Recently, it’s come to my attention that several people aren’t particularly happy with my article about the NES and the Master System in the UK market, finding the evidence supporting the argument to be lacking. Placed in such a position, there’s only really one reasonable thing to do: put forward an even more ridiculous convention-challenging article in its place.
So then, without further ado, let’s look at two new combatants. In the red corner we have the Nintendo Entertainment System. As we’ve seen before, this grey box managed to punch surprisingly far above its weight despite the UK being an unquestionably Sega-leaning territory. In the blue corner we have the Commodore Amiga, grandad of the 16-bit era, and a computer that inspired such loyalty its remaining fans will hunt you down with cudgels for merely suggesting that the Mega Drive had the superior version of Zool. Given the amount of forest needing to be felled to send the several thousand UK-based Amiga magazines to print, there can’t really be any sort of challenge here at all. Or can there?
One thing both the NES and the Amiga have in common is the difficulty in obtaining authoritative sales information that won’t be challenged by some snotty-nosed brat (er, middle-aged man) on one of the last, dying, internet forums. Thankfully, difficult doesn’t necessarily mean impossible, so let’s see what we can tease out about the hardware sales of both machines.
When it comes to the NES, the narrative is one we’ve seen before. With 1987’s launch badly botched by Mattel, the NES had hardly any UK sales in the 1980s but took off quite significantly once San Serif took over distribution in 1990. We’ve already seen a document – purporting to represent some internal information from EA – that supports this narrative, but if we take this to be unreliable, we other sources do we have?
Arriving in December 1991 (and dated January 1992,) Total! Was a magazine from Future Publishing that would spend most of its time covering the Super Nintendo and Gameboy but had a fair amount of original NES content as well. Interestingly in their launch issue, they included a news story from Nintendo that claimed 50,000 NES units had been sold in 1989, just under 200,000 had been sold in 1990, and a whopping 550,000 had been sold in 1991. Crikey.
Oh, I get it, you’re a suspicious person, aren’t you? You’re thinking “Nick, Nintendo weren’t self-distributing in Europe, so why were THEY providing the stats?” This is a reasonable enough question. Thankfully, the same figures also were picked up by US outlet Electronic Gaming Retail News, which (as a slightly more serious publication), reveals that these figures came from Nintendo’s then European representative, Bandai – rather than Nintendo themselves – and were also based on a combination of their own sales figures and G & A Lek Trak, to which Sega was also subscribed. The figures look legit.
Can they be trusted? In the original form repeated in Electronic Gaming Retail News, Bandai made some interesting claims – most notably boasting of a significantly larger install base than the Sega Master System. Is this claim correct? It does feel a little off – though without authoritative Master System results it’s unclear whether they represent a gross inflation of Nintendo’s 90s tally or an ‘accidental’ tuck of Sega’s impressive late-1980s sales.
If we disregard the EA figures as untrustworthy AND think the Bandai figures have a definite pong about them, is there a third source? Thankfully yes, we have another set! You see, though the playground battles between Sega and Nintendo might have been fiercely fought, The UK government weren’t convinced that the two hardware giants were competing as hard as they could be. The Monopolies and Mergers Commission were tasked with launching an investigation into the two companies and in 1995 – just in time for Nintendo and Sega to both be stomped by Sony – they published their report.
Helpfully, the MMC do not appear to have been convinced by the G & A Lek Trak figures used by Sega and Nintendo. Though they did make use of official figures taken Sega and Nintendo, these were augmented by ‘industry and retail sources’, making for a different (and seemingly more reliable) data set.
According to the MMC then, the figures Bandai were floating for 1991 were quite possibly inflated, with Nintendo selling 325,000 units of NES hardware to customers. Mind you, the MMC report also suggests that Nintendo maintained most of 1991’s momentum into the following years, selling 322,000 in 1992 and 273,000 in 1993. Even If hardware sales dropped to just 100,000 in 1994, sales for the NES would have eclipsed one million units based on the period of 1991-4 alone. Factor in the console’s strong second half of 1990 and the much smaller sales from the late 80s and the 1.5 million user base suggested by the EA figures suddenly doesn’t seem so unreasonable at all. Crucially, although we would say each set might be unreliable in their own way, the pictures they draw are reasonably consistent, give or take a few hundred sales of course.
As with the NES, the Amiga has no single unquestioned authority we can turn to for how many systems were sold in the UK. Worse, in this case it seems that the MMC weren’t able to start an investigation into whether Commodore were monopolising the market on unhinged fanboys (Note: this is just a joke. Like the CD32. No, no, I love you really Commodore fans. Please do not break into my house and bludgeon my cats to death with an unsold A600).
