Living by numbers – the Amiga games chart database

Wait! It’s not another review, don’t run away! Girl Scout’s Promise (™).

What’s jolly annoying is that if you want to know where (or if) a particular game entered the olde worlde, pre-internet charts you have to delve through dozens of magazines published approximately around the time of its release. This is a real chore because not all magazines printed these league tables, and the ones that did, didn’t always do so. Some started including them as a regular feature amidst their news or reviews sections several years into their run, and stopped well before they bit the dust.

In any case, this is a totally unreliable method because often games would make an unexpected resurgence many months or years after their initial release as a result of the development of a sequel, or a new marketing push such as bundling several related games together and selling them at a knockdown price. How would you go about capturing that phenomenon? Re-read every magazine every published?

My solution is a database… isn’t it always? I’ve got quite a collection now. What I’ve done is enter chart data from one of a selection of magazines for every month the Amiga was an active/viable platform between June 1987 and July 1996.

Some of this information was originally compiled by Gallup, some by HMV or MicroByte et al. Some pollsters maintained top 40 lists, others top 30, 20 or 10, whilst others spanned their sales data across the various Amiga platforms or price bands; standard Amiga, A1200 or CD32, premium range or budget.

<span style="font-size: 10pt">An extract from Amiga Format’s April 1995 issue.</span>

I tapped whatever Miggy pulp I could lay my hands on so the jigsaw is comprised of a complete hodgepodge of source material. I could have included every chart from every magazine so we’d be able to cross reference using the overlapping data, but then I only have so many decades left on planet earth and there’s only one of me. 😉 That said, if anyone knows where I can find data prior to June 1987, I’m all ears.

Having ploughed through over two and a half thousand entries, a few interesting trends become apparent:-

  • Sports games will always sell no matter what, often for months on end. It doesn’t matter in the least if they are almost identical to last year’s edition.
  • It’s not unusual for highly acclaimed games to sell poorly, and leave the charts early.
  • Games with a recognised license attached will usually sell no matter how awful the result.
  • The majority of critics panning a game won’t necessarily sound its death knell. Some games rose above the mockery carried by the hype-train alone. Not everyone subscribed to gaming magazines, and many people bought games for others with no prior knowledge of their quality.

True, more of a curious reminder than a startling revelation, interesting nonetheless I think. If anyone agrees I’ll put the database online once I’ve finished getting everything in place. Otherwise it will be kept as an internal EA research tool for our own amusement. I have no idea if anyone cares about this or not so speak now or forever hold your peace.

Okey doke, decision made… here’s the Zoho report.

You count the days, but does it all add up to you?

29 thoughts on “Living by numbers – the Amiga games chart database

  • August 23, 2017 at 5:28 am
    Permalink

    This is an AWESOME idea! That’d be a great resource! I care!

    • August 25, 2017 at 4:33 am
      Permalink

      We care, we care!
      At least the collection of kittens who live inside me do care!
      I suppose that database contains the list of sources so we could expand the data if we have magazines that you did not get to peruse yet? (Also, suggestion: keep a picture of the source data in the database so errors correction and updates can be done quickly.)

      And my, sports games, license games? Urgh… although we all suspected it, it is important to have the data to prove it.
      At least we can now affirm for good that we owe uninformed parents for the torrent of cynical mediocrity that US Gold, Tiertex and others of the same ilk forced onto our unsuspecting – now forever traumatized- inner kittens! 😉

      Fantastic job! You Sir, as usual are a man ch fluffy kitten!

      • August 25, 2017 at 10:54 am
        Permalink

        Thanks. I’m not sure I understand all the kitten talk, but I’m glad yours are doing well and like the database idea. 😉 I’m more of a puppy guy myself.

        I’ve added a link to the report because WordPress won’t let me embed it at the moment. Anyway, yes, sources and dates are cited so anyone could open up the mag the data came from to see the charts for themselves.

