 - 5 weeks. For that reason I would consider the Top 10 to be an irregular chart and just include the Top 10 unit sales in the corresponding weekly chart. Where known we should also note the dates that the chart covered. During holidays Japanese "weekly" charts can be double or triple weeks. Whilst monthly charts can be calender months, they're generally retail months. That means they're either 4 or 5 weeks long (generally the last month of each quarter is 5 weeks long), it allows for like for like comparisons over the previous year as each year the same month contains the same number of weekends etc. That does mean that every 5-6 years is a 53 week year. There is an ISO standard for that which Gallup/Chart-Track in the UK and Media Create in Japan follow (and I suspect other European chart compilers), but as usual US has their own slightly different system which NPD uses. Famitsu meanwhile seem to have made up their own non-standard variant. Can we not just ignore unnecessary data on the reception page, but keep it on the chart page? For example NPD has the average selling price of each title, which might be useful if someone wanted to look at how quickly a title dropped in price, but probably isn't really important for a quick overview of it's sales on the reception page. I think the actual unit sales numbers should be included on the reception page. If we wanted to simplify it we could just include cumulative sales. We have a lot of those for Japan, quite a few have been leaked for US (including lifetime sales for all DC games), and first week sales were often published for the UK. For now we have very few for other European countries, but they were available in the subscriber reports, so it's possible that some of those get scanned one day.
Yeah I suppose it's worth keeping any data that was published in a chart on the chart pages (within reason). Number of sales is an awkward one because there is a separate sales table on reception pages (the idea being, when a publisher sends out press releases saying "this sold 3490248203 units", we've got a place to put it). The system's currently in use with hardware but not so much software... because we don't have many numbers. I guess my plan would be that any charts that supply sales information... would populate that table too. But then I guess there is a difference between units sold and revenue generated, and if the numbers only apply to a sub-set of retailers... hmm. Best plan is just to put as much information on the chart pages as possible, and I'll work out a solution to the above later.
My suggestion would be to use sales tables for official sales numbers (ie sales from publishers to retailers/distributors reported in press releases and financial results). The charts should only be tracking sales from retailers to consumers. Whilst some of these did have different charts for units and revenue, there only seems to be publicly available data for unit sales. Your last point is a good one ... no sales tracker tracks 100% of sales. For example NPD started software tracking in 1994 with ~62% coverage .... they generally upweighted their numbers to 100% (ie they just multiply by 1.6), but it's not always clear if they did. Fortunately I do have access to some of these ... on condition that I don't publicly share them ... which obviously isn't great for a sourced wiki. On the other hand Gallup/Chart-Track in the UK reported the "raw" numbers that they tracked with an estimate of the market coverage that they had. It generally increased over time from ~75% for second half of 1994 to 95% a few years ago. The problem with this is that the numbers we got were usually through the trade media who had access to the original reports. Sometimes they just reported the raw numbers, other times times they extrapolated to 100% themselves. So it's not always clear which method they used. We could always just state whether those numbers are "raw", "extrapolated", or "unknown". The problem is that this kind of data was proprietary, publishers paid tens of thousands of dollars for it, and staff were under NDAs. So it's not always clear exactly what the publicly available data represents. Japan is a bit different ... whilst there was one professional tracking company (Media Create), there was also numerous magazines publishers also publishing weekly charts (with 100% extrapolated numbers) who also sold their data to the trade. The main one being Famitsu, followed by Dengeki, and Softbank did it for a while too. So Japanese numbers were more open, and all extrapolated to 100%. Some were more accurate than others, depending on market coverage.
https://archive.org/details/dengeki-dreamcast-issue-1-december-1998/page/114/mode/2up I still want a Dreamcast tank.