Thankfully, Amiga magazines from the period did at least carry several news articles and features which can be of use to us. In the June 1993 volume of Amiga Format, for example, the AF team put together an interesting article charting the state of different Amiga markets across the world. For the UK, they claimed the install base was 1,500,000 Amigas – a figure which was more than double the 700,000-strong US audience. Unfortunately, the article is very much a cursory overview of each region, without any details of how any of these totals were reached.
Helpfully, we also have some figures which come vaguely from the horse’s mouth (or at least the horse’s, er, email) via Dr Peter Kittel, an employee of Commodore Germany. Dr Kittel’s figures are very specific for the number of Commodores sold in Germany (1,680,480) but vaguer for everyone else (2,675,000 for the rest of Europe, 1,240,000 for the USA.) As the UK was Commodore’s other main market in Europe, Kittel’s numbers would suggest the Amiga Format figures are correct for the end of 1993, as long as we presume that they themselves carry on to the end of 1994.
Alongside Kittel’s figures, Amiga Format also elaborated a bit further in their 1993 annual (published, of course, at the end of 1992.) In the annual, they reported that Commodore claimed to have shipped a million Amigas in by February 1992, with three million sold worldwide. Better than that, however, they also penned a short history of the Amiga in the UK, which included periodical sales figures for the all-important A500 (and unloved stepchild A600 variant) from Commodore. According to those figures, 40,000 Amigas were sold between summer of 1987 to the summer of 1988, 160,000 were sold the following year, 200,000 the year after that, 250,000 between July 1990 and June 1991 and 300,000 both for the period covering 1991 -1992 and projected for the period ending in the summer 1993, taking projected sales up to 1,250,000 units. These are broadly in line with the other figures given above when you factor in the potential (more limited) sales of other Amiga devices, like the dead-on-arrival CDTV console and the recently released high-end disappointment, the A1200.
Looking at the figures then, this ridiculous confrontation doesn’t look so ridiculous after all. Going just by the hardware figures, the Amiga and the NES seem to have surprisingly similar stories – albeit starting a year apart. Just as the NES ceded early ground to the Master System, Commodore’s original £1,475 Amiga 1000 was targeted mainly at the creative and business sectors they’d had success with in the US, giving Atari’s ST free reign to mop up the domestic and gaming side of things. Both Nintendo and Commodore had to ditch the strategies they’d made a success of in the US in order to sell machines in the UK, and when they did so the amount they sold seem remarkably similar – with both machines selling around 300,000 units in 1992 and 1993.
Though ownership doesn’t necessarily map onto popularity (I would suspect that everyone who owned a copy of Street Fighter 2 on the SNES would have swapped their game cartridge for a Street Fighter 2 arcade machine in a heartbeat), nonetheless it is worth noting that, if you took a random sample of the population of the UK, you’d have just as much chance of finding a NES owner as you would someone who gamed on an Amiga.
Hardware figures are great focus points for internet forum bun fights, but they don’t alone present an entire picture. Though numerically a single sale at the beginning of a machine’s life is the same size as a single sale at the end, a machine bought at the beginning to be used as a family’s main console for many years will be worth much more financially than a heavily discounted moribund machine bought on a whim so someone can try out one or two big games. We’ve seen that the NES and the Amiga shifted around the same number of hardware units, but were people actively using them and buying software for them?
With a quick Google search, you can easily find the global sales tallies for all the best-selling NES titles. Unfortunately for us, no one seems to have thought to break these totals down for individual countries. On a title-by-title basis we don’t really can’t say how the best-selling games sold in the UK, which is irritating to say the least.
Though we don’t have a lot of data for individual games, we have better luck when we look at the library as a whole. As we can see from the articles we examined above, in 1991 Bandai claimed to have shifted a whopping 2,000,000 units of NES software over the previous twelve months. These figures should be treated with caution, however, as manufacturers often blurred the lines between reporting the number of units sold to retailers with units sold through to the customer. Technically they aren’t strictly false, but they aren’t giving an authentic view of what people were actively playing either.
To play it safe, we should probably turn to the more conservative, retail-sourced numbers supplied by the MMC committee report. It does have the downside of being focused only on Sega and Nintendo – with other publishers’ titles lumped together into one in a way which makes unclear which platforms their sales belong to – but it should give us a lowball figure for Nintendo game sales.