        I definitely like the idea of being able to click an attachment to see the original tables, but that would mean adding 2500 images (as the db stands), and 10 – 60 of those would be duplicates per mag because each game in a given chart has its own entry. I did it that way so they could be sorted by position, date or mag.

        To do it the other way – one db entry per chart with a single pic of the data attached – I’d have to add the entries in a free form notes field so they could be searched, but not sorted. Ah, pros and cons… cons and pros.

        Something else I noticed – when World Cup 94 kicked off there was a massive spike in sales of football games. How depressing and predictable.

        I got so sick of typing the same thing I resorted to a Chrome extension called Text Expander that replaces acronyms or whatever with predetermined text so you can type faster. Chrome’s autofill function decided it would randomly choose which entries to remember… and that was only about 5% of them. Brilliant!

        • September 2, 2017 at 7:51 pm
          Permalink

          (Responding in reverse paragraph order).
          I cannot even comprehend how you managed to do all this but I am grateful for your sacrifice!

          Hum, you are right obviously, storing images comes with its own set of issues, some kind of hyperlinks would be nice but these things don’t come without additional work indeed. But I guess I am probably too eager to make suggestions before actually using the DB, usage is usually the best guide.

          As for the World Cup, rahhhh, I concur, it is so sad to witness that adaptation of real life games are more popular than digital native ones (I am misusing the term but it lends itself nicely to that context), especially given that these buyers very likely do not play football games but just watch it on TV.
          On the other hand, this might also tell us something about gameplay: are there games of which the rules are polished enough that they are entertaining enough that one will want to come back to them again and again? Sensible Soccer was popular because it was very playable but also because its core gameplay has been polished in the real world for quite some time. Can we say that virtual sports games are as polished as football is? During the Amiga era, this very likely wasn’t the case.
          Even Speedball II which is an absolute blast, is still very primitive in terms of the styles of play and strategies it allows.
          Something to ponder during the cold winter nights next to the chimney fire for sure. 😉

          The kittens? They control me!
          😉
          (Nah, I just like these creatures, and incidentally, they form the elementary brick of my personal cosmology but that is a story for another time.)

          • September 4, 2017 at 11:45 am
            Permalink

            That would make sense, great idea: upload images to the server and then post links to them within the db entries. I could do bulk edits to add the link to 10 – 100 entries simultaneously with one amendment, which would be a lot less work than adding the same pic individually, and mean there would be no duplicates. I noticed Amiga Force included a top 100 in July 1993 so I expanded the range to include that data.

            Actually, hang on a sec, I’ll use this as a test case. If you search Amiga Force entries in July 1993 you’ll see a link to the magazine extract alongside each of the 100 entries. That seems to work pretty well so I’ll add images for all the rest at some point too.

            Football games are a weird phenomenon. People buy Sensi Soccer and Kick Off because they’re great games, but lots of the others seem to be bought only because they’re football games and it’s the national sport… it’s verging on a mindless impulse. Very odd when the premium price ones would cost £25!

            I suppose all computer football games are a distant approximation of the real thing if you have a passion for the latter. For everyone else, the arcade equivalent may be good enough, unless you’re a stickler for simulation level experiences. It takes all sorts I imagine. I don’t keep up to date with more contemporary football games, I just know they’ve come a long way since the days of Sensi, which targeted the fun factor above all else.

  • August 23, 2017 at 3:16 pm
    Permalink

    Thanks. Glad to hear I haven’t wasted my time. 🙂

    June 1987 to January 1996 is now complete. Amiga Format published charts beyond this, but I think this may be a good cut-off point. I’m up to 1908 entries and starting to crack up just a wee tad of a smidgen.

  • August 24, 2017 at 1:44 am
    Permalink

    that is a great idea and resource, how you not jumped off a cliff with the insanity of doing it all is beyond me !

  • August 24, 2017 at 12:39 pm
    Permalink

    Thanks. The trick is to kid yourself into thinking it won’t be such a big deal before you get stuck in, and by the time you realise you’ve bitten off more than you can chew, you’re in too deep to give up.

