Retail PC Game Sales – Still On Life Support

July 3rd, 2008

A while back, I posted here about the sharp decline in US PC retail game sales (using data that had mostly been helpfully collected and posted by Rob Merritt).

That was two years ago, so the topic deserved a revisit.

The good news is, sales aren’t falling as fast. The bad news is, they’re still falling, down about 4% (10% inflation adjusted) since that last post. Overall, as the graph below shows, in inflation adjusted terms, these sales are down about 61% from their peak in 1999.

US retail PC game sales 1994-2007
(See Graph Data and Sources below)

What’s a little more depressing is that the sales that are being made are almost entirely from old franchises. See these top ten lists for 2007 and 2006. Virtually every game on both lists is based on an old franchise. (I’m counting World of Warcraft as based on the old Warcraft franchise. For that matter, WoW itself is pretty old now. And there is a title “Star Wars: Empire at War that appears to be a non-sequel, but of course, it’s based on Star Wars.) Sims/SimCity titles make up 5 of the 10 entries on both lists.

In short, the US retail PC game market is much smaller than it once was, and not really open to breakthrough new (original) titles.

Even though I’m a developer, I’m probably characteristic of many gamers as well. I have only bought one or two new retail PC games in the last few years. I prefer downloadable casual titles or console titles.

I don’t read the print magazines about PC gaming anymore. Well, we’re down to only one such magazine anyways (PC Gamer is the only one left, I think).

Perhaps I’ll write a longer post in the future about *why* (in my opinion) this decline occurred. Perhaps I’ll also gather data on some more promising areas (casual games, console games, MMORPGs, etc.)

Graph Data and Sources
Data:
Year/Raw Sales/Inflation Adjusted Sales
1994    $966  $1,352
1995  $1,400  $1,905
1996  $1,700  $2,247
1997  $1,800  $2,325
1998  $1,800  $2,290
1999  $1,900  $2,365
2000  $1,600  $1,927
2001  $1,750  $2,049
2002  $1,400  $1,614
2003  $1,200  $1,352
2004  $1,080  $1,185
2005    $953  $1,012
2006    $970    $998
2007    $911    $911

Sources:
2007 data
2006 data
My old post on this topic, with older sources.
My old post was, in turn, largely based on data and sources collected by Rob Merritt, at this link.
CPI data, used for inflation adjustments

Links and such

July 3rd, 2008

Catching up on other blogs a bit. Some of these links are a bit old, but still interesting (to me, anyways):

Greg Costikyan has a great post on his experiences presenting at a venture capital conference. I haven’t attended a VC conference, but parts of his post accurately echo my experiences at various game-oriented conferences.

He also writes a bit about another VC’s comments on the casual game industry.

Besides this blog, I run (through New Crayon Games) the site CasualGameBlogs.com, which aggregates content (feeds) from a bunch of casual game type blogs. I did some updating of that site recently, and added Bret on Social Games. I suppose it’s not a perfect fit, because, as I understand things, “Social Games” are a bit different from casual games, with the former primarily run through social networking sites like Facebook. I’m way behind the times on these social sites (I don’t even have a MySpace page – the SHAME!!!), but probably need to bone up a bit on this stuff. Bret’s site may be helpful for this.

AOL appears to be making efforts to improve its casual games site (Games.com). The community manager, Laurent Courtines, had asked me to add the blog for this site to CasualGameBlogs.com. I declined – blog.games.com is a bit too commercialized for what I want for CGB, but it’s still an interesting read, with a loto of top X lists of various kinds for Games.com.

Finally, while this site has been around for a while (originally under a slightly different name), for those who haven’t found it, CasualCharts.com is a great way to see which games are on top now, and what has fared well historically.

Games, Games, Games

July 1st, 2008

In theory, I should probably create a separate post for each game that I want to comment on. Feh… Here’s some game comments from the last year or two.

Casual Games:
Build-a-Lot – Excellent game. Clever idea, well implemented.
Build-a-Lot 2 – Disappointing sequel. Didn’t add enough new stuff. Still, I bought it and played it a while.
Peggle – Go play this game now, if you haven’t already. Brilliant.
Monopoly by Parker Brothers – The newest version of Monopoly, aimed at the casual market. I’m a sucker for Monopoly, and I did buy this game, but I’m pretty disappointed. As beautiful as the graphics are, the AI is awful, there appears to be no system for internet matchmaking (or did I miss it somehow?), and the graphics are nice, but intrusive (too many animations that can’t easily be skipped, slowing the game).
Ticket To Ride – Still my favorite. Barely fits in the casual genre, but a lot of fun. Apparently there’s a new version out for X-Box Live Arcade, but I’ve been playing the PC version.