Even without taking account of those limitations, NES sales seem surprisingly strong. If we go by the Swedish release list (Predictably a similar document doesn’t seem to be available for the UK, curse you ABBA) and the MMC sales data we can see that, in 1991, 31,410 copies of Nintendo-published titles were sold for each of the 39 new PAL versions that were made available in that year (making for a total of 1,224,990 sold through to customers.) For the following year there was a reasonable decrease in both metrics – 880,000 sales across 36 new games resulted in sales of about 24,444 per Nintendo-published game – but only a shallow decline the year after (539,000 sales against 24 new games, resulting in 22,458 per new Ninty release.) Impressive stuff considering that, until 1992, the games were locked at an eye watering RRP of £39.99. For perspective, the 1992 equivalent of a ‘triple-A’ game on something like the Spectrum would be £10.99, or around £20-25 on the Amiga.
20,000-30,000 per game seems pretty impressive in itself. However, if we assume there were only around 500,000 NES units in bedrooms by the end of 1991, we can say that Nintendo-published software had an attach rate of around two and a half games per NES owner. Not bad for a period when 2.5 games would have set you back a hundred quid. This seems impressive, but how does it compare to the Amiga?
In a case of being careful what you wish for, the Amiga presents with the opposite problems of the ones we had with the NES. Though there are numerous Amiga developers who’ve released their sales figures over the years, Commodore didn’t rule their platform with the clunky iron fist deployed by Nintendo. Consequently, we have precise figures for a few heavy hitters, but broader figures covering the entire platform are harder to come by.
On the self-selecting front, some of the figures were very impressive. In his history of DMA Design, Mike Dailly recalls how the release-day success took everyone – including publisher Psygnosis – by surprise:
Though they’ve dropped off the face of the earth today, in the 1990s the Lemmings were an absolute phenomenon. The game was converted to anything that could meet its (unsurprisingly meagre) demands and went on to shift north of 15,000,000 copies across all platforms.
So then, 55,000 copies in just a single day was highly remarkable for any game on any platform, and to put this into context Mike helpfully provides us with lifetime sales of two earlier DMA titles that didn’t quite have Lemmings’ success: Menace sold a ‘whopping’ 20,000 copies on the Amiga but Blood Money blew it out of the water with 40,000 – figures which are much closer to the average sales figure for a Nintendo-published title.
However, other tent-pole titles don’t seem to have done so well on the Amiga. The original version of smash-hit Worms sold just 10,000 copies according to Team 17 Founder Martyn Brown, which was about the same number of sales as their ambitious Doom competitor Alien Breed 3D (even though the latter was restricted to the advanced A1200 system only.) Going toe-to-toe with the 16-bit consoles, Electronic Arts also only seem to have managed to sell only 20,000 copies apiece of the Amiga ports of Jungle Strike and PGA European Tour.
Mind you, with that there were a number of titles that sold a lot better. Zool was ambitiously touted as the Amiga’s answer to Mario and Sonic the Hedgehog, and this push seems to led it to sales of around 90,000 copies. Sensible Software were rockstar (the concept, not the publisher) developers on the Amiga and their sales figures proved to be consistently mammoth – Cannon Fodder (complete with tabloid baiting poppy-based ad campaign) shifted 100,000, while Sensible World of Soccer went even further, selling 120,000. Blimey.
Unfortunately, there’s one additional complication that impedes the usefulness of our Amiga figures as well: piracy. Though there is significant disagreement over how bad it actually was (publishers pushing for government intervention suggested it could be as bad as 15 unauthorised copies per sale, though Electronic Arts PR manager Simon Jeffries suggested to CU Amiga magazine that it was more like 6:1), with market stalls across the country selling the latest Amiga titles for just £1 it definitely had some effect on the market.
In a wide-reaching article that interviewed publishers, developers, enforcement officers and even the pirates themselves, Amiga Format presciently warned their readers that they risked killing the platform by buying pirated software. Though this seems hyperbolic, the article contained a few warnings that turned out to be more than idle threats. Peter Molyneux from Bullfrog, for example, warned that though the company had always been Amiga-first, they would likely stop publishing for the platform. Though an A1200 version of their next title was prototyped, 1994’s Magic Carpet would debut on the (Pentium!) PC and next generation of consoles.
Thankfully, it’s a bit beyond our scope to decide whether piracy killed the Amiga – we’re just here to gather a broad-stroke image of how the platform was utilised. Still, the piracy angle does make our job a bit trickier. In their piracy article Amiga Format claimed that by 1993 the average Amiga title sold just 8,000 copies. If we take this figure as gospel (and I’m not saying that it is) a title that sold 8,000 copies could have anything from 8,000 to 120,000 copies actually in circulation, making a straight comparison with other systems difficult. And that’s before we get into the whole popularity angle as well. Should one person playing a copied game just because they could count as much as another who saved for weeks for a specific £39.99 title on the NES?