    I’ve extended the data to May 1996 now and added loads more stats on budget games.

    • August 25, 2017 at 4:50 am
      Permalink

      Once again, wow, great job!
      Budget games are bound to reveal another kind of dynamic since the good ones should recapture a part of the sales lost to piracy or, on the contrary demonstrate that competition with free pirate versions is a death blow from which few games recover.
      Oh, how much I would love to have valid data about “played pirated copies”. 😀

      • August 25, 2017 at 11:06 am
        Permalink

        Yep, that would be a fascinating, but no doubt impossible poll to organise.

        I’ve noticed some of the budget entries are far more interesting than the premium price games. For instance, the Monkey Island games really found their feet when they were re-released as cheapies. They hardly made much of a dent in the full price charts upon initial release. Not in terms of the long-term staying power you’d expect anyhow.

        It does all help to explain why so many studios put out what would sell, rather than games that would go down in history as classics. When you’re coding for a living you’ve got to put food on the table and if that means lowering your standards, that’s what lots of them did.

        • September 2, 2017 at 8:08 pm
          Permalink

          Yup. Although there are indirect ways to measure the impact of piracy on sales. I will refer you on that topic to this fantastic article of the digital antiquarian : http://www.filfre.net/2016/01/a-pirates-life-for-me-part-3-case-studies-in-copy-protection/ , the part from Doug Bell of FTL/Dungeon Master fame is highly informative in this regard.
          With this in mind, it should be able to gauge the impact of piracy by releasing two paying games of equal quality at the same time, one without protection, and the other with a very solid one. If piracy does not impact sales, they should sell approximately the same number of units. If it does, then it should be pretty obvious. 😉

          As to why studios did often produce very low quality games I think that the phenomenon that you point out is indeed interesting but not necessarily linked to low sales. I suppose that US Gold arcade conversions did overall very well in terms of sales despite being objective stinkers with astounding regularity. I suppose that many studios produced crap because the market was growing and far from saturated and they could get away with it anyway, especially the ones which were subcontracted by license holders.

          • September 4, 2017 at 8:54 am
            Permalink

            Fascinating article, thanks. Yes, it does prove that DRM can be effective enough to hinder piracy as long as economically necessary, and ‘uncrackable’ certainly doesn’t have to be the Holy Grail to strive for from a publisher’s point of view.

            That would be an interesting experiment if any publishers were brave enough to try it with their triple A titles. Some Amiga games were released without any protection at all, but mostly I think because they couldn’t afford to pay the copyright protection licensor the mandatory fee to use it, and didn’t have the dev budget or time to create their own system.

  • August 30, 2017 at 12:04 pm
    Permalink

    Yep, it’s definitely a tall order, and that’s just covering the UK-based press. I imagine other countries had their own polling outlets, which would make for an interesting comparison if I ever found the time to broach that angle.

    Anyway, for now the initial goal has been accomplished… providing at least one chart for every month when the Amiga was in-vogue. You get a different impression of the popularity of games depending on which chart you use so at some point it would be worth going back and working more data into the mix. For instance HMV might have sold more fighting games because – for whatever reason – their stores attracted people who like violent sports. Boots may cater for another audience entirely; more mums perhaps who made some impulse buys in the software department when shopping for pharmaceuticals. Then there’s Gallup who polled multiple stores so may give a more rounded appraisal.

  • September 2, 2017 at 8:17 pm
    Permalink

    I started looking very briefly at the DB and there is an enormous amount of information to peruse which would benefit (that’s the coder in me speaking) from a query language interface. Is there a way to query the DB with something resembling SQL?

    This would be very useful to extract specific information and possibly make visually informative graphs out of it.

    I cannot promise I would have much time to do so but if assistance is helpful in this regard do not hesitate to poke me at my email address!