One of my big complaints with people on the internet is that they don't write anything down. Even the tedious things. So let's not repeat that mistake: Teletext again. I've just brought Sega Retro up to date (at least in terms of game reviews) with happenings on Super Page 58, the top fan site for Channel 4's old Digitiser service. They archive pages as screenshots, using clever techniques to rebuild data from VHS recordings. So why are the images coming through with different fonts and crazy resolutions? All three of the above are just as correct (although secretly we'd want the underlying source so you could tweak things as you want). While teletext was standardised, the means of displaying it was not - just like computers having slightly different looking ASCII character sets, every television rendered the service ever so slightly differently, because manufacturers had different ideas. The third one here is more common, and was probably half-invented by the BBC who pushed this service - certainly slightly blocky sans serif is how I remember Teletext, though I've seen all three of these in real life (and with some TVs you may have even been able to choose!). (and the inevitable shot of that one wacky TV that let you print pages) It's a similar story for aspect ratios and borders and how to deal with widescreen - is slightly higher resolution text a selling point, or does it just increase manufacturing costs? You're always going to be stuck with six colours and 1970s-era "pseudo-graphics". If you're wondering how much of Teletext has been archived, the answer is "not much". Even with Digitiser all these Sega reviews are unaccounted for, and those are pages that are actively being sought.
So I was flipping through a scan of Gamest magazine on archive.org because I love going through old Japanese gaming magazines when I saw this article about Virtua Cop. Towards the bottom of the page, I see the plaque from Sega's old headquarters renamed to "Miracle Co. Ltd". I tried using machine translators for this, and I get something along the lines of "the familiar front door to Sega's...oh wait I mean Miracle Co. Ltd headquarters." Not sure if this renaming was just for this issue, but I thought it was pretty interesting to find nonetheless. This is from issue 119 from July 1994.
oh come on From Bargain Hunt this afternoon. A Sega slot machine in the background???? Maybe! Or not. But probably, unless it isn't. The Sega Bell is a piece of history the company would rather forget about. Service Games had a wonderful system in place where they imported loads of "High Top" stock from the Mills Bell-O-Matic factory in Reno, Nevada, made any missing (or replacement) parts in Japan, and sold rebadged units across US occupied islands in the Pacific, and bits of Europe (and apparently North Africa). The UK was one such market, so what you're seeing here could be a mash-up between genuine Mills parts from the US and cheapo Sega parts from Japan, probably from the early 1960s. They had no permission from Mills to do this, but they made a lot of money from the venture before the law caught up. Of course, Mills did eventually start exporting equipment on their own terms, and there's no guarantee this is a British model, but yeah. And yes they did lose money on that cabinet.
That teletext review of F1 Challenge is a bit harsh , like it didn't set the world on fire but I don't remember it being that bad!
I twokked this 1995 product catalogue: https://retrocdn.net/File:1995SegaProductCatalog_US.pdf Not quite sure where it's from (E3 maybe?), but it's got a few fun faces: Free Runner! 32 Xtreme! Virtua Hamster! 32X X-Men! I've mirrored some of the details but there's plenty more fact checking to do.
Something easily missed: Three things you need to know about VR Troopers: 1) it existed b) it was comprised of footage from two different Japanese Metal Hero shows, which means they had to contrive reasons for the three main characters to never fight together (aka what was the point) iii) there was a 32X game in development: I assumed this was one of many lazy Mega Drive -> 32X conversions, i.e. give the the 16-bit version a bit of colour and call it a day, but according to the descriptions, no - this was set to feature real time 3D worlds with vehicles, spread across 30 levels. There are no screenshots or footage, but if the marketing is to be believed, this might have been one of the most ambitious 32X projects of the time. You can take that with an enourmous pinch of salt, because "the time" would have been late 1995, after a drought of 32X releases and even fewer that actually bothered to tap into the 3D capabilities. To put this into perspective, the three "big" 32X games mentioned in the above catalogue are Virtua Fighter (which is fair enough, but late given the Saturn version was out), Spider-Man: Web of Fire (which is horrendous) and Prime Time NFL Football (which wasn't released!) That being said, 32 Xtreme was going to feature "Paintball snowboarding". And Ratchet and Bolt was due to take place in A.D. 2101... war was beginning what happen somebody set up us the bomb