Traditional PC Games:
This is a bit embarrassing, but I’ve barely played any traditional (boxed retail) PC games in the last year or two. I bought Flight Simulator X, but only toyed with it a bit. I played a fair amount of SimCity 4, but I haven’t bought the newest SimCity title yet.

Console Games:
My 7 year old son has taken the title of biggest console gamer in the family away from me. Lately, he’s been playing the Lego series of games (the two Lego Star Wars games and the more recent Lego Indiana Jones). These games are brilliant.

We bought a Wii, and it’s about what I expected. A good party game system, with some interesting game mechanics. Most of the games we’ve played are a bit shallow, but we’ve still been having fun with this system. The biggest disappointment is that the remote control thingie doesn’t seem to have a good sense where on the screen you’re pointing to. I guess this is because there’s no calibration step in setting up the device (you know, where the system would tell you to aim at each corner of the screen and press a button), so the Wii doesn’t know if you’re playing on a 25″ TV or a 55″ TV. Perhaps, given the technology the Wii uses, such calibration wouldn’t even be possible. In any case, it makes some games harder and less intuitive than they should be, because where you point the control doesn’t correspond super-closely with where the system thinks you’re pointing (for our TV anyways).

Our family doesn’t have an XBox 360 or PS3 yet, so no comments there.

No Comment

July 1st, 2008

I admit I’m a bit of a sucker for online conversations. For a while, I posted my own thoughts with some regularity here. I still post regularly on a couple of forums, and I occasionally will comment on other people’s blogs.

That said, some time ago I disabled comments on this blog. Why? The blog was getting flooded with spam. Little if any of it showed up on the site, but I was getting hammered with e-mail notifications of new (95%+ spam) comments coming in, and didn’t want to waste time sorting the legitimate comments from the “Free ringtones!” offers (and many that were far less tame).

Despite a lot of filtering and deleting since I started this blog, I’m showing 8113 comments for this blog awaiting moderation. Since I think I shut off all ways that real live human beings would enter comments, I think these are all spam that the bots are trying to submit via some method I don’t have figured out. That’s 8113 comments from 12/3/2007 through 7/1/2008. Yowza.

Yeah, I know there are programs that can help with the filtering, and I used to use one, but it didn’t work very well, and I’m not motivated enough to find a better solution. So, no comments for the time being. (You can still send me e-mails via the spam-resistant address shown on the ‘About’ page).

A (Brief?) Return…

July 1st, 2008

If you’ve been reading amateur blogs for a while, you may be familiar with a certain cycle.

  • Blogger creates blog.
  • A huge flurry of initial posts ensues. Blogger describes such mundane topics as taking his family to McDonald’s.
  • Blogger realizes that perhaps the rest of the world isn’t so interested in his musings on McDonald’s.
  • Posting frequency drops off.
  • A last post or two.
  • Blog goes into hiatus.
  • Anyways, for the time being, I’m adding the following:

  • After some time away, blogger returns with a few posts.
  • It’s probably a doomed effort, but for the near future, I’ll try to post a bit more.

    Casual Games Book

    February 26th, 2007

    Perhaps this is an ominous sign for the industry – someone has recently released a book entitled Creating Casual Games for Profit & Fun. (Amazon link). This is the first I’ve heard of any books specifically aimed at our little corner of the games industry.

    I haven’t read the book, so I have no idea if it’s any good.

    Fun Physics Game

    December 15th, 2006

    Very simple game here. See if you can beat 27.314 seconds…

    A Flash Mini-Game

    December 11th, 2006

    I took some time the week before last to play around with Flash some more. Specifically, I ported one of the mini-games from Banana Bay to Flash, to be played in a browser.

    You can try the result here:

    Banana Bay Mini-Game

    Banana Bay – Mini Game

    While it’s probably on par with a lot of other simple Flash games out there, I’m not entirely thrilled with the results. I was porting relevant sections of my C/C++ codebase into Flash, but some paradigms didn’t translate well (graphical effects mostly). While I went out of my way to implement a few things in alternate ways, overall, I simplified the graphics and sound, and cut out the music, so the resulting mini-game feels a little flat to me. Note that this is NOT the main Banana Bay game, but rather, one of the 7 mini-games that I implemented in Flash.

    Overall, it took me about 25-30 hours to take it to Flash, but I probably over-engineered the framework, and my Flash knowledge starting out was pretty limited. On the other hand, I was taking an existing game (mostly) and existing artwork, so I economized on actual design time.

    Casual Games Biz Article

    November 27th, 2006

    A pretty lengthy, and solid, article on the casual games biz, here.