All we can say objectively and with certainty is that, in the early 90s, both the Amiga and the NES were shifting around 200,000 hardware units per year and were selling games that were able to rack up sales that measured, on average, in the low tens of thousands. If we set the 58,000 units the Megadrive was able to sell per new title in 1993 as our benchmark, we can also say that both the Amiga and NES were probably closer to each other than they were to Sega’s 16-bit behemoth.
We have a bit of a conundrum then. Regardless of which system might have edged out the other, we can say that the platforms sold similar numbers of hardware, shifted software in the same region and were both on largely similar trajectories towards obsolescence. We can objectively say that if you were to grab a random cross-section of the population in 1993, you shouldn’t expect to find any more Amiga owners than you would NES owners.
However, even as the author of this article, writing this just feels wrong. Why? Even just going by my own experience, my expanded circle of friends in the early 90s contained precisely one NES (replacement for a lemonade-spoiled Spectrum) and one Amiga (hand-me-down from an older cousin.) Why does it feel like the Amiga should be so much common?
I’ve seen it suggested that the number of single format magazines as a good way to judge a machine’s popularity. However, I don’t think this is necessarily a great metric. As we’ve seen above, though there is a bit of controversy around just how many Amigas Commodore sold, there can be little doubt that the Amiga sold more units in the UK than the US – and that’s before you consider that Commodore reached around one in every fifty people in the UK but just one in every three hundred people in the US.
You wouldn’t know this, however, if you looked purely at the number of Amiga magazines available in both regions. In fact, despite the UK Amiga representing a much, much larger slice of the market the lion’s share of the magazines probably falls on the US side. The reason seems to be one of demographics. Though in the UK the Amiga leant more towards older teenagers, in the US Commodore did a better job of marketing their machine as a business/productivity tool. Consequently, while in the UK we had a number of big, gaming and general-computing Amiga mags, in the US there was room for more focussed Amiga magazines that covered its various business and design usages. It would be wrong to assume that the Amiga was less popular in the UK simply because the overall number of magazines was slightly smaller.
In the UK then there were a couple of reasons why Amiga mags might be more appealing to a publisher than a magazine that covered an individual games console. The first of these was an unbroken trend of computer magazines being providers of software. In the early 80s this was done via code listings printed in the magazine itself, but as Spectrum and Commodore mags became targeted at a younger readership, magazines started aping the giveaways found in kids comics and begun simply mounting tapes full of games to their front covers.
Though the full Spectrum/C64 games provided by Your Sinclair/Zzap64 were mostly replaced by demos on the Amiga, the presence of software provided an incentive that console-focussed mags simply couldn’t match, as nice as the game novelisations provided by Sega Power were. No one really needed three different opinions on whether Chakan the Forever Man was a bit dull, but, as we’ve seen before, there was always room for an extra computer mag if the games it gave away were good.
Not only were console-focussed mags unable to give away free software, but another issue (ahem) was the difficulty in setting them up. Because console games came on cartridges, console mags required special hardware to play preview releases and needed the cooperation of publishers and – preferably – the console manufacturer to make themselves viable. This all represented a lot of risk, particularly if you were going toe-to-toe with ‘official’ Sega or Nintendo magazines. If you were making a magazine for the Amiga or the Atari ST, you simply required the computer itself and the games on standard floppy disks. Consequently, I would have been much more surprised if the number of Sega magazines available at any one time had eclipsed the number available for the Amiga.
From a practical perspective, it’s easy to see why a computer-based mag would be more attractive than a console based one, and this also goes a long way to explaining why the Amiga seemed so ubiquitous. Standing in front of a newsagent stand in 1992, you would have been confronted with seeming endless computer-focused mags: Amiga Power, Amiga Format, ST Format, Amiga Action, Amiga Computer, ST User, Amiga Force, ST Action and (depending on where you lived) potentially even more.
The reasons for this also extend beyond the practical. Looking back today, it’s easy for us to draw a nice, simple picture of how the UK went from a micro-dominated market to one that had fallen in love with consoles. If we look at media from the time we can see this picture isn’t entirely accurate. From C&VG’s first look at the new consoles in October 1986, it’s clear the new consoles weren’t seen directly as the future or even as some sort of crazy futuristic experiment; instead they were seen as a throwback to the early 80s:
“Do you pensioners out there remember the dedicated video games consoles? Remember all those really brill games on the Coleco, Vectrex and Atari VCS? Don’t know about you, but I miss them SO much!”