    • September 4, 2017 at 8:51 am
      Permalink

      Thanks for the offer, though I’m really not sure what the possibilities might be. Zoho provide the infrastructure and offer their services on multiple levels between free and paid. My config is based on the free model so is more limited, yet fine for my purposes.

      You build up your applications using drag and drop functions or by inserting your own code if you know what you’re doing. I’m using a tiny fraction of the available options so what you describe may well already come as standard as long you choose to implement it. Much of it goes over my head to be honest… I’m certainly no programmer. I’d have to go through the help pages to find out about the provision for graphs and whatnot.

      • September 24, 2017 at 6:39 pm
        Permalink

        I do not know what technical possibilities Zoho offers so I would be at pain to extrapolate what is actually possible with this platform but I will try to give a rough example below.

        Say that we would like to map out the popularity of shoot’em up games over time so we first gather a list of all shoot’em up by crossing the Hall of Light and Lemon Amiga’s databases and now we want to visualize on a graph how their rankings evolved.

        If we had to do that manually, this would require using the Zoho web interface to:
        1- search for each individual game in that list
        2- for each line of the result, write down the result in a Google Doc spreadsheet, noting down the date and the ranking
        3- generate a graph ordered by time, with rankings on the vertical axis, and each game with a different color

        Step 3 is the easiest by far, but step 1 and 2 are more tedious, especially step 2 which is essentially soul devouring copy-paste from the Zoho results to the spreadsheet and very error prone.

        With a query language, we could have step 1 and 2 done with a simple query such as :
        select name, rank where name in (“battle squadron”, “silkworm”, [all other shoot’em ups]).
        The result of which could then be exported as a “.csv” file which could simply be imported into the google doc spreadsheet.

        The gain is quite great because this makes a tedious 30min task into a 1 minute one.

        Alas, I know absolutely nothing about Zoho and whether it actually does support this kind of programmatic querying of the database. I might look into it if I have a bit of time but so far my time has been very limited in this area so do not hold your breath. 😉

        • September 25, 2017 at 4:09 am
          Permalink

          Interesting idea, it would be great if you could pull that off. I could give you admin rights so you can have a dabble and see what’s possible. I know Zoho is a very capable, tweakable platform so I wouldn’t be surprised if they have a way of implementing this. They have a forum where this kind of thing is discussed and staff who are very receptive to feedback otherwise.

          • September 25, 2017 at 8:42 pm
            Permalink

            I have given zoho a deeper look Sunday evening and it looks like they have some way of querying the data but I am not sure yet if it has the flexibility I am hoping to find.

            Will update when I found out more about it.

  • September 16, 2017 at 2:22 am
    Permalink

    Thanks for the tip. It would definitely be interesting to compare how tastes differed from one country to another. I suppose a lot more American football games were sold in the US and hockey games in Canada for instance. We’d likely discover some titles that revolve around obscure franchises that never left their country of origin too.

  • September 24, 2017 at 7:00 pm
    Permalink

    A few remarks after using the database for a bit:

    1- Wow, the screenshots are indeed really useful as they provide a way to have a quick access to the whole top 20/50/whatever of the moment which is really useful to get some additional context. Some also have a ST or other chart on the same page which is also really nice. Also, they provide a way to double check the data on the fly so any eventual error can be signaled and thus corrected. (I haven’t noticed any so far. 😉 ) Thanks for adding them!

    2- After checking the screenshots, I noticed that some of the charts have different sources (you mention it your main article but I seemed to have forgotten it :p). It could be useful (for statistical purposes if we ever reach that stage) to include the source of the data in the database. I realize that this is obviously additional work, but since the most of the data is in we can add that information piece by piece: create a new column “source” and populate it by default with “not inputted yet” and allow some users to fill it in whenever they get an occasion to.
    I get that maybe the logistics of allowing input by multiple people might be complicated but if the system supports that I would definitely be ready to spend a bit of time doing my part in that regard. 😉

    Now that the DB is in the wild, it could become a crowd-managed/grown project, allowing other people to build on top of your (massive) work.