    Some interesting excerpts:

    The sequel to one of PopCap’s popular word puzzles, “Bookworm Adventures,” is expected to be the most expensive title produced for the casual game genre. PopCap, which has offices in San Francisco, Seattle and Ireland, spent $700,000 over 2 1/2 years developing the game. It’s set to debut online Tuesday at $30 per download.

    $700K is a lot to develop a casual game. I’m anxious to take a look at this one when it comes out. (I don’t see it on PopCap’s site yet – sometimes they do limited releases a week or so before doing a general release to the public…)

    Research group DFC Intelligence estimates that revenues from casual games worldwide will grow to $953 million this year, from $713 million last year. They were $228 million in 2002. Those numbers don’t include casual games played on handheld devices.

    That’s pretty significant – a quadrupling in revenue in 4 years. Also, the last time I saw, traditional (non-casual, non-subscription) PC games were at about $800-900 million annually, in the US. You can probably assume the world-wide totals for traditional PC games are about twice that (probably a bit under $2 billion). Taking the analyst’s estimate and my guesswork at face value, casual games are now about half as big as traditional PC games, and growing much faster.

    Of course, console gaming dwarfs both of the above sectors…

    Is Sony Losing Their Shirt On Each PS3?

    November 17th, 2006

    There’s been a lot of industry speculation that Sony is losing a lot of money with each PS/3 sold.

    This article claims the the total cost of the main unit alone, EXCLUDING controllers, cables, packaging and any profit margin for the retailers (i.e. Wal-Mart’s gotta make money to sell these things), is $805 to $840 (for the low and high end PS/3’s respectively). Shipping cost to the retailers is also apparently excluded. Adding in these costs, (even with a very modest assumed retailer markup) and you get to about $900-$950, which, with the current retail prices at $499 and $599, implies a $300-350 loss per unit.

    I’m not an expert in component pricing and manufacturing costs, but still, I’m skeptical.

    Let’s look at some specific prices in that cost estimate:

    Manufacturing Costs: $40
    Really? Assuming the components come pre-assembled as modules (very likely), then assembling the parts is sort of like putting a Lego toy together – plug everything together, screw a few things together, and done. I’d make a wild guess that a trained worker could assemble 2 units an hour – that’s probably very conservative. The units are apparently being manufactured by a Taiwanese company (Asustek). By my wild guess, that would seem to imply labor costs of about $10/hour, or roughly $5/unit. If Asustek in turn pushes the manufacturing to a plant in South China, it would be even less. Note that in the second table, they estimate XBox 360’s manufacturing cost at only $6.10 per unit. No explanation is given for the enormous discrepency between the systems.

    Enclosure: $31-33
    Really? The plastic/alloy box around the PS/3 costs $31 in million unit bulk order pricing? I doubt it. Considering you can buy computer cases, at quantity 1, at retail price of $15 after rebate from NewEgg, I doubt Sony is paying more than $10-15 for the enclosure

    Power Supply: $37.50
    Hello, I’d like to order 5 million power supplies – how much will that be, per? $37.50? Wait a sec – I can buy a SINGLE 400 watt power supply at NewEgg for $20. You’re telling me my five million unit order is at almost twice that, per unit? Yes, I know the unit in the PS/3 is a ‘slim line’ power supply, but I hardly think that accounts for this price estimate. Thanks, but I’ll find a different supplier.

    Other Components and Manufacturing: $148

    Weren’t we already charged for manufacturing? What are these ‘other components’? The table already appears to list all the major chips, memory, hardware, optical and hard drive…

    Combined price of 4 primary chips (graphics, CPU, I/O, ‘Reality Synthesizer’): $305
    OK, here’s the deal with technology – I think it applies well to these chips, and to a lesser extent to the other items above.

    Chipmaking has 3 main costs:
    1) R & D – How much to invent it? This is basically a single large lump sum invested up front.
    2) Manufacturing ramp-up – Initial yields are low and it takes time and money to figure out how to manufacture items in bulk. Again, this is basically a single large lump sum invested up front.
    3) Per unit costs – Once you’ve got it developed and you have your manufacturing going, how much does it cost to stamp out each chip? This cost declines slowly over the life of manufacturing, but is much more stable (per unit) than the first two costs.

    Now, a chipmaker has to make a profit per chip. Clearly, that entails charging more than the per-unit cost in item 3, but it also involves making enough money over the life of the chip run to recoup (and hopefully make a profit on), items 1 and 2.