RE: Slot machines, because idk I figured it might be wise to work out why there's so many variants. It's the early 1960s and these are mechanical devices - you wouldn't think there'd be much room for innovation... and there isn't. But meh. After the Sega Bell, there was the Star series, which could be entirely original Sega products that pre-date the Sega 1000, but we don't know for sure (yet). They're called Stars because of their super duper new feature: the star. Although Sega Bells might have also had stars. But whatever, these ones definitely do. Slot machines rules are typically quite simple (I mean they usually printed them on the front of the case). You're matching fruit and/or bells and/or BAR symbols on a horizontal "pay line", and there are payouts with certain combinations. The machine only works in "coins", should you line up three stars, you get 200 coins on top of whatever else you've won. At least I think that's the idea. Why stars? Because... uh. It's more common to see "7" these days, though maybe "lucky number 7" didn't mean much to the markets Sega were targeting. Who knows. The Bonus Star has a light up BONUS banner that means something. Maybe if you win when all the letters are lit you'll get a bit more money (18 coins seems to have been the default) - it's not super clear from the material we have. The Progressive Star makes more sense to me - there's a separate counter that increases every time the game is played, which is added onto the aforementioned star bonus. So in reality, it's a Diamond 3 Star + a little bit more. The Bonanza Star has three rows (and diagonals) rather than a single pay line. It should therefore be paying out more often, but probably isn't because gambling. And the Mad Money Star seems to be the same as the Bonanza Star, except it has Alfred E. Neuman's face (from MAD magazine) on the reels that is worth even more. This is the one where Sega used the face without seeking permission and got sued. There's also the "Console Sega" which is some confusing multi-player thing that probably isn't connected (outside of looks), the "Olympia Star" which has buttons in an attempt to appease Japanese regulators, and a couple of cost-reduced "Starlets" which came out much later. We also have adverts for the "Multi-Bell 35", "Paybak Star", "Diamond 4 Star" and "Double Pay Star" but I'm not sure any of these were actually manufactured.
...or even better! This is a "Lucky Devil" version. What does that mean? Instone & Ashby Ltd., "Mills & Sega Specialists". They'll take your slot machine and modify it, so it's a US-Japan-UK mashup. Well, it's certainly a thing.
Just doing a bit of public service since I'm here - this "Royal Crown" slot machine and its sibling "Imperial Crown" are commonly mislabeled as Sega machines. They're similar, from around the mid-1960s, and may even use identical parts (whose designs were probably borrowed from Mills), but contrary to what many auction sites and message boards believe, they're not from Sega. I think they might be British made ("crown" might be a clue there). They were certainly sold over here, and exported to US territories... somewhere. Sega liked to label things as as theirs, whereas other than crown marks, these have no indication of manufacturer or date.
Make Legs a Manhattan So this industry is stupid - the internet is telling me that this is a British knock-off of a Sega Mad Money Star, presumably to undercut an already cost-reduced machine. Given Sega were using Alfred E. Neuman without a license, it's copying copyright infringement. At least that's the story - I'm not super convinced this isn't another modified Sega machine, but it's very difficult to know which models are original because after 60 years, parts have been replaced, shells repainted and none of the original manufacturers exist anymore. There are slot machine communities, but they often point to Sega Retro for advice*, so that's hot helpful. I think the answers lie in trade magazines, or a specific Britsh magazine, (The) World's Fair, where these products were often advertised. Problem is, the first issue of this magazine came out in 1904, and it's still in print now, so it's a vast archive that's not online (and of course "World's Fair" isn't a particularly useful search term in Google). *I am reasonably confident in saying that if it's a 1950s/1960s slot machine and not on our wiki, it's probably not an "official" Sega one. Lots of slot machines are attributed to Sega (or Mills), but I think we're dealing with an industry that worked with psuedo-standard parts, because nobody wanted to design new internal mechanisms. At least for mechanical units - once solid state electronics became viable in the 1970s, things diverge again.
I did look into this, but I've yet to find one in the wild - my guess is they're either exceedingly rare or indistinguishable from "normal" Sega Bells.
No problem, man... I made a search too... but as you said it's either exceedingly rare or indistinguishable from "normal" Sega Bells... and you are our "slot machines" expert here... so ... I just wanted to warn you of this...