Beyond hardware and software price points, the Amiga and NES were in very different places. Though – as we’ve seen – in the end they never generated the sales to deserve it, at the end of the 80s the Amiga and the ST were the self-evident future of home gaming. Publishers who released games for the Spectrum and C64 supported the Amiga and Atari almost unanimously, and the Amiga and ST versions of games were taken as the lead platforms by the gaming press. The Amiga played everything your C64 did but played it better and had advanced titles your old 8-bit machine couldn’t replicate. At the time, the upgrade seemed as natural and obvious as switching from a PS4 to PS5 today.
The consoles, meanwhile, were in an awkward place. Though technically they lived in that 8-bit bracket, the action games they carried often played better than comparable titles on the Amiga and ST, although they naturally didn’t look or sound as good and weren’t capable of recreating the advanced polygon-based games available on the computers. From 1987, consoles were held at arm’s length by the gaming media, with reviews held in separate sections like C&VG’s Mean Machines and their game sales graded in separate sales charts. It wasn’t till Sega’s runaway success of the Megadrive that the industry realised games consoles would make up a fundamental part of the industry’s future and shifted their focus accordingly. Even without its botched launch, the NES would have had to have sold double or triple what it did if it was to be taken seriously by the games media and given a similar amount of weight to Commodore’s machine.
Overall, I hope we’ve taken something away from what seems on the surface as a bit of a ridiculous comparison. As much fun as it is to make fun of the ridiculous and obviously erroneous ‘takes’ of ‘people’ (well, ‘YouTubers’) who were born years after the events they discuss, this particular subject seems to be one specifically designed to trip up those with first- hand experience. At the time, the Amiga was so ubiquitous in the UK gaming scene the idea that it could have only had middling sales would have seemed ridiculous. However, as I think we’ve seen, an unpredictable conspiracy of circumstances made the microcomputer seem far more widespread in the wild than it actually was. As absurd as it may seem to claim the NES was as popular (or at least almost as popular) in the UK as the Commodore Amiga, if we discard media coverage, I don’t see any evidence that suggests this wasn’t actually the case.
Standing in front of a newsagent stand in 1992, it’s easy to understand my anyone would come away thinking they lived in an Amiga-focussed world. Sure, you would have had Total, Mean Machines and Sega Power, but the computer-based mags were endless.
Even with the multi-format mags, it was easy to be led into thinking that the Amiga was the dominant force in gaming. If you were buying games in 1990, you would most likely be doing so on a ZX Spectrum, Commodore 64 or an Amstrad CPC. You wouldn’t know that from looking at C&VG mind you – where most of the screenshots of multiformat releases would inevitably come from the Amiga and the ST.
It would only be by looking closely at the Gallup charts that you would get an idea of the true shape of market. In the December 1990 issue of C&VG, for example, we can see that the top 5 Amiga games were Corporation, Kick Off 2, Immortal, Treasure Island Dizzy and Shadow of the Beast 2. Mostly 16-bit exclusives, none of these titles made it into the all-format top twenty. Looking down the page at the aged Commodore 64, we find all five of the top games (Run the Gauntlet, Hong Kong Phooey, Guardian Angel Pro Boxing and Salamander) landing in the top 20 overall titles. Notably, all five of those titles were part of the £2.99 budget lines. The following year, the powers that be became so irked at the dominance of budget tapes they removed them from the chart entirely, turning it into a full-price top twenty.
Despite the relatively poor sales, it was entirely understandable that magazines would defer to the Amiga and ST. Though the best-selling games on the Amiga may have failed to chart on the wider front, the Amiga represented the future. From the very first edition of C&VGs’s Mean Machines segment (which started as a section rather than its own magazine) it was obvious that the new consoles were viewed as an anachronism, a throwback to the old Atari console. Until the Megadrive forced the industry’s hand, they were seen as inhabiting their own sphere quite separate from the microcomputer scene (though C&VG had a separate Megadrive chart, they still had nothing for either the NES or the Master System.)
As dominant as the Spectrum and the C64 were, they were coming to the end of their useful lives, a hangover from a surprisingly elongated past. By the close of the 80s the Amiga was clearly the anointed successor. The Amiga 500 might have been pricey, but it was in the same region as the Commodore 64 had been back when it launched in 1982. With the Amiga’s wide variety of software support there was no reason to believe that history wouldn’t repeat itself, even if the Amiga had got off to a slow start. As it happened the games console was not quite the dinosaur it appeared to be. Whilst the NES and Master System led inauspicious lives in this market, a keen observer could have looked across the Atlantic to see the shape of things to come.