    I have no doubt that it cost Sony and their partners an enormous sum to develop these chips and their manufacturing processes. But I really doubt the true per-unit cost (i.e. item 3) is anywhere near $300 for the lot of them. Perhaps whoever created the table in question tried to amortize a significant chunk of items 1 and 2 against the first few hundred thousand chips off the line. But that just seems wrong to me – Sony, internally, is probably amortizing those costs against a projected run of 50-100 million PS/3s, lifetime (maybe more). Taking a very wild guess, I’d say the $305 figure is off by a factor of 2 or so.

    To repeat, I am NOT an expert in these things, and may be wildly wrong. But it seems that analysts are applying cost pricing roughly equivalent to building a single machine with parts from NewEgg or Fry’s, rather than bulk pricing, bulk manufacturing, and amortizing R & D cost over 50-100 million PS/3s that will likely be built, lifetime.

    How much money is Sony really losing, per unit? I don’t know, and perhaps Sony doesn’t really know either. Computing a per unit cost on the first batch of PS/3s is a bit of a silly exercise anyways. They’ll probably lose money in the PS/3’s first year, due to all that R&D, low software sales to a small initial base of PS/3 owners, and declining PS/2 sales. But my wild guess is that Sony will make a lot of money over the course of the PS/3 generation, and also will likely make more money than Microsoft.

    Ticket To Ride Updated

    November 9th, 2006

    For a bit over a year, I’ve been addicted to the online version of Ticket To Ride (from Days of Wonder), which was originally a popular railroad board game, but has a free online version as well.

    While you can play the web version for free, you get certain additional perks by buying a 1 year ‘web card’ for ~$20, and also by buying the CD-ROM version (also ~$20, though they sent me one free last year as an early purchaser of a web card).

    Anyways, they’ve just updated the CD-ROM version to include several new variants, based on their board game expansion pack ‘USA 1910′. If by chance you have the CD-ROM version of the game, you can now download the updates, and even if you’re only playing the web version, you can play the new variants, if someone with the CD-ROM version starts a 1910 game.

    And, for the ~95% of you who don’t know what I’m talking about, I strongly encourage you to try the game – again, it’s free to try. I’ve possibly spent more time playing this game in the last year than any other computer game I’ve played, including Civilization and SimCity.

    Windows Vista Goes Gold

    November 9th, 2006

    Well, I guess I was wrong a couple weeks ago in thinking Vista was not that close to going gold. It’s been announced on the Vista Blog (link via Slashdot) that Vista has been released to manufacturing – aka gone gold.

    Congrats to the Vista team.

    New beta build

    October 28th, 2006

    There’s a new Banana Bay beta build up (try saying that 5 times fast). If you’ve played previous betas, you can grab it from the same location, with the same login.

    For this beta, I’m not adding any new beta testers, but I should have another version later next week that I will open to new testers, so if you’d like to be in it, drop me an e-mail at psteinmeyerpublic A T charter D O T net.

    Windows Vista, R.C. 1

    October 27th, 2006

    You can download for free the first release candidate for Windows Vista.

    I did so last night to try to see if Banana Bay would work alright under Vista.

    Some observations:

    It took a long time to download – it’s about 2.5 GB, and took about 2 hours or so for me to download on my fast cable connection.

    I went to install it on a ~3 year old Windows XP machine. This machine has decent specs – AMD Athlon XP 3000+ CPU, GeForce 5700 video, and 1 GB RAM.

    I wanted to install it to a hard drive partition and keep my old XP install around. This machine is not my primary, and had no important data, but I still wanted to keep XP on it. The Vista RC 1 expires on 5/31/07, and I didn’t really want to be left with a PC with an expired OS (yes, I could reinstall XP later, but that’s a pain).

    I tried the option to install Vista onto a partition, but although there were some vague indications from the language in the setup that I could do this, in fact, it seemed that your drive would already have to be properly partitioned for this to work. Although my hard drive was probably only about 20% full, I couldn’t find a method to partition it as part of the Vista install, and didn’t want to invest hours and/or $$$ finding a 3rd party drive partitioner.

    So I opted for an overall ‘Upgrade’ – i.e. replace my XP install with Vista, but keep my programs intact.

    It didn’t like one program I had installed (Norton), but after I spent 20 minutes or so removing Norton (what a pain), I got Vista to start it’s install/upgrade process again.

    It ran ok for about 3 hours, then stalled in the last of the 5? stages it listed (finishing installation?)

    I let it run overnight, but when it was still dead this morning, I rebooted. After some more complaints from Vista about an incomplete install, I went the 3rd route, and chose the option that amounted to ‘blow away everything on the drive and give me a clean Vista install.