    3- I have been itching to do “genre” related searches but this data is not in the DB, so I have found the idea of crossing the DB’s data with the Hall of Light and Lemon Amiga databases (and maybe the Open Amiga Game Database used by FS-UAE) very seducting. I have no idea how that could be done but this would certainly be useful when doing research.

    4- Man, I wish I had more time to invest in this fantastic project. Once again, great job! This is certainly a fantastic stepping stone for Amiga games research.

    • September 25, 2017 at 4:23 am
      Permalink

      No problem, it definitely adds an extra dimension having the source screencaps. There is one top 20 chart which is actually a top 19 because they skipped a number somehow, so if you come across a gap, that’s deliberate.

      Zoho would easily allow for multiple editors, I’d just have to invite them one at a time. I don’t know how I’d manage a Wikipedia style system where it’s totally open access though. That’s probably not a good idea.

      I can see how source data would be useful, good idea. For some mags it might be consistent throughout so all fields could be bulk edited with Gallup, Woolworths or whatever.

      I did post this on EAB (without making any suggestions re: some sort of collaboration), but there wasn’t much interest at all. Strange how that happens sometimes when you’d think they’d be the perfect target audience.

      • September 25, 2017 at 8:39 pm
        Permalink

        Indeed, I would have expected many to bounce around in joy given the variety of creatures roaming around the board. I have seen more excitement on console sites where you would expect the public to be less data oriented and inclined to appreciate this kind of initiative.

        Let us hope that this is just because they have not yet realized the implications.
        I suppose that a few graphical visualizations will do the DB’s advertising job better than a million words. 😉

  • September 24, 2017 at 9:17 pm
    Permalink

    Ah! few others points I wanted to make but forgot:

    1- I noticed you encoded the platform of the game in its title, maybe this should be a separate column to allow finer filtering when needed? Of course this is additional work but as I suggested in my previous comment maybe we can “out-crowd” this work? 😉

    2- Judging from the charts, it seems that we have to face the depressive fact that Rise of the Robots was in fact quite popular. Search it in the results, then cry…
    This said, this is a very informative data point which proves that marketing (which was almost exclusively via the printed press at the time if I recall correctly) works: gamers bought the game because it was hyped for months by gaming magazines. It is highly unlikely that enough parents and grand parents learned about the game to decide to buy it so we must recognize that the gaming press likely had an influence over purchases…
    Which, likely, also shines a light on the success of (bad) arcade ports.
    And incidentally on the corruption of the gaming press by their advertisers.

    There was another point I wanted to mention alas it escapes me a the moment. Hopefully it will pop back up when I get to use the DB again. 😉

    Note: maybe we should move these comments to another place best suited to discussion? Like maybe the EverythingAmiga (currently deserted) forums? Or the EAB?

    • September 24, 2017 at 9:23 pm
      Permalink

      And of course, minutes after posting this, I remembered what my last point was! 🙂

      I came to realize that aside from the article which this comment replies to, the database is not available anywhere from the everythingamiga.com site, which significantly decreases its odds to be discovered by passing visitors as they would have to read this very specific article to learn of its existence.

      Maybe the DB should be referenced more prominently in one of the site main sections?
      Or have a permanent place somewhere in the menus?
      It seems that such a useful project should have more visibility.

      Cheers,
      Laurent

    • September 25, 2017 at 4:32 am
      Permalink

      Having a checkbox to delineate whether entries are for the budget, A1200 etc. release could be useful, but in a lot of cases charts are mixed and they don’t specify which is which. Some have prices which give the game away, others leave you guessing. With some entries it’s even difficult to tell which game they’re referring to e.g. Indy Jones could mean a number of different things.

      RotR is indeed a sad case. Some people seem to have judged it by how good they expected or wanted it to be rather than how it actually turned out. Some critics also reviewed the hype and visuals forgetting to actually assess the game itself.

Leave a Reply