    This took only about an hour, and succeeded, mostly…

    Once Vista fired up and prompted me for my user name and the like, it seemed pretty solid. Internet access worked right off the bat (I didn’t need to enter any settings). The OS seemed solid, except that it was running at 1024 x 768 resolution. I have a Dell 2005 LCD monitor, and it’s native rez (and my preferred rez) is 1600 x 1200. Unfortunately, Vista would not let me bump the rez past 1024 x 768. The problem is that it apparently couldn’t identify my monitor, possible because the PC is not directly connected to the monitor, but rather, goes through a KVM switcher (allowing me to hook 4 PCs to the same monitor, keyboard and mouse).

    It wouldn’t let me manually specify the monitor – it just had the generic default monitor configuration.

    I was able to find a switch for ‘Hide modes that this monitor cannot display’, which was checked by default. I unchecked it, which gave me some more, higher rez modes to select, but they were all widescreen modes, like 1980 x 1080. I ended up leaving things at 1024 x 768. Maybe I’ll fight this battle another day.

    One of the first things you notice with Vista is that it pops up a LOT of warning dialogs when you change just about any setting. I suppose this keeps newbie users out of trouble to an extent, but it’s annoying to a savvy user. I could probably find a way to switch into administrator mode and cut back on this, but I wanted to experience the OS as most users would.

    As for the actual OS – it looks very cool. I’m surprised to admit this, since I thought the default XP look (candy-coated) was awful, but the default Vista look is very nice – better than Mac OS-X in my opinion.

    I went to install my game, which runs quite well in Windows XP, and is very vanilla (does not even use DirectX, but rather, the more basic Windows functions of Waveout for sound, and BitBlt for graphics). I got a few semi-scary warnings when I went through the install process. This is going to scare inexperienced users away from games and such by smaller developers. You can address it to some extent by digitally signing your installers, but that’s expensive, cumbersome, and somewhat oriented towards larger developers, IMO.

    After the install and the various warnings, I went to play the game.

    It all played fine, except the sound was very ‘skippy/stuttering’. Usually this means the sound buffers are too small. I recompiled the game with bigger buffers, and that resolved the problem, but at the cost of much higher sound latency (i.e. the lag between when you, say, press a button, and when you hear the appropriate sound). I added some debug code and determined that I was filling the buffers at proper intervals. Then I added some more debug code to see if the buffers were ever ‘running out’ (which would cause stuttering).

    To do this, I called waveOutGetPosition, which you can use to see if the buffers have data in them. In theory, this function is simply a ‘read-only’ function – it tells you something about the state of audio playback, but should not actually CHANGE anything about audio playback.

    And yet, once I started calling this function, the stuttering went away. I confirmed this by adding a switch by which I could, in game, toggle this function call on and off. I do not know why calling a query function such as this appears to affect the way that buffers are refilled, and it makes me uncomfortable that, at the moment, the ability of my game to play sound correctly in Vista appears to depend on this oddity, but that’s the way it is.

    Otherwise, I haven’t spent too much time using Vista. I will say that my initial experiences with it leave me rather skeptical that this it will really be ready for prime time in the next few weeks (I believe the final is supposed to be seeded to corporate customers in November, and to be pre-installed on consumer PCs in January). We’ll see, but a multi-month slippage in release dates would not surprise me. I doubt they will have a version good enough that I’d want to make it my primary OS come January.

    Edit – One further note: Since I set this Vista machine up about 6 hours ago, its hard drive has been chattering most of the day, even when idle. Hopefully this is just a one time indexing thing or something like that (though the machine is now clean except for the OS and my game – shouldn’t take too long). If Vista is always active and pounding on the hard drive even when you’re leaving it alone, that could become very annoying very quickly.

    Edit 2 – apparently R.C. 1 is NOT brand new but has been around for at least a little bit. R.C. 2 was apparently recently leaked and is better, but is not available at the site above. So, if this version I’m looking at is several months old, it’s more believable that Microsoft could hit a final in a few weeks with a decent version.

    Ayiti – The Cost of Life

    October 25th, 2006

    In my last post I mentioned an interesting serious game in beta. It’s now been released. Ayiti – The Cost of Life is a game where you try to shepherd your poor family in Haiti through the many problems they face. A light graphical touch softens a tough topic – give it a look if the subject interests you.

    On a lighter note, this game shows that even dirt-simple graphics and mechanics can be amusing. It takes about 10 seconds to learn and you’ll probably be done with it in 10 minutes or less, but it’s fun. (My record is 9.1 seconds)

    Finally, check out Funny Farm. A simple word game/trivia puzzle – type in words to expand the grid. For example – cow connects to ‘On the farm’ and cowboy connects to cow. Post your save game link here and you can merge it with others working on the puzzle (it’s a BIG puzzle – you’ll want to work with others